release 2013.12.02

In this case, it looks rather harmless (since the conditions for --restrict-filenames should not change while a process is running), but just to be sure.
This also simplifies the interface for callers, who can just pass in the idiomatic None for "I don't care, whatever is the default".

README.md

@@ -56,7 +56,8 @@ which means you can modify it, redistribute it or use it however you like.
     --no-playlist              download only the currently playing video
     --age-limit YEARS          download only videos suitable for the given age
     --download-archive FILE    Download only videos not present in the archive
-                               file. Record all downloaded videos in it.
+                               file. Record the IDs of all downloaded videos in
+                               it.
 
 ## Download Options:
     -r, --rate-limit LIMIT     maximum download rate in bytes per second (e.g.
@@ -123,17 +124,18 @@ which means you can modify it, redistribute it or use it however you like.
     --get-description          simulate, quiet but print video description
     --get-filename             simulate, quiet but print output filename
     --get-format               simulate, quiet but print output format
+    -j, --dump-json            simulate, quiet but print JSON information
     --newline                  output progress bar as new lines
     --no-progress              do not print progress bar
     --console-title            display progress in console titlebar
     -v, --verbose              print various debugging information
     --dump-intermediate-pages  print downloaded pages to debug problems(very
                                verbose)
-    --write-pages              Write downloaded pages to files in the current
+    --write-pages              Write downloaded intermediary pages to files in
-                               directory
+                               the current directory to debug problems
 
 ## Video Format Options:
-    -f, --format FORMAT        video format code, specifiy the order of
+    -f, --format FORMAT        video format code, specify the order of
                                preference using slashes: "-f 22/17/18". "-f mp4"
                                and "-f flv" are also supported
     --all-formats              download all available video formats

devscripts/bash-completion.in

@@ -1,10 +1,21 @@
 __youtube_dl()
 {
-    local cur prev opts
+    local cur prev opts fileopts diropts keywords
     COMPREPLY=()
     cur="${COMP_WORDS[COMP_CWORD]}"
+    prev="${COMP_WORDS[COMP_CWORD-1]}"
     opts="{{flags}}"
-    keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater"
+    keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater :ythistory"
+    fileopts="-a|--batch-file|--download-archive|--cookies"
+    diropts="--cache-dir"
+
+    if [[ ${prev} =~ ${fileopts} ]]; then
+        COMPREPLY=( $(compgen -f -- ${cur}) )
+        return 0
+    elif [[ ${prev} =~ ${diropts} ]]; then
+        COMPREPLY=( $(compgen -d -- ${cur}) )
+        return 0
+    fi
 
     if [[ ${cur} =~ : ]]; then
         COMPREPLY=( $(compgen -W "${keywords}" -- ${cur}) )

setup.py

@@ -48,7 +48,7 @@ else:
         'data_files': [  # Installing system-wide would require sudo...
             ('etc/bash_completion.d', ['youtube-dl.bash-completion']),
             ('share/doc/youtube_dl', ['README.txt']),
-            ('share/man/man1/', ['youtube-dl.1'])
+            ('share/man/man1', ['youtube-dl.1'])
         ]
     }
     if setuptools_available:

test/helper.py

@@ -12,10 +12,6 @@ from youtube_dl import YoutubeDL
 from youtube_dl.utils import preferredencoding
 
 
-def global_setup():
-    youtube_dl._setup_opener(timeout=10)
-
-
 def get_params(override=None):
     PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
                                    "parameters.json")

test/parameters.json

@@ -39,5 +39,6 @@
     "writeinfojson": true, 
     "writesubtitles": false,
     "allsubtitles": false,
-    "listssubtitles": false
+    "listssubtitles": false,
+    "socket_timeout": 20
 }

test/test_age_restriction.py

@@ -6,8 +6,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import global_setup, try_rm
+from test.helper import try_rm
-global_setup()
 
 
 from youtube_dl import YoutubeDL
@@ -24,7 +23,7 @@ def _download_restricted(url, filename, age):
     }
     ydl = YoutubeDL(params)
     ydl.add_default_info_extractors()
-    json_filename = filename + '.info.json'
+    json_filename = os.path.splitext(filename)[0] + '.info.json'
     try_rm(json_filename)
     ydl.download([url])
     res = os.path.exists(json_filename)

test/test_all_urls.py

@@ -100,10 +100,15 @@ class TestAllURLsMatching(unittest.TestCase):
     def test_keywords(self):
         self.assertMatch(':ytsubs', ['youtube:subscriptions'])
         self.assertMatch(':ytsubscriptions', ['youtube:subscriptions'])
-        self.assertMatch(':thedailyshow', ['ComedyCentral'])
+        self.assertMatch(':ythistory', ['youtube:history'])
-        self.assertMatch(':tds', ['ComedyCentral'])
+        self.assertMatch(':thedailyshow', ['ComedyCentralShows'])
-        self.assertMatch(':colbertreport', ['ComedyCentral'])
+        self.assertMatch(':tds', ['ComedyCentralShows'])
-        self.assertMatch(':cr', ['ComedyCentral'])
+        self.assertMatch(':colbertreport', ['ComedyCentralShows'])
+        self.assertMatch(':cr', ['ComedyCentralShows'])
+
+    def test_vimeo_matching(self):
+        self.assertMatch('http://vimeo.com/channels/tributes', ['vimeo:channel'])
+        self.assertMatch('http://vimeo.com/user7108434', ['vimeo:user'])
 
 
 if __name__ == '__main__':

test/test_download.py

@@ -9,12 +9,10 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from test.helper import (
     get_params,
     get_testcases,
-    global_setup,
     try_rm,
     md5,
     report_warning
 )
-global_setup()
 
 
 import hashlib
@@ -103,7 +101,7 @@ def generator(test_case):
                 tc_filename = get_tc_filename(tc)
                 try_rm(tc_filename)
                 try_rm(tc_filename + '.part')
-                try_rm(tc_filename + '.info.json')
+                try_rm(os.path.splitext(tc_filename)[0] + '.info.json')
         try_rm_tcs_files()
         try:
             try_num = 1
@@ -130,11 +128,12 @@ def generator(test_case):
                 if not test_case.get('params', {}).get('skip_download', False):
                     self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
                     self.assertTrue(tc_filename in finished_hook_called)
-                self.assertTrue(os.path.exists(tc_filename + '.info.json'))
+                info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
+                self.assertTrue(os.path.exists(info_json_fn))
                 if 'md5' in tc:
                     md5_for_file = _file_md5(tc_filename)
                     self.assertEqual(md5_for_file, tc['md5'])
-                with io.open(tc_filename + '.info.json', encoding='utf-8') as infof:
+                with io.open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)
                 for (info_field, expected) in tc.get('info_dict', {}).items():
                     if isinstance(expected, compat_str) and expected.startswith('md5:'):

test/test_playlists.py

@@ -8,20 +8,21 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL, global_setup
+from test.helper import FakeYDL
-global_setup()
 
 
 from youtube_dl.extractor import (
     DailymotionPlaylistIE,
     DailymotionUserIE,
     VimeoChannelIE,
+    VimeoUserIE,
     UstreamChannelIE,
     SoundcloudSetIE,
     SoundcloudUserIE,
     LivestreamIE,
     NHLVideocenterIE,
     BambuserChannelIE,
+    BandcampAlbumIE
 )
 
 
@@ -54,6 +55,14 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['title'], u'Vimeo Tributes')
         self.assertTrue(len(result['entries']) > 24)
 
+    def test_vimeo_user(self):
+        dl = FakeYDL()
+        ie = VimeoUserIE(dl)
+        result = ie.extract('http://vimeo.com/nkistudio/videos')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], u'Nki')
+        self.assertTrue(len(result['entries']) > 65)
+
     def test_ustream_channel(self):
         dl = FakeYDL()
         ie = UstreamChannelIE(dl)
@@ -101,7 +110,15 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://bambuser.com/channel/pixelversity')
         self.assertIsPlaylist(result)
         self.assertEqual(result['title'], u'pixelversity')
-        self.assertTrue(len(result['entries']) >= 66)
+        self.assertTrue(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')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], u'Nightmare Night EP')
+        self.assertTrue(len(result['entries']) >= 4)
 
 if __name__ == '__main__':
     unittest.main()

test/test_subtitles.py

@@ -6,8 +6,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL, global_setup, md5
+from test.helper import FakeYDL, md5
-global_setup()
 
 
 from youtube_dl.extractor import (
@@ -73,7 +72,7 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.DL.params['writesubtitles'] = True
         self.DL.params['subtitlesformat'] = 'vtt'
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '356cdc577fde0c6783b9b822e7206ff7')
+        self.assertEqual(md5(subtitles['en']), '3cb210999d3e021bd6c7f0ea751eab06')
 
     def test_youtube_list_subtitles(self):
         self.DL.expect_warning(u'Video doesn\'t have automatic captions')

test/test_utils.py

@@ -24,6 +24,8 @@ from youtube_dl.utils import (
     xpath_with_ns,
     smuggle_url,
     unsmuggle_url,
+    shell_quote,
+    encodeFilename,
 )
 
 if sys.version_info < (3, 0):
@@ -170,6 +172,10 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(res_url, url)
         self.assertEqual(res_data, None)
 
+    def test_shell_quote(self):
+        args = ['ffmpeg', '-i', encodeFilename(u'ñ€ß\'.mp4')]
+        self.assertEqual(shell_quote(args), u"""ffmpeg -i 'ñ€ß'"'"'.mp4'""")
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_write_annotations.py

@@ -7,8 +7,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import get_params, global_setup, try_rm
+from test.helper import get_params, try_rm
-global_setup()
 
 
 import io

test/test_write_info_json.py

@@ -7,8 +7,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import get_params, global_setup
+from test.helper import get_params
-global_setup()
 
 
 import io
@@ -31,7 +30,7 @@ params = get_params({
 
 
 TEST_ID = 'BaW_jenozKc'
-INFO_JSON_FILE = TEST_ID + '.mp4.info.json'
+INFO_JSON_FILE = TEST_ID + '.info.json'
 DESCRIPTION_FILE = TEST_ID + '.mp4.description'
 EXPECTED_DESCRIPTION = u'''test chars:  "'/\ä↭𝕐
 

test/test_youtube_lists.py

@@ -6,8 +6,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL, global_setup
+from test.helper import FakeYDL
-global_setup()
 
 
 from youtube_dl.extractor import (
@@ -27,7 +26,7 @@ class TestYoutubeLists(unittest.TestCase):
     def test_youtube_playlist(self):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
-        result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')[0]
+        result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
         self.assertIsPlaylist(result)
         self.assertEqual(result['title'], 'ytdl test PL')
         ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
@@ -44,13 +43,13 @@ class TestYoutubeLists(unittest.TestCase):
     def test_issue_673(self):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
-        result = ie.extract('PLBB231211A4F62143')[0]
+        result = ie.extract('PLBB231211A4F62143')
         self.assertTrue(len(result['entries']) > 25)
 
     def test_youtube_playlist_long(self):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
-        result = ie.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')[0]
+        result = ie.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
         self.assertIsPlaylist(result)
         self.assertTrue(len(result['entries']) >= 799)
 
@@ -58,7 +57,7 @@ class TestYoutubeLists(unittest.TestCase):
         #651
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
-        result = ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')[0]
+        result = ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
         ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
         self.assertFalse('pElCt5oNDuI' in ytie_results)
         self.assertFalse('KdPEApIVdWM' in ytie_results)
@@ -66,7 +65,7 @@ class TestYoutubeLists(unittest.TestCase):
     def test_youtube_playlist_empty(self):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
-        result = ie.extract('https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx')[0]
+        result = ie.extract('https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx')
         self.assertIsPlaylist(result)
         self.assertEqual(len(result['entries']), 0)
 
@@ -74,7 +73,7 @@ class TestYoutubeLists(unittest.TestCase):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
         # TODO find a > 100 (paginating?) videos course
-        result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')[0]
+        result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
         entries = result['entries']
         self.assertEqual(YoutubeIE()._extract_id(entries[0]['url']), 'j9WZyLZCBzs')
         self.assertEqual(len(entries), 25)
@@ -84,22 +83,22 @@ class TestYoutubeLists(unittest.TestCase):
         dl = FakeYDL()
         ie = YoutubeChannelIE(dl)
         #test paginated channel
-        result = ie.extract('https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w')[0]
+        result = ie.extract('https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w')
         self.assertTrue(len(result['entries']) > 90)
         #test autogenerated channel
-        result = ie.extract('https://www.youtube.com/channel/HCtnHdj3df7iM/videos')[0]
+        result = ie.extract('https://www.youtube.com/channel/HCtnHdj3df7iM/videos')
         self.assertTrue(len(result['entries']) >= 18)
 
     def test_youtube_user(self):
         dl = FakeYDL()
         ie = YoutubeUserIE(dl)
-        result = ie.extract('https://www.youtube.com/user/TheLinuxFoundation')[0]
+        result = ie.extract('https://www.youtube.com/user/TheLinuxFoundation')
         self.assertTrue(len(result['entries']) >= 320)
 
     def test_youtube_safe_search(self):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
-        result = ie.extract('PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl')[0]
+        result = ie.extract('PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl')
         self.assertEqual(len(result['entries']), 2)
 
     def test_youtube_show(self):
@@ -108,5 +107,14 @@ class TestYoutubeLists(unittest.TestCase):
         result = ie.extract('http://www.youtube.com/show/airdisasters')
         self.assertTrue(len(result) >= 3)
 
+    def test_youtube_mix(self):
+        dl = FakeYDL()
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('http://www.youtube.com/watch?v=lLJf9qJHR3E&list=RDrjFaenf1T-Y')
+        entries = result['entries']
+        self.assertTrue(len(entries) >= 20)
+        original_video = entries[0]
+        self.assertEqual(original_video['id'], 'rjFaenf1T-Y')
+
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_signature.py

@@ -6,9 +6,6 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import global_setup
-global_setup()
-
 
 import io
 import re

youtube_dl/FileDownloader.py

@@ -1,19 +1,16 @@
-import math
 import os
 import re
 import subprocess
 import sys
 import time
 
-if os.name == 'nt':
-    import ctypes
-
 from .utils import (
     compat_urllib_error,
     compat_urllib_request,
     ContentTooShortError,
     determine_ext,
     encodeFilename,
+    format_bytes,
     sanitize_open,
     timeconvert,
 )
@@ -56,20 +53,6 @@ class FileDownloader(object):
         self._progress_hooks = []
         self.params = params
 
-    @staticmethod
-    def format_bytes(bytes):
-        if bytes is None:
-            return 'N/A'
-        if type(bytes) is str:
-            bytes = float(bytes)
-        if bytes == 0.0:
-            exponent = 0
-        else:
-            exponent = int(math.log(bytes, 1024.0))
-        suffix = ['B','KiB','MiB','GiB','TiB','PiB','EiB','ZiB','YiB'][exponent]
-        converted = float(bytes) / float(1024 ** exponent)
-        return '%.2f%s' % (converted, suffix)
-
     @staticmethod
     def format_seconds(seconds):
         (mins, secs) = divmod(seconds, 60)
@@ -120,7 +103,7 @@ class FileDownloader(object):
     def format_speed(speed):
         if speed is None:
             return '%10s' % '---b/s'
-        return '%10s' % ('%s/s' % FileDownloader.format_bytes(speed))
+        return '%10s' % ('%s/s' % format_bytes(speed))
 
     @staticmethod
     def best_block_size(elapsed_time, bytes):
@@ -151,16 +134,8 @@ class FileDownloader(object):
     def to_stderr(self, message):
         self.ydl.to_screen(message)
 
-    def to_cons_title(self, message):
+    def to_console_title(self, message):
-        """Set console/terminal window title to message."""
+        self.ydl.to_console_title(message)
-        if not self.params.get('consoletitle', False):
-            return
-        if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
-            # c_wchar_p() might not be necessary if `message` is
-            # already of type unicode()
-            ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
-        elif 'TERM' in os.environ:
-            self.to_screen('\033]0;%s\007' % message, skip_eol=True)
 
     def trouble(self, *args, **kargs):
         self.ydl.trouble(*args, **kargs)
@@ -249,7 +224,7 @@ class FileDownloader(object):
         else:
             self.to_screen(u'\r%s[download] %s of %s at %s ETA %s' %
                 (clear_line, percent_str, data_len_str, speed_str, eta_str), skip_eol=True)
-        self.to_cons_title(u'youtube-dl - %s of %s at %s ETA %s' %
+        self.to_console_title(u'youtube-dl - %s of %s at %s ETA %s' %
                 (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip()))
 
     def report_resuming_byte(self, resume_len):
@@ -281,6 +256,61 @@ class FileDownloader(object):
                 (clear_line, data_len_str, self.format_seconds(tot_time)))
 
     def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url, live):
+        def run_rtmpdump(args):
+            start = time.time()
+            resume_percent = None
+            resume_downloaded_data_len = None
+            proc = subprocess.Popen(args, stderr=subprocess.PIPE)
+            cursor_in_new_line = True
+            proc_stderr_closed = False
+            while not proc_stderr_closed:
+                # read line from stderr
+                line = u''
+                while True:
+                    char = proc.stderr.read(1)
+                    if not char:
+                        proc_stderr_closed = True
+                        break
+                    if char in [b'\r', b'\n']:
+                        break
+                    line += char.decode('ascii', 'replace')
+                if not line:
+                    # proc_stderr_closed is True
+                    continue
+                mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec \(([0-9]{1,2}\.[0-9])%\)', line)
+                if mobj:
+                    downloaded_data_len = int(float(mobj.group(1))*1024)
+                    percent = float(mobj.group(2))
+                    if not resume_percent:
+                        resume_percent = percent
+                        resume_downloaded_data_len = downloaded_data_len
+                    eta = self.calc_eta(start, time.time(), 100-resume_percent, percent-resume_percent)
+                    speed = self.calc_speed(start, time.time(), downloaded_data_len-resume_downloaded_data_len)
+                    data_len = None
+                    if percent > 0:
+                        data_len = int(downloaded_data_len * 100 / percent)
+                    data_len_str = u'~' + format_bytes(data_len)
+                    self.report_progress(percent, data_len_str, speed, eta)
+                    cursor_in_new_line = False
+                    self._hook_progress({
+                        'downloaded_bytes': downloaded_data_len,
+                        'total_bytes': data_len,
+                        'tmpfilename': tmpfilename,
+                        'filename': filename,
+                        'status': 'downloading',
+                        'eta': eta,
+                        'speed': speed,
+                    })
+                elif self.params.get('verbose', False):
+                    if not cursor_in_new_line:
+                        self.to_screen(u'')
+                    cursor_in_new_line = True
+                    self.to_screen(u'[rtmpdump] '+line)
+            proc.wait()
+            if not cursor_in_new_line:
+                self.to_screen(u'')
+            return proc.returncode
+
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
         test = self.params.get('test', False)
@@ -291,12 +321,11 @@ class FileDownloader(object):
         except (OSError, IOError):
             self.report_error(u'RTMP download detected but "rtmpdump" could not be run')
             return False
-        verbosity_option = '--verbose' if self.params.get('verbose', False) else '--quiet'
 
         # Download using rtmpdump. rtmpdump returns exit code 2 when
         # the connection was interrumpted and resuming appears to be
         # possible. This is part of rtmpdump's normal usage, AFAIK.
-        basic_args = ['rtmpdump', verbosity_option, '-r', url, '-o', tmpfilename]
+        basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename]
         if player_url is not None:
             basic_args += ['--swfVfy', player_url]
         if page_url is not None:
@@ -310,30 +339,48 @@ class FileDownloader(object):
         if live:
             basic_args += ['--live']
         args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)]
+
+        if sys.platform == 'win32' and sys.version_info < (3, 0):
+            # Windows subprocess module does not actually support Unicode
+            # on Python 2.x
+            # See http://stackoverflow.com/a/9951851/35070
+            subprocess_encoding = sys.getfilesystemencoding()
+            args = [a.encode(subprocess_encoding, 'ignore') for a in args]
+        else:
+            subprocess_encoding = None
+
         if self.params.get('verbose', False):
+            if subprocess_encoding:
+                str_args = [
+                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
+                    for a in args]
+            else:
+                str_args = args
             try:
                 import pipes
-                shell_quote = lambda args: ' '.join(map(pipes.quote, args))
+                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
             except ImportError:
                 shell_quote = repr
-            self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(args))
+            self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(str_args))
-        retval = subprocess.call(args)
+
+        retval = run_rtmpdump(args)
+
         while (retval == 2 or retval == 1) and not test:
             prevsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True)
+            self.to_screen(u'[rtmpdump] %s bytes' % prevsize)
             time.sleep(5.0) # This seems to be needed
-            retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
+            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
             cursize = os.path.getsize(encodeFilename(tmpfilename))
             if prevsize == cursize and retval == 1:
                 break
              # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
             if prevsize == cursize and retval == 2 and cursize > 1024:
-                self.to_screen(u'\r[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
+                self.to_screen(u'[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
                 retval = 0
                 break
         if retval == 0 or (test and retval == 2):
             fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'\r[rtmpdump] %s bytes' % fsize)
+            self.to_screen(u'[rtmpdump] %s bytes' % fsize)
             self.try_rename(tmpfilename, filename)
             self._hook_progress({
                 'downloaded_bytes': fsize,
@@ -536,7 +583,7 @@ class FileDownloader(object):
                 self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
                 return False
 
-        data_len_str = self.format_bytes(data_len)
+        data_len_str = format_bytes(data_len)
         byte_counter = 0 + resume_len
         block_size = self.params.get('buffersize', 1024)
         start = time.time()

youtube_dl/YoutubeDL.py

@@ -5,17 +5,53 @@ from __future__ import absolute_import
 
 import errno
 import io
+import json
 import os
+import platform
 import re
 import shutil
+import subprocess
 import socket
 import sys
 import time
 import traceback
 
-from .utils import *
+if os.name == 'nt':
+    import ctypes
+
+from .utils import (
+    compat_cookiejar,
+    compat_http_client,
+    compat_print,
+    compat_str,
+    compat_urllib_error,
+    compat_urllib_request,
+    ContentTooShortError,
+    date_from_str,
+    DateRange,
+    determine_ext,
+    DownloadError,
+    encodeFilename,
+    ExtractorError,
+    format_bytes,
+    locked_file,
+    make_HTTPS_handler,
+    MaxDownloadsReached,
+    PostProcessingError,
+    platform_name,
+    preferredencoding,
+    SameFileError,
+    sanitize_filename,
+    subtitles_filename,
+    takewhile_inclusive,
+    UnavailableVideoError,
+    write_json_file,
+    write_string,
+    YoutubeDLHandler,
+)
 from .extractor import get_info_extractor, gen_extractors
 from .FileDownloader import FileDownloader
+from .version import __version__
 
 
 class YoutubeDL(object):
@@ -57,6 +93,7 @@ class YoutubeDL(object):
     forcethumbnail:    Force printing thumbnail URL.
     forcedescription:  Force printing description.
     forcefilename:     Force printing final filename.
+    forcejson:         Force printing info_dict as JSON.
     simulate:          Do not download the video files.
     format:            Video format code.
     format_limit:      Highest quality format to try.
@@ -68,6 +105,7 @@ class YoutubeDL(object):
     playlistend:       Playlist item to end at.
     matchtitle:        Download only matching titles.
     rejecttitle:       Reject downloads for matching titles.
+    logger:            Log messages to a logging.Logger instance.
     logtostderr:       Log messages to stderr instead of stdout.
     writedescription:  Write the video description to a .description file
     writeinfojson:     Write the video description to a .info.json file
@@ -88,9 +126,13 @@ class YoutubeDL(object):
     noplaylist:        Download single video instead of a playlist if in doubt.
     age_limit:         An integer representing the user's age in years.
                        Unsuitable videos for the given age are skipped.
-    downloadarchive:   File name of a file where all downloads are recorded.
+    download_archive:   File name of a file where all downloads are recorded.
                        Videos already present in the file are not downloaded
                        again.
+    cookiefile:        File name where cookies should be read from and dumped to.
+    nocheckcertificate:Do not verify SSL certificates
+    proxy:             URL of the proxy server to use
+    socket_timeout:    Time to wait for unresponsive hosts, in seconds
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
@@ -105,7 +147,7 @@ class YoutubeDL(object):
     _num_downloads = None
     _screen_file = None
 
-    def __init__(self, params):
+    def __init__(self, params=None):
         """Create a FileDownloader object with the given options."""
         self._ies = []
         self._ies_instances = {}
@@ -114,6 +156,7 @@ class YoutubeDL(object):
         self._download_retcode = 0
         self._num_downloads = 0
         self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
+        self.params = {} if params is None else params
 
         if (sys.version_info >= (3,) and sys.platform != 'win32' and
                 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
@@ -123,14 +166,15 @@ class YoutubeDL(object):
                 u'Assuming --restrict-filenames since file system encoding '
                 u'cannot encode all charactes. '
                 u'Set the LC_ALL environment variable to fix this.')
-            params['restrictfilenames'] = True
+            self.params['restrictfilenames'] = True
 
-        self.params = params
         self.fd = FileDownloader(self, self.params)
 
-        if '%(stitle)s' in self.params['outtmpl']:
+        if '%(stitle)s' in self.params.get('outtmpl', ''):
             self.report_warning(u'%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
 
+        self._setup_opener()
+
     def add_info_extractor(self, ie):
         """Add an InfoExtractor object to the end of the list."""
         self._ies.append(ie)
@@ -163,7 +207,9 @@ class YoutubeDL(object):
 
     def to_screen(self, message, skip_eol=False):
         """Print message to stdout if not in quiet mode."""
-        if not self.params.get('quiet', False):
+        if self.params.get('logger'):
+            self.params['logger'].debug(message)
+        elif not self.params.get('quiet', False):
             terminator = [u'\n', u''][skip_eol]
             output = message + terminator
             write_string(output, self._screen_file)
@@ -171,14 +217,47 @@ class YoutubeDL(object):
     def to_stderr(self, message):
         """Print message to stderr."""
         assert type(message) == type(u'')
-        output = message + u'\n'
+        if self.params.get('logger'):
-        if 'b' in getattr(self._screen_file, 'mode', '') or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
+            self.params['logger'].error(message)
-            output = output.encode(preferredencoding())
+        else:
-        sys.stderr.write(output)
+            output = message + u'\n'
+            if 'b' in getattr(self._screen_file, 'mode', '') or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
+                output = output.encode(preferredencoding())
+            sys.stderr.write(output)
 
-    def fixed_template(self):
+    def to_console_title(self, message):
-        """Checks if the output template is fixed."""
+        if not self.params.get('consoletitle', False):
-        return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None)
+            return
+        if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
+            # c_wchar_p() might not be necessary if `message` is
+            # already of type unicode()
+            ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
+        elif 'TERM' in os.environ:
+            write_string(u'\033]0;%s\007' % message, self._screen_file)
+
+    def save_console_title(self):
+        if not self.params.get('consoletitle', False):
+            return
+        if 'TERM' in os.environ:
+            # Save the title on stack
+            write_string(u'\033[22;0t', self._screen_file)
+
+    def restore_console_title(self):
+        if not self.params.get('consoletitle', False):
+            return
+        if 'TERM' in os.environ:
+            # Restore the title from stack
+            write_string(u'\033[23;0t', self._screen_file)
+
+    def __enter__(self):
+        self.save_console_title()
+        return self
+
+    def __exit__(self, *args):
+        self.restore_console_title()
+    
+        if self.params.get('cookiefile') is not None:
+            self.cookiejar.save()
 
     def trouble(self, message=None, tb=None):
         """Determine action to take when a download problem appears.
@@ -254,7 +333,7 @@ class YoutubeDL(object):
         """Report file has already been fully downloaded."""
         try:
             self.to_screen(u'[download] %s has already been downloaded' % file_name)
-        except (UnicodeEncodeError) as err:
+        except UnicodeEncodeError:
             self.to_screen(u'[download] The file has already been downloaded')
 
     def increment_downloads(self):
@@ -295,15 +374,17 @@ class YoutubeDL(object):
     def _match_entry(self, info_dict):
         """ Returns None iff the file should be downloaded """
 
-        title = info_dict['title']
+        if 'title' in info_dict:
-        matchtitle = self.params.get('matchtitle', False)
+            # This can happen when we're just evaluating the playlist
-        if matchtitle:
+            title = info_dict['title']
-            if not re.search(matchtitle, title, re.IGNORECASE):
+            matchtitle = self.params.get('matchtitle', False)
-                return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
+            if matchtitle:
-        rejecttitle = self.params.get('rejecttitle', False)
+                if not re.search(matchtitle, title, re.IGNORECASE):
-        if rejecttitle:
+                    return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
-            if re.search(rejecttitle, title, re.IGNORECASE):
+            rejecttitle = self.params.get('rejecttitle', False)
-                return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
+            if rejecttitle:
+                if re.search(rejecttitle, title, re.IGNORECASE):
+                    return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
         date = info_dict.get('upload_date', None)
         if date is not None:
             dateRange = self.params.get('daterange', DateRange())
@@ -314,8 +395,8 @@ class YoutubeDL(object):
             if age_limit < info_dict.get('age_limit', 0):
                 return u'Skipping "' + title + '" because it is age restricted'
         if self.in_download_archive(info_dict):
-            return (u'%(title)s has already been recorded in archive'
+            return (u'%s has already been recorded in archive'
-                    % info_dict)
+                    % info_dict.get('title', info_dict.get('id', u'video')))
         return None
 
     @staticmethod
@@ -385,7 +466,7 @@ class YoutubeDL(object):
         result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
         if result_type == 'video':
             self.add_extra_info(ie_result, extra_info)
-            return self.process_video_result(ie_result)
+            return self.process_video_result(ie_result, download=download)
         elif result_type == 'url':
             # We have to add extra_info to the results because it may be
             # contained in a playlist
@@ -394,7 +475,7 @@ class YoutubeDL(object):
                                      ie_key=ie_result.get('ie_key'),
                                      extra_info=extra_info)
         elif result_type == 'playlist':
-            self.add_extra_info(ie_result, extra_info)
+
             # We process each entry in the playlist
             playlist = ie_result.get('title', None) or ie_result.get('id', None)
             self.to_screen(u'[download] Downloading playlist: %s' % playlist)
@@ -424,6 +505,12 @@ class YoutubeDL(object):
                     'webpage_url': ie_result['webpage_url'],
                     'extractor_key': ie_result['extractor_key'],
                 }
+
+                reason = self._match_entry(entry)
+                if reason is not None:
+                    self.to_screen(u'[download] ' + reason)
+                    continue
+
                 entry_result = self.process_ie_result(entry,
                                                       download=download,
                                                       extra_info=extra)
@@ -579,7 +666,7 @@ class YoutubeDL(object):
 
         # Forced printings
         if self.params.get('forcetitle', False):
-            compat_print(info_dict['title'])
+            compat_print(info_dict['fulltitle'])
         if self.params.get('forceid', False):
             compat_print(info_dict['id'])
         if self.params.get('forceurl', False):
@@ -593,6 +680,8 @@ class YoutubeDL(object):
             compat_print(filename)
         if self.params.get('forceformat', False):
             compat_print(info_dict['format'])
+        if self.params.get('forcejson', False):
+            compat_print(json.dumps(info_dict))
 
         # Do nothing else if in simulate mode
         if self.params.get('simulate', False):
@@ -655,7 +744,7 @@ class YoutubeDL(object):
                     return
 
         if self.params.get('writeinfojson', False):
-            infofn = filename + u'.info.json'
+            infofn = os.path.splitext(filename)[0] + u'.info.json'
             self.report_writeinfojson(infofn)
             try:
                 json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle'])
@@ -706,13 +795,15 @@ class YoutubeDL(object):
 
     def download(self, url_list):
         """Download a given list of URLs."""
-        if len(url_list) > 1 and self.fixed_template():
+        if (len(url_list) > 1 and
+                '%' not in self.params['outtmpl']
+                and self.params.get('max_downloads') != 1):
             raise SameFileError(self.params['outtmpl'])
 
         for url in url_list:
             try:
                 #It also downloads the videos
-                videos = self.extract_info(url)
+                self.extract_info(url)
             except UnavailableVideoError:
                 self.report_error(u'unable to download video')
             except MaxDownloadsReached:
@@ -744,11 +835,26 @@ class YoutubeDL(object):
             except (IOError, OSError):
                 self.report_warning(u'Unable to remove downloaded video file')
 
+    def _make_archive_id(self, info_dict):
+        # Future-proof against any change in case
+        # and backwards compatibility with prior versions
+        extractor = info_dict.get('extractor_key')
+        if extractor is None:
+            if 'id' in info_dict:
+                extractor = info_dict.get('ie_key')  # key in a playlist
+        if extractor is None:
+            return None  # Incomplete video information
+        return extractor.lower() + u' ' + info_dict['id']
+
     def in_download_archive(self, info_dict):
         fn = self.params.get('download_archive')
         if fn is None:
             return False
-        vid_id = info_dict['extractor'] + u' ' + info_dict['id']
+
+        vid_id = self._make_archive_id(info_dict)
+        if vid_id is None:
+            return False  # Incomplete video information
+
         try:
             with locked_file(fn, 'r', encoding='utf-8') as archive_file:
                 for line in archive_file:
@@ -763,12 +869,15 @@ class YoutubeDL(object):
         fn = self.params.get('download_archive')
         if fn is None:
             return
-        vid_id = info_dict['extractor'] + u' ' + info_dict['id']
+        vid_id = self._make_archive_id(info_dict)
+        assert vid_id
         with locked_file(fn, 'a', encoding='utf-8') as archive_file:
             archive_file.write(vid_id + u'\n')
 
     @staticmethod
     def format_resolution(format, default='unknown'):
+        if format.get('vcodec') == 'none':
+            return 'audio only'
         if format.get('_resolution') is not None:
             return format['_resolution']
         if format.get('height') is not None:
@@ -781,23 +890,124 @@ class YoutubeDL(object):
         return res
 
     def list_formats(self, info_dict):
-        def line(format):
+        def format_note(fdict):
-            return (u'%-20s%-10s%-12s%s' % (
+            res = u''
+            if fdict.get('format_note') is not None:
+                res += fdict['format_note'] + u' '
+            if (fdict.get('vcodec') is not None and
+                    fdict.get('vcodec') != 'none'):
+                res += u'%-5s' % fdict['vcodec']
+            elif fdict.get('vbr') is not None:
+                res += u'video'
+            if fdict.get('vbr') is not None:
+                res += u'@%4dk' % fdict['vbr']
+            if fdict.get('acodec') is not None:
+                if res:
+                    res += u', '
+                res += u'%-5s' % fdict['acodec']
+            elif fdict.get('abr') is not None:
+                if res:
+                    res += u', '
+                res += 'audio'
+            if fdict.get('abr') is not None:
+                res += u'@%3dk' % fdict['abr']
+            if fdict.get('filesize') is not None:
+                if res:
+                    res += u', '
+                res += format_bytes(fdict['filesize'])
+            return res
+
+        def line(format, idlen=20):
+            return ((u'%-' + compat_str(idlen + 1) + u's%-10s%-12s%s') % (
                 format['format_id'],
                 format['ext'],
                 self.format_resolution(format),
-                format.get('format_note', ''),
+                format_note(format),
-                )
+            ))
-            )
 
         formats = info_dict.get('formats', [info_dict])
-        formats_s = list(map(line, formats))
+        idlen = max(len(u'format code'),
+                    max(len(f['format_id']) for f in formats))
+        formats_s = [line(f, idlen) for f in formats]
         if len(formats) > 1:
-            formats_s[0] += (' ' if formats[0].get('format_note') else '') + '(worst)'
+            formats_s[0] += (' ' if format_note(formats[0]) else '') + '(worst)'
-            formats_s[-1] += (' ' if formats[-1].get('format_note') else '') + '(best)'
+            formats_s[-1] += (' ' if format_note(formats[-1]) else '') + '(best)'
 
         header_line = line({
             'format_id': u'format code', 'ext': u'extension',
-            '_resolution': u'resolution', 'format_note': u'note'})
+            '_resolution': u'resolution', 'format_note': u'note'}, idlen=idlen)
         self.to_screen(u'[info] Available formats for %s:\n%s\n%s' %
                        (info_dict['id'], header_line, u"\n".join(formats_s)))
+
+    def urlopen(self, req):
+        """ Start an HTTP download """
+        return self._opener.open(req)
+
+    def print_debug_header(self):
+        if not self.params.get('verbose'):
+            return
+        write_string(u'[debug] youtube-dl version ' + __version__ + u'\n')
+        try:
+            sp = subprocess.Popen(
+                ['git', 'rev-parse', '--short', 'HEAD'],
+                stdout=subprocess.PIPE, stderr=subprocess.PIPE,
+                cwd=os.path.dirname(os.path.abspath(__file__)))
+            out, err = sp.communicate()
+            out = out.decode().strip()
+            if re.match('[0-9a-f]+', out):
+                write_string(u'[debug] Git HEAD: ' + out + u'\n')
+        except:
+            try:
+                sys.exc_clear()
+            except:
+                pass
+        write_string(u'[debug] Python version %s - %s' %
+                     (platform.python_version(), platform_name()) + u'\n')
+
+        proxy_map = {}
+        for handler in self._opener.handlers:
+            if hasattr(handler, 'proxies'):
+                proxy_map.update(handler.proxies)
+        write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n')
+
+    def _setup_opener(self):
+        timeout_val = self.params.get('socket_timeout')
+        timeout = 600 if timeout_val is None else float(timeout_val)
+
+        opts_cookiefile = self.params.get('cookiefile')
+        opts_proxy = self.params.get('proxy')
+
+        if opts_cookiefile is None:
+            self.cookiejar = compat_cookiejar.CookieJar()
+        else:
+            self.cookiejar = compat_cookiejar.MozillaCookieJar(
+                opts_cookiefile)
+            if os.access(opts_cookiefile, os.R_OK):
+                self.cookiejar.load()
+
+        cookie_processor = compat_urllib_request.HTTPCookieProcessor(
+            self.cookiejar)
+        if opts_proxy is not None:
+            if opts_proxy == '':
+                proxies = {}
+            else:
+                proxies = {'http': opts_proxy, 'https': opts_proxy}
+        else:
+            proxies = compat_urllib_request.getproxies()
+            # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
+            if 'http' in proxies and 'https' not in proxies:
+                proxies['https'] = proxies['http']
+        proxy_handler = compat_urllib_request.ProxyHandler(proxies)
+        https_handler = make_HTTPS_handler(
+            self.params.get('nocheckcertificate', False))
+        opener = compat_urllib_request.build_opener(
+            https_handler, proxy_handler, cookie_processor, YoutubeDLHandler())
+        # Delete the default user-agent header, which would otherwise apply in
+        # cases where our custom HTTP handler doesn't come into play
+        # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
+        opener.addheaders = []
+        self._opener = opener
+
+        # TODO remove this global modification
+        compat_urllib_request.install_opener(opener)
+        socket.setdefaulttimeout(timeout)

youtube_dl/__init__.py

@@ -34,50 +34,42 @@ __authors__  = (
     'Andras Elso',
     'Jelle van der Waa',
     'Marcin Cieślak',
+    'Anton Larionov',
+    'Takuya Tsuchida',
 )
 
 __license__ = 'Public Domain'
 
 import codecs
-import collections
 import getpass
 import optparse
 import os
 import random
 import re
 import shlex
-import socket
 import subprocess
 import sys
-import traceback
-import platform
 
 
 from .utils import (
-    compat_cookiejar,
     compat_print,
-    compat_str,
-    compat_urllib_request,
     DateRange,
     decodeOption,
     determine_ext,
     DownloadError,
     get_cachedir,
-    make_HTTPS_handler,
     MaxDownloadsReached,
-    platform_name,
     preferredencoding,
     SameFileError,
     std_headers,
     write_string,
-    YoutubeDLHandler,
 )
 from .update import update_self
-from .version import __version__
 from .FileDownloader import (
     FileDownloader,
 )
 from .extractor import gen_extractors
+from .version import __version__
 from .YoutubeDL import YoutubeDL
 from .PostProcessor import (
     FFmpegMetadataPP,
@@ -206,6 +198,9 @@ def parseOpts(overrideArguments=None):
     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=optparse.SUPPRESS_HELP)
 
 
     selection.add_option('--playlist-start',
@@ -214,7 +209,9 @@ def parseOpts(overrideArguments=None):
             dest='playlistend', metavar='NUMBER', help='playlist video to end at (default is last)', default=-1)
     selection.add_option('--match-title', dest='matchtitle', metavar='REGEX',help='download only matching titles (regex or caseless sub-string)')
     selection.add_option('--reject-title', dest='rejecttitle', metavar='REGEX',help='skip download for matching titles (regex or caseless sub-string)')
-    selection.add_option('--max-downloads', metavar='NUMBER', dest='max_downloads', help='Abort after downloading NUMBER files', default=None)
+    selection.add_option('--max-downloads', metavar='NUMBER',
+                         dest='max_downloads', type=int, default=None,
+                         help='Abort after downloading NUMBER files')
     selection.add_option('--min-filesize', metavar='SIZE', dest='min_filesize', help="Do not download any videos smaller than SIZE (e.g. 50k or 44.6m)", default=None)
     selection.add_option('--max-filesize', metavar='SIZE', dest='max_filesize', help="Do not download any videos larger than SIZE (e.g. 50k or 44.6m)", default=None)
     selection.add_option('--date', metavar='DATE', dest='date', help='download only videos uploaded in this date', default=None)
@@ -226,7 +223,7 @@ def parseOpts(overrideArguments=None):
                          default=None, type=int)
     selection.add_option('--download-archive', metavar='FILE',
                          dest='download_archive',
-                         help='Download only videos not present in the archive file. Record all downloaded videos in it.')
+                         help='Download only videos not present in the archive file. Record the IDs of all downloaded videos in it.')
 
 
     authentication.add_option('-u', '--username',
@@ -241,7 +238,7 @@ def parseOpts(overrideArguments=None):
 
     video_format.add_option('-f', '--format',
             action='store', dest='format', metavar='FORMAT', default='best',
-            help='video format code, specifiy the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
+            help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
     video_format.add_option('--all-formats',
             action='store_const', dest='format', help='download all available video formats', const='all')
     video_format.add_option('--prefer-free-formats',
@@ -306,6 +303,9 @@ def parseOpts(overrideArguments=None):
     verbosity.add_option('--get-format',
             action='store_true', dest='getformat',
             help='simulate, quiet but print output format', default=False)
+    verbosity.add_option('-j', '--dump-json',
+            action='store_true', dest='dumpjson',
+            help='simulate, quiet but print JSON information', default=False)
     verbosity.add_option('--newline',
             action='store_true', dest='progress_with_newline', help='output progress bar as new lines', default=False)
     verbosity.add_option('--no-progress',
@@ -320,7 +320,7 @@ def parseOpts(overrideArguments=None):
             help='print downloaded pages to debug problems(very verbose)')
     verbosity.add_option('--write-pages',
             action='store_true', dest='write_pages', default=False,
-            help='Write downloaded pages to files in the current directory')
+            help='Write downloaded intermediary pages to files in the current directory to debug problems')
     verbosity.add_option('--youtube-print-sig-code',
             action='store_true', dest='youtube_print_sig_code', default=False,
             help=optparse.SUPPRESS_HELP)
@@ -447,19 +447,6 @@ def _real_main(argv=None):
 
     parser, opts, args = parseOpts(argv)
 
-    # Open appropriate CookieJar
-    if opts.cookiefile is None:
-        jar = compat_cookiejar.CookieJar()
-    else:
-        try:
-            jar = compat_cookiejar.MozillaCookieJar(opts.cookiefile)
-            if os.access(opts.cookiefile, os.R_OK):
-                jar.load()
-        except (IOError, OSError) as err:
-            if opts.verbose:
-                traceback.print_exc()
-            write_string(u'ERROR: unable to open cookie file\n')
-            sys.exit(101)
     # Set user agent
     if opts.user_agent is not None:
         std_headers['User-Agent'] = opts.user_agent
@@ -491,8 +478,6 @@ def _real_main(argv=None):
     all_urls = batchurls + args
     all_urls = [url.strip() for url in all_urls]
 
-    opener = _setup_opener(jar=jar, opts=opts)
-
     extractors = gen_extractors()
 
     if opts.list_extractors:
@@ -547,7 +532,7 @@ def _real_main(argv=None):
     if opts.retries is not None:
         try:
             opts.retries = int(opts.retries)
-        except (TypeError, ValueError) as err:
+        except (TypeError, ValueError):
             parser.error(u'invalid retry count specified')
     if opts.buffersize is not None:
         numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
@@ -558,13 +543,13 @@ def _real_main(argv=None):
         opts.playliststart = int(opts.playliststart)
         if opts.playliststart <= 0:
             raise ValueError(u'Playlist start must be positive')
-    except (TypeError, ValueError) as err:
+    except (TypeError, ValueError):
         parser.error(u'invalid playlist start number specified')
     try:
         opts.playlistend = int(opts.playlistend)
         if opts.playlistend != -1 and (opts.playlistend <= 0 or opts.playlistend < opts.playliststart):
             raise ValueError(u'Playlist end must be greater than playlist start')
-    except (TypeError, ValueError) as err:
+    except (TypeError, ValueError):
         parser.error(u'invalid playlist end number specified')
     if opts.extractaudio:
         if opts.audioformat not in ['best', 'aac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']:
@@ -603,13 +588,12 @@ def _real_main(argv=None):
                      u' file! Use "%%(ext)s" instead of %r' %
                      determine_ext(outtmpl, u''))
 
-    # YoutubeDL
+    ydl_opts = {
-    ydl = YoutubeDL({
         'usenetrc': opts.usenetrc,
         'username': opts.username,
         'password': opts.password,
         'videopassword': opts.videopassword,
-        'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat),
+        'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.dumpjson),
         'forceurl': opts.geturl,
         'forcetitle': opts.gettitle,
         'forceid': opts.getid,
@@ -617,8 +601,9 @@ def _real_main(argv=None):
         'forcedescription': opts.getdescription,
         'forcefilename': opts.getfilename,
         'forceformat': opts.getformat,
+        'forcejson': opts.dumpjson,
         'simulate': opts.simulate,
-        'skip_download': (opts.skip_download or opts.simulate or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat),
+        'skip_download': (opts.skip_download or opts.simulate or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.dumpjson),
         'format': opts.format,
         'format_limit': opts.format_limit,
         'listformats': opts.listformats,
@@ -667,102 +652,47 @@ def _real_main(argv=None):
         'youtube_print_sig_code': opts.youtube_print_sig_code,
         'age_limit': opts.age_limit,
         'download_archive': opts.download_archive,
-        })
+        'cookiefile': opts.cookiefile,
+        'nocheckcertificate': opts.no_check_certificate,
+        'proxy': opts.proxy,
+        'socket_timeout': opts.socket_timeout,
+    }
+
+    with YoutubeDL(ydl_opts) as ydl:
+        ydl.print_debug_header()
+        ydl.add_default_info_extractors()
+
+        # PostProcessors
+        # Add the metadata pp first, the other pps will copy it
+        if opts.addmetadata:
+            ydl.add_post_processor(FFmpegMetadataPP())
+        if opts.extractaudio:
+            ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites))
+        if opts.recodevideo:
+            ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo))
+        if opts.embedsubtitles:
+            ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat))
+
+        # Update version
+        if opts.update_self:
+            update_self(ydl.to_screen, opts.verbose)
+
+        # Maybe do nothing
+        if len(all_urls) < 1:
+            if not opts.update_self:
+                parser.error(u'you must provide at least one URL')
+            else:
+                sys.exit()
 
-    if opts.verbose:
-        write_string(u'[debug] youtube-dl version ' + __version__ + u'\n')
         try:
-            sp = subprocess.Popen(
+            retcode = ydl.download(all_urls)
-                ['git', 'rev-parse', '--short', 'HEAD'],
+        except MaxDownloadsReached:
-                stdout=subprocess.PIPE, stderr=subprocess.PIPE,
+            ydl.to_screen(u'--max-download limit reached, aborting.')
-                cwd=os.path.dirname(os.path.abspath(__file__)))
+            retcode = 101
-            out, err = sp.communicate()
-            out = out.decode().strip()
-            if re.match('[0-9a-f]+', out):
-                write_string(u'[debug] Git HEAD: ' + out + u'\n')
-        except:
-            try:
-                sys.exc_clear()
-            except:
-                pass
-        write_string(u'[debug] Python version %s - %s' %(platform.python_version(), platform_name()) + u'\n')
-
-        proxy_map = {}
-        for handler in opener.handlers:
-            if hasattr(handler, 'proxies'):
-                proxy_map.update(handler.proxies)
-        write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n')
-
-    ydl.add_default_info_extractors()
-
-    # PostProcessors
-    # Add the metadata pp first, the other pps will copy it
-    if opts.addmetadata:
-        ydl.add_post_processor(FFmpegMetadataPP())
-    if opts.extractaudio:
-        ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites))
-    if opts.recodevideo:
-        ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo))
-    if opts.embedsubtitles:
-        ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat))
-
-    # Update version
-    if opts.update_self:
-        update_self(ydl.to_screen, opts.verbose)
-
-    # Maybe do nothing
-    if len(all_urls) < 1:
-        if not opts.update_self:
-            parser.error(u'you must provide at least one URL')
-        else:
-            sys.exit()
-
-    try:
-        retcode = ydl.download(all_urls)
-    except MaxDownloadsReached:
-        ydl.to_screen(u'--max-download limit reached, aborting.')
-        retcode = 101
-
-    # Dump cookie jar if requested
-    if opts.cookiefile is not None:
-        try:
-            jar.save()
-        except (IOError, OSError):
-            sys.exit(u'ERROR: unable to save cookie jar')
 
     sys.exit(retcode)
 
 
-def _setup_opener(jar=None, opts=None, timeout=300):
-    if opts is None:
-        FakeOptions = collections.namedtuple(
-            'FakeOptions', ['proxy', 'no_check_certificate'])
-        opts = FakeOptions(proxy=None, no_check_certificate=False)
-
-    cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
-    if opts.proxy is not None:
-        if opts.proxy == '':
-            proxies = {}
-        else:
-            proxies = {'http': opts.proxy, 'https': opts.proxy}
-    else:
-        proxies = compat_urllib_request.getproxies()
-        # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
-        if 'http' in proxies and 'https' not in proxies:
-            proxies['https'] = proxies['http']
-    proxy_handler = compat_urllib_request.ProxyHandler(proxies)
-    https_handler = make_HTTPS_handler(opts)
-    opener = compat_urllib_request.build_opener(
-        https_handler, proxy_handler, cookie_processor, YoutubeDLHandler())
-    # Delete the default user-agent header, which would otherwise apply in
-    # cases where our custom HTTP handler doesn't come into play
-    # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
-    opener.addheaders = []
-    compat_urllib_request.install_opener(opener)
-    socket.setdefaulttimeout(timeout)
-    return opener
-
-
 def main(argv=None):
     try:
         _real_main(argv)

youtube_dl/extractor/__init__.py

@@ -1,5 +1,6 @@
 from .appletrailers import AppleTrailersIE
 from .addanime import AddAnimeIE
+from .anitube import AnitubeIE
 from .archiveorg import ArchiveOrgIE
 from .ard import ARDIE
 from .arte import (
@@ -10,7 +11,7 @@ from .arte import (
 )
 from .auengine import AUEngineIE
 from .bambuser import BambuserIE, BambuserChannelIE
-from .bandcamp import BandcampIE
+from .bandcamp import BandcampIE, BandcampAlbumIE
 from .bliptv import BlipTVIE, BlipTVUserIE
 from .bloomberg import BloombergIE
 from .breakcom import BreakIE
@@ -19,12 +20,15 @@ from .c56 import C56IE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .cinemassacre import CinemassacreIE
+from .clipfish import ClipfishIE
+from .clipsyndicate import ClipsyndicateIE
 from .cnn import CNNIE
 from .collegehumor import CollegeHumorIE
-from .comedycentral import ComedyCentralIE
+from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
 from .condenast import CondeNastIE
 from .criterion import CriterionIE
 from .cspan import CSpanIE
+from .d8 import D8IE
 from .dailymotion import (
     DailymotionIE,
     DailymotionPlaylistIE,
@@ -68,6 +72,7 @@ from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
 from .hypem import HypemIE
 from .ign import IGNIE, OneUPIE
+from .imdb import ImdbIE
 from .ina import InaIE
 from .infoq import InfoQIE
 from .instagram import InstagramIE
@@ -96,11 +101,13 @@ from .nba import NBAIE
 from .nbc import NBCNewsIE
 from .newgrounds import NewgroundsIE
 from .nhl import NHLIE, NHLVideocenterIE
+from .niconico import NiconicoIE
 from .nowvideo import NowVideoIE
 from .ooyala import OoyalaIE
 from .orf import ORFIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
+from .podomatic import PodomaticIE
 from .pornhub import PornHubIE
 from .pornotube import PornotubeIE
 from .rbmaradio import RBMARadioIE
@@ -116,19 +123,24 @@ from .slashdot import SlashdotIE
 from .slideshare import SlideshareIE
 from .sohu import SohuIE
 from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE
-from .southparkstudios import SouthParkStudiosIE
+from .southparkstudios import (
+    SouthParkStudiosIE,
+    SouthparkDeIE,
+)
 from .space import SpaceIE
 from .spankwire import SpankwireIE
 from .spiegel import SpiegelIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .statigram import StatigramIE
 from .steam import SteamIE
+from .streamcloud import StreamcloudIE
 from .sztvhu import SztvHuIE
 from .teamcoco import TeamcocoIE
 from .techtalks import TechTalksIE
 from .ted import TEDIE
 from .tf1 import TF1IE
 from .thisav import ThisAVIE
+from .toutv import TouTvIE
 from .traileraddict import TrailerAddictIE
 from .trilulilu import TriluliluIE
 from .tube8 import Tube8IE
@@ -147,8 +159,13 @@ from .viddler import ViddlerIE
 from .videodetective import VideoDetectiveIE
 from .videofyme import VideofyMeIE
 from .videopremium import VideoPremiumIE
-from .vimeo import VimeoIE, VimeoChannelIE
+from .vimeo import (
+    VimeoIE,
+    VimeoChannelIE,
+    VimeoUserIE,
+)
 from .vine import VineIE
+from .viki import VikiIE
 from .vk import VKIE
 from .wat import WatIE
 from .websurg import WeBSurgIE
@@ -159,7 +176,11 @@ from .xhamster import XHamsterIE
 from .xnxx import XNXXIE
 from .xvideos import XVideosIE
 from .xtube import XTubeIE
-from .yahoo import YahooIE, YahooSearchIE
+from .yahoo import (
+    YahooIE,
+    YahooNewsIE,
+    YahooSearchIE,
+)
 from .youjizz import YouJizzIE
 from .youku import YoukuIE
 from .youporn import YouPornIE
@@ -176,6 +197,7 @@ from .youtube import (
     YoutubeTruncatedURLIE,
     YoutubeWatchLaterIE,
     YoutubeFavouritesIE,
+    YoutubeHistoryIE,
 )
 from .zdf import ZDFIE
 

youtube_dl/extractor/anitube.py

@@ -0,0 +1,53 @@
+import re
+
+from .common import InfoExtractor
+
+
+class AnitubeIE(InfoExtractor):
+    IE_NAME = u'anitube.se'
+    _VALID_URL = r'https?://(?:www\.)?anitube\.se/video/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.anitube.se/video/36621',
+        u'md5': u'59d0eeae28ea0bc8c05e7af429998d43',
+        u'file': u'36621.mp4',
+        u'info_dict': {
+            u'id': u'36621',
+            u'ext': u'mp4',
+            u'title': u'Recorder to Randoseru 01',
+        },
+        u'skip': u'Blocked in the US',
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        key = self._html_search_regex(r'http://www\.anitube\.se/embed/([A-Za-z0-9_-]*)',
+                                      webpage, u'key')
+
+        config_xml = self._download_xml('http://www.anitube.se/nuevo/econfig.php?key=%s' % key,
+                                                key)
+
+        video_title = config_xml.find('title').text
+
+        formats = []
+        video_url = config_xml.find('file')
+        if video_url is not None:
+            formats.append({
+                'format_id': 'sd',
+                'url': video_url.text,
+            })
+        video_url = config_xml.find('filehd')
+        if video_url is not None:
+            formats.append({
+                'format_id': 'hd',
+                'url': video_url.text,
+            })
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'formats': formats
+        }

youtube_dl/extractor/arte.py

@@ -1,7 +1,6 @@
 # encoding: utf-8
 import re
 import json
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -69,7 +68,7 @@ class ArteTvIE(InfoExtractor):
             lang = mobj.group('lang')
             return self._extract_liveweb(url, name, lang)
 
-        if re.search(self._LIVE_URL, video_id) is not None:
+        if re.search(self._LIVE_URL, url) is not None:
             raise ExtractorError(u'Arte live streams are not yet supported, sorry')
             # self.extractLiveStream(url)
             # return
@@ -78,8 +77,7 @@ class ArteTvIE(InfoExtractor):
         """Extract from videos.arte.tv"""
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
-        ref_xml = self._download_webpage(ref_xml_url, video_id, note=u'Downloading metadata')
+        ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'Downloading metadata')
-        ref_xml_doc = xml.etree.ElementTree.fromstring(ref_xml)
         config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
         config_xml_url = config_node.attrib['ref']
         config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
@@ -109,13 +107,12 @@ class ArteTvIE(InfoExtractor):
         """Extract form http://liveweb.arte.tv/"""
         webpage = self._download_webpage(url, name)
         video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, u'event id')
-        config_xml = self._download_webpage('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
+        config_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
                                             video_id, u'Downloading information')
-        config_doc = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
         event_doc = config_doc.find('event')
         url_node = event_doc.find('video').find('urlHd')
         if url_node is None:
-            url_node = video_doc.find('urlSd')
+            url_node = event_doc.find('urlSd')
 
         return {'id': video_id,
                 'title': event_doc.find('name%s' % lang.capitalize()).text,

youtube_dl/extractor/auengine.py

@@ -1,10 +1,10 @@
-import os.path
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
-    compat_urllib_parse_urlparse,
+    determine_ext,
+    ExtractorError,
 )
 
 class AUEngineIE(InfoExtractor):
@@ -25,22 +25,25 @@ class AUEngineIE(InfoExtractor):
         title = self._html_search_regex(r'<title>(?P<title>.+?)</title>',
                 webpage, u'title')
         title = title.strip()
-        links = re.findall(r'[^A-Za-z0-9]?(?:file|url):\s*["\'](http[^\'"&]*)', webpage)
+        links = re.findall(r'\s(?:file|url):\s*["\']([^\'"]+)["\']', webpage)
-        links = [compat_urllib_parse.unquote(l) for l in links]
+        links = map(compat_urllib_parse.unquote, links)
+
+        thumbnail = None
+        video_url = None
         for link in links:
-            root, pathext = os.path.splitext(compat_urllib_parse_urlparse(link).path)
+            if link.endswith('.png'):
-            if pathext == '.png':
                 thumbnail = link
-            elif pathext == '.mp4':
+            elif '/videos/' in link:
-                url = link
+                video_url = link
-                ext = pathext
+        if not video_url:
+            raise ExtractorError(u'Could not find video URL')
+        ext = u'.' + determine_ext(video_url)
         if ext == title[-len(ext):]:
             title = title[:-len(ext)]
-        ext = ext[1:]
+
-        return [{
+        return {
             'id':        video_id,
-            'url':       url,
+            'url':       video_url,
-            'ext':       ext,
             'title':     title,
             'thumbnail': thumbnail,
-        }]
+        }

youtube_dl/extractor/bambuser.py

@@ -15,7 +15,8 @@ class BambuserIE(InfoExtractor):
 
     _TEST = {
         u'url': u'http://bambuser.com/v/4050584',
-        u'md5': u'fba8f7693e48fd4e8641b3fd5539a641',
+        # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388
+        #u'md5': u'fba8f7693e48fd4e8641b3fd5539a641',
         u'info_dict': {
             u'id': u'4050584',
             u'ext': u'flv',
@@ -24,6 +25,11 @@ class BambuserIE(InfoExtractor):
             u'uploader': u'pixelversity',
             u'uploader_id': u'344706',
         },
+        u'params': {
+            # It doesn't respect the 'Range' header, it would download the whole video
+            # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59
+            u'skip_download': True,
+        },
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/bandcamp.py

@@ -3,13 +3,16 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    compat_str,
+    compat_urlparse,
     ExtractorError,
 )
 
 
 class BandcampIE(InfoExtractor):
+    IE_NAME = u'Bandcamp'
     _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
-    _TEST = {
+    _TESTS = [{
         u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
         u'file': u'1812978515.mp3',
         u'md5': u'cdeb30cdae1921719a3cbcab696ef53c',
@@ -17,7 +20,7 @@ class BandcampIE(InfoExtractor):
             u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad"
         },
         u'skip': u'There is a limit of 200 free downloads / month for the test song'
-    }
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -26,6 +29,23 @@ class BandcampIE(InfoExtractor):
         # We get the link to the free download page
         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
         if m_download is None:
+            m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
+        if m_trackinfo:
+            json_code = m_trackinfo.group(1)
+            data = json.loads(json_code)
+
+            for d in data:
+                formats = [{
+                    'format_id': 'format_id',
+                    'url': format_url,
+                    'ext': format_id.partition('-')[0]
+                } for format_id, format_url in sorted(d['file'].items())]
+                return {
+                    'id': compat_str(d['id']),
+                    'title': d['title'],
+                    'formats': formats,
+                }
+        else:
             raise ExtractorError(u'No free songs found')
 
         download_link = m_download.group(1)
@@ -61,3 +81,49 @@ class BandcampIE(InfoExtractor):
                       }
 
         return [track_info]
+
+
+class BandcampAlbumIE(InfoExtractor):
+    IE_NAME = u'Bandcamp:album'
+    _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)'
+
+    _TEST = {
+        u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
+        u'playlist': [
+            {
+                u'file': u'1353101989.mp3',
+                u'md5': u'39bc1eded3476e927c724321ddf116cf',
+                u'info_dict': {
+                    u'title': u'Intro',
+                }
+            },
+            {
+                u'file': u'38097443.mp3',
+                u'md5': u'1a2c32e2691474643e912cc6cd4bffaa',
+                u'info_dict': {
+                    u'title': u'Kero One - Keep It Alive (Blazo remix)',
+                }
+            },
+        ],
+        u'params': {
+            u'playlistend': 2
+        },
+        u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        title = mobj.group('title')
+        webpage = self._download_webpage(url, title)
+        tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
+        if not tracks_paths:
+            raise ExtractorError(u'The page doesn\'t contain any track')
+        entries = [
+            self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
+            for t_path in tracks_paths]
+        title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title')
+        return {
+            '_type': 'playlist',
+            'title': title,
+            'entries': entries,
+        }

youtube_dl/extractor/brightcove.py

@@ -75,16 +75,22 @@ class BrightcoveIE(InfoExtractor):
         params = {'flashID': object_doc.attrib['id'],
                   'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
                   }
-        playerKey = find_xpath_attr(object_doc, './param', 'name', 'playerKey')
+        def find_param(name):
+            node = find_xpath_attr(object_doc, './param', 'name', name)
+            if node is not None:
+                return node.attrib['value']
+            return None
+        playerKey = find_param('playerKey')
         # Not all pages define this value
         if playerKey is not None:
-            params['playerKey'] = playerKey.attrib['value']
+            params['playerKey'] = playerKey
-        videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer')
+        # The three fields hold the id of the video
+        videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID')
         if videoPlayer is not None:
-            params['@videoPlayer'] = videoPlayer.attrib['value']
+            params['@videoPlayer'] = videoPlayer
-        linkBase = find_xpath_attr(object_doc, './param', 'name', 'linkBaseURL')
+        linkBase = find_param('linkBaseURL')
         if linkBase is not None:
-            params['linkBaseURL'] = linkBase.attrib['value']
+            params['linkBaseURL'] = linkBase
         data = compat_urllib_parse.urlencode(params)
         return cls._FEDERATED_URL_TEMPLATE % data
 

youtube_dl/extractor/canalplus.py

@@ -1,10 +1,10 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import unified_strdate
 
+
 class CanalplusIE(InfoExtractor):
     _VALID_URL = r'https?://(www\.canalplus\.fr/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>\d+))'
     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s'
@@ -25,16 +25,15 @@ class CanalplusIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = mobj.groupdict().get('id')
         if video_id is None:
             webpage = self._download_webpage(url, mobj.group('path'))
             video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id')
         info_url = self._VIDEO_INFO_TEMPLATE % video_id
-        info_page = self._download_webpage(info_url,video_id, 
+        doc = self._download_xml(info_url,video_id, 
                                            u'Downloading video info')
 
         self.report_extraction(video_id)
-        doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8'))
         video_info = [video for video in doc if video.find('ID').text == video_id][0]
         infos = video_info.find('INFOS')
         media = video_info.find('MEDIA')

youtube_dl/extractor/clipfish.py

@@ -0,0 +1,58 @@
+import re
+import time
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class ClipfishIE(InfoExtractor):
+    IE_NAME = u'clipfish'
+
+    _VALID_URL = r'^https?://(?:www\.)?clipfish\.de/.*?/video/(?P<id>[0-9]+)/'
+    _TEST = {
+        u'url': u'http://www.clipfish.de/special/game-trailer/video/3966754/fifa-14-e3-2013-trailer/',
+        u'file': u'3966754.mp4',
+        u'md5': u'2521cd644e862936cf2e698206e47385',
+        u'info_dict': {
+            u'title': u'FIFA 14 - E3 2013 Trailer',
+            u'duration': 82,
+        },
+        u'skip': 'Blocked in the US'
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+
+        info_url = ('http://www.clipfish.de/devxml/videoinfo/%s?ts=%d' %
+                    (video_id, int(time.time())))
+        doc = self._download_xml(
+            info_url, video_id, note=u'Downloading info page')
+        title = doc.find('title').text
+        video_url = doc.find('filename').text
+        if video_url is None:
+            xml_bytes = xml.etree.ElementTree.tostring(doc)
+            raise ExtractorError(u'Cannot find video URL in document %r' %
+                                 xml_bytes)
+        thumbnail = doc.find('imageurl').text
+        duration_str = doc.find('duration').text
+        m = re.match(
+            r'^(?P<hours>[0-9]+):(?P<minutes>[0-9]{2}):(?P<seconds>[0-9]{2}):(?P<ms>[0-9]*)$',
+            duration_str)
+        if m:
+            duration = (
+                (int(m.group('hours')) * 60 * 60) +
+                (int(m.group('minutes')) * 60) +
+                (int(m.group('seconds')))
+            )
+        else:
+            duration = None
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'thumbnail': thumbnail,
+            'duration': duration,
+        }

youtube_dl/extractor/clipsyndicate.py

@@ -0,0 +1,52 @@
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    find_xpath_attr,
+)
+
+
+class ClipsyndicateIE(InfoExtractor):
+    _VALID_URL = r'http://www\.clipsyndicate\.com/video/play(list/\d+)?/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.clipsyndicate.com/video/play/4629301/brick_briscoe',
+        u'md5': u'4d7d549451bad625e0ff3d7bd56d776c',
+        u'info_dict': {
+            u'id': u'4629301',
+            u'ext': u'mp4',
+            u'title': u'Brick Briscoe',
+            u'duration': 612,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        js_player = self._download_webpage(
+            'http://eplayer.clipsyndicate.com/embed/player.js?va_id=%s' % video_id,
+            video_id, u'Downlaoding player')
+        # it includes a required token
+        flvars = self._search_regex(r'flvars: "(.*?)"', js_player, u'flvars')
+
+        playlist_page = self._download_webpage(
+            'http://eplayer.clipsyndicate.com/osmf/playlist?%s' % flvars,
+            video_id, u'Downloading video info') 
+        # Fix broken xml
+        playlist_page = re.sub('&', '&amp;', playlist_page)
+        pdoc = xml.etree.ElementTree.fromstring(playlist_page.encode('utf-8'))
+
+        track_doc = pdoc.find('trackList/track')
+        def find_param(name):
+            node = find_xpath_attr(track_doc, './/param', 'name', name)
+            if node is not None:
+                return node.attrib['value']
+
+        return {
+            'id': video_id,
+            'title': find_param('title'),
+            'url': track_doc.find('location').text,
+            'thumbnail': find_param('thumbnail'),
+            'duration': int(find_param('duration')),
+        }

youtube_dl/extractor/cnn.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import determine_ext
@@ -33,8 +32,7 @@ class CNNIE(InfoExtractor):
         path = mobj.group('path')
         page_title = mobj.group('title')
         info_url = u'http://cnn.com/video/data/3.0/%s/index.xml' % path
-        info_xml = self._download_webpage(info_url, page_title)
+        info = self._download_xml(info_url, page_title)
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
 
         formats = []
         for f in info.findall('files/file'):

youtube_dl/extractor/collegehumor.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -46,11 +45,10 @@ class CollegeHumorIE(InfoExtractor):
 
         self.report_extraction(video_id)
         xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
-        metaXml = self._download_webpage(xmlUrl, video_id,
+        mdoc = self._download_xml(xmlUrl, video_id,
                                          u'Downloading info XML',
                                          u'Unable to download video info XML')
 
-        mdoc = xml.etree.ElementTree.fromstring(metaXml)
         try:
             videoNode = mdoc.findall('./video')[0]
             youtubeIdNode = videoNode.find('./youtubeID')
@@ -65,16 +63,13 @@ class CollegeHumorIE(InfoExtractor):
 
         if next_url.endswith(u'manifest.f4m'):
             manifest_url = next_url + '?hdcore=2.10.3'
-            manifestXml = self._download_webpage(manifest_url, video_id,
+            adoc = self._download_xml(manifest_url, video_id,
                                          u'Downloading XML manifest',
                                          u'Unable to download video info XML')
 
-            adoc = xml.etree.ElementTree.fromstring(manifestXml)
             try:
-                media_node = adoc.findall('./{http://ns.adobe.com/f4m/1.0}media')[0]
-                node_id = media_node.attrib['url']
                 video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text
-            except IndexError as err:
+            except IndexError:
                 raise ExtractorError(u'Invalid manifest file')
             url_pr = compat_urllib_parse_urlparse(info['thumbnail'])
             info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')

youtube_dl/extractor/comedycentral.py

@@ -1,7 +1,7 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
+from .mtv import MTVIE, _media_xml_tag
 from ..utils import (
     compat_str,
     compat_urllib_parse,
@@ -11,7 +11,37 @@ from ..utils import (
 )
 
 
-class ComedyCentralIE(InfoExtractor):
+class ComedyCentralIE(MTVIE):
+    _VALID_URL = r'http://www.comedycentral.com/(video-clips|episodes|cc-studios)/(?P<title>.*)'
+    _FEED_URL = u'http://comedycentral.com/feeds/mrss/'
+
+    _TEST = {
+        u'url': u'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
+        u'md5': u'4167875aae411f903b751a21f357f1ee',
+        u'info_dict': {
+            u'id': u'cef0cbb3-e776-4bc9-b62e-8016deccb354',
+            u'ext': u'mp4',
+            u'title': u'Uncensored - Greg Fitzsimmons - Too Good of a Mother',
+            u'description': u'After a certain point, breastfeeding becomes c**kblocking.',
+        },
+    }
+    # Overwrite MTVIE properties we don't want
+    _TESTS = []
+
+    def _get_thumbnail_url(self, uri, itemdoc):
+        search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
+        return itemdoc.find(search_path).attrib['url']
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        title = mobj.group('title')
+        webpage = self._download_webpage(url, title)
+        mgid = self._search_regex(r'data-mgid="(?P<mgid>mgid:.*?)"',
+                                  webpage, u'mgid')
+        return self._get_videos_info(mgid)
+
+
+class ComedyCentralShowsIE(InfoExtractor):
     IE_DESC = u'The Daily Show / Colbert Report'
     # urls can be abbreviations like :thedailyshow or :colbert
     # urls for episodes like:
@@ -127,13 +157,12 @@ class ComedyCentralIE(InfoExtractor):
 
         uri = mMovieParams[0][1]
         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
-        indexXml = self._download_webpage(indexUrl, epTitle,
+        idoc = self._download_xml(indexUrl, epTitle,
                                           u'Downloading show index',
                                           u'unable to download episode index')
 
         results = []
 
-        idoc = xml.etree.ElementTree.fromstring(indexXml)
         itemEls = idoc.findall('.//item')
         for partNum,itemEl in enumerate(itemEls):
             mediaId = itemEl.findall('./guid')[0].text
@@ -144,10 +173,9 @@ class ComedyCentralIE(InfoExtractor):
 
             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
                         compat_urllib_parse.urlencode({'uri': mediaId}))
-            configXml = self._download_webpage(configUrl, epTitle,
+            cdoc = self._download_xml(configUrl, epTitle,
                                                u'Downloading configuration for %s' % shortMediaId)
 
-            cdoc = xml.etree.ElementTree.fromstring(configXml)
             turls = []
             for rendition in cdoc.findall('.//rendition'):
                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)

youtube_dl/extractor/common.py

@@ -4,11 +4,11 @@ import re
 import socket
 import sys
 import netrc
+import xml.etree.ElementTree
 
 from ..utils import (
     compat_http_client,
     compat_urllib_error,
-    compat_urllib_request,
     compat_str,
 
     clean_html,
@@ -19,6 +19,7 @@ from ..utils import (
     unescapeHTML,
 )
 
+
 class InfoExtractor(object):
     """Information Extractor class.
 
@@ -71,6 +72,11 @@ class InfoExtractor(object):
                                 ("3D" or "DASH video")
                     * width     Width of the video, if known
                     * height    Height of the video, if known
+                    * abr       Average audio bitrate in KBit/s
+                    * acodec    Name of the audio codec in use
+                    * vbr       Average video bitrate in KBit/s
+                    * vcodec    Name of the video codec in use
+                    * filesize  The number of bytes, if known in advance
     webpage_url:    The url to the video webpage, if given to youtube-dl it
                     should allow to get the same result again. (It will be set
                     by YoutubeDL if it's missing)
@@ -152,7 +158,7 @@ class InfoExtractor(object):
         elif note is not False:
             self.to_screen(u'%s: %s' % (video_id, note))
         try:
-            return compat_urllib_request.urlopen(url_or_request)
+            return self._downloader.urlopen(url_or_request)
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
             if errnote is None:
                 errnote = u'Unable to download webpage'
@@ -204,6 +210,12 @@ class InfoExtractor(object):
         """ Returns the data of the page as a string """
         return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
 
+    def _download_xml(self, url_or_request, video_id,
+                      note=u'Downloading XML', errnote=u'Unable to download XML'):
+        """Return the xml as an xml.etree.ElementTree.Element"""
+        xml_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
+
     def to_screen(self, msg):
         """Print msg to screen, prefixing it with '[ie_name]'"""
         self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
@@ -225,12 +237,14 @@ class InfoExtractor(object):
         self.to_screen(u'Logging in')
 
     #Methods for following #608
-    def url_result(self, url, ie=None):
+    def url_result(self, url, ie=None, video_id=None):
         """Returns a url that points to a page that should be processed"""
         #TODO: ie should be the class used for getting the info
         video_info = {'_type': 'url',
                       'url': url,
                       'ie_key': ie}
+        if video_id is not None:
+            video_info['id'] = video_id
         return video_info
     def playlist_result(self, entries, playlist_id=None, playlist_title=None):
         """Returns a playlist"""
@@ -315,13 +329,19 @@ class InfoExtractor(object):
 
     # Helper functions for extracting OpenGraph info
     @staticmethod
-    def _og_regex(prop):
+    def _og_regexes(prop):
-        return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop)
+        content_re = r'content=(?:"([^>]+?)"|\'(.+?)\')'
+        property_re = r'property=[\'"]og:%s[\'"]' % re.escape(prop)
+        template = r'<meta[^>]+?%s[^>]+?%s'
+        return [
+            template % (property_re, content_re),
+            template % (content_re, property_re),
+        ]
 
     def _og_search_property(self, prop, html, name=None, **kargs):
         if name is None:
             name = 'OpenGraph %s' % prop
-        escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
+        escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs)
         if escaped is None:
             return None
         return unescapeHTML(escaped)
@@ -336,10 +356,21 @@ 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_regex('video')]
+        regexes = self._og_regexes('video')
-        if secure: regexes.insert(0, self._og_regex('video:secure_url'))
+        if secure: regexes = self._og_regexes('video:secure_url') + regexes
         return self._html_search_regex(regexes, html, name, **kargs)
 
+    def _html_search_meta(self, name, html, display_name=None):
+        if display_name is None:
+            display_name = name
+        return self._html_search_regex(
+            r'''(?ix)<meta(?=[^>]+(?:name|property)=["\']%s["\'])
+                    [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
+            html, display_name, fatal=False)
+
+    def _dc_search_uploader(self, html):
+        return self._html_search_meta('dc.creator', html, 'uploader')
+
     def _rta_search(self, html):
         # See http://www.rtalabel.org/index.php?content=howtofaq#single
         if re.search(r'(?ix)<meta\s+name="rating"\s+'
@@ -348,6 +379,23 @@ class InfoExtractor(object):
             return 18
         return 0
 
+    def _media_rating_search(self, html):
+        # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/
+        rating = self._html_search_meta('rating', html)
+
+        if not rating:
+            return None
+
+        RATING_TABLE = {
+            'safe for kids': 0,
+            'general': 8,
+            '14 years': 14,
+            'mature': 17,
+            'restricted': 19,
+        }
+        return RATING_TABLE.get(rating.lower(), None)
+
+
 
 class SearchInfoExtractor(InfoExtractor):
     """

youtube_dl/extractor/d8.py

@@ -0,0 +1,22 @@
+# encoding: utf-8
+from .canalplus import CanalplusIE
+
+
+class D8IE(CanalplusIE):
+    _VALID_URL = r'https?://www\.d8\.tv/.*?/(?P<path>.*)'
+    _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/d8/%s'
+    IE_NAME = u'd8.tv'
+
+    _TEST = {
+        u'url': u'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
+        u'file': u'966289.flv',
+        u'info_dict': {
+            u'title': u'Campagne intime - Documentaire exceptionnel',
+            u'description': u'md5:d2643b799fb190846ae09c61e59a859f',
+            u'upload_date': u'20131108',
+        },
+        u'params': {
+            # rtmp
+            u'skip_download': True,
+        },
+    }

youtube_dl/extractor/dailymotion.py

@@ -186,7 +186,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
             webpage = self._download_webpage(request,
                                              id, u'Downloading page %s' % pagenum)
 
-            playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
+            playlist_el = get_element_by_attribute(u'class', u'row video_list', webpage)
             video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el))
 
             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:

youtube_dl/extractor/daum.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -32,14 +31,12 @@ class DaumIE(InfoExtractor):
         full_id = self._search_regex(r'<link rel="video_src" href=".+?vid=(.+?)"',
             webpage, u'full id')
         query = compat_urllib_parse.urlencode({'vid': full_id})
-        info_xml = self._download_webpage(
+        info = self._download_xml(
             'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
             u'Downloading video info')
-        urls_xml = self._download_webpage(
+        urls = self._download_xml(
             'http://videofarm.daum.net/controller/api/open/v1_2/MovieData.apixml?' + query,
             video_id, u'Downloading video formats info')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
-        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
 
         self.to_screen(u'%s: Getting video urls' % video_id)
         formats = []
@@ -49,10 +46,9 @@ class DaumIE(InfoExtractor):
                 'vid': full_id,
                 'profile': profile,
             })
-            url_xml = self._download_webpage(
+            url_doc = self._download_xml(
                 'http://videofarm.daum.net/controller/api/open/v1_2/MovieLocation.apixml?' + format_query,
                 video_id, note=False)
-            url_doc = xml.etree.ElementTree.fromstring(url_xml.encode('utf-8'))
             format_url = url_doc.find('result/url').text
             formats.append({
                 'url': format_url,

youtube_dl/extractor/dreisat.py

@@ -1,7 +1,6 @@
 # coding: utf-8
 
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -30,8 +29,7 @@ class DreiSatIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
-        details_xml = self._download_webpage(details_url, video_id, note=u'Downloading video details')
+        details_doc = self._download_xml(details_url, video_id, note=u'Downloading video details')
-        details_doc = xml.etree.ElementTree.fromstring(details_xml.encode('utf-8'))
 
         thumbnail_els = details_doc.findall('.//teaserimage')
         thumbnails = [{

youtube_dl/extractor/ebaumsworld.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import determine_ext
@@ -21,9 +20,8 @@ class EbaumsWorldIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        config_xml = self._download_webpage(
+        config = self._download_xml(
             'http://www.ebaumsworld.com/video/player/%s' % video_id, video_id)
-        config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
         video_url = config.find('file').text
 
         return {

youtube_dl/extractor/eighttracks.py

@@ -1,4 +1,3 @@
-import itertools
 import json
 import random
 import re

youtube_dl/extractor/escapist.py

@@ -11,11 +11,11 @@ from ..utils import (
 
 
 class EscapistIE(InfoExtractor):
-    _VALID_URL = r'^(https?://)?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$'
+    _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$'
     _TEST = {
         u'url': u'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
         u'file': u'6618-Breaking-Down-Baldurs-Gate.mp4',
-        u'md5': u'c6793dbda81388f4264c1ba18684a74d',
+        u'md5': u'ab3a706c681efca53f0a35f1415cf0d1',
         u'info_dict': {
             u"description": u"Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.", 
             u"uploader": u"the-escapist-presents", 
@@ -25,50 +25,60 @@ class EscapistIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
         showName = mobj.group('showname')
         videoId = mobj.group('episode')
 
         self.report_extraction(videoId)
         webpage = self._download_webpage(url, videoId)
 
-        videoDesc = self._html_search_regex('<meta name="description" content="([^"]*)"',
+        videoDesc = self._html_search_regex(
+            r'<meta name="description" content="([^"]*)"',
             webpage, u'description', fatal=False)
 
-        playerUrl = self._og_search_video_url(webpage, name='player url')
+        playerUrl = self._og_search_video_url(webpage, name=u'player URL')
 
-        title = self._html_search_regex('<meta name="title" content="([^"]*)"',
+        title = self._html_search_regex(
-            webpage, u'player url').split(' : ')[-1]
+            r'<meta name="title" content="([^"]*)"',
+            webpage, u'title').split(' : ')[-1]
 
-        configUrl = self._search_regex('config=(.*)$', playerUrl, u'config url')
+        configUrl = self._search_regex('config=(.*)$', playerUrl, u'config URL')
         configUrl = compat_urllib_parse.unquote(configUrl)
 
-        configJSON = self._download_webpage(configUrl, videoId,
+        formats = []
-                                            u'Downloading configuration',
-                                            u'unable to download configuration')
 
-        # Technically, it's JavaScript, not JSON
+        def _add_format(name, cfgurl):
-        configJSON = configJSON.replace("'", '"')
+            configJSON = self._download_webpage(
+                cfgurl, videoId,
+                u'Downloading ' + name + ' configuration',
+                u'Unable to download ' + name + ' configuration')
 
+            # Technically, it's JavaScript, not JSON
+            configJSON = configJSON.replace("'", '"')
+
+            try:
+                config = json.loads(configJSON)
+            except (ValueError,) as err:
+                raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err))
+            playlist = config['playlist']
+            formats.append({
+                'url': playlist[1]['url'],
+                'format_id': name,
+            })
+
+        _add_format(u'normal', configUrl)
+        hq_url = (configUrl +
+                  ('&hq=1' if '?' in configUrl else configUrl + '?hq=1'))
         try:
-            config = json.loads(configJSON)
+            _add_format(u'hq', hq_url)
-        except (ValueError,) as err:
+        except ExtractorError:
-            raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err))
+            pass  # That's fine, we'll just use normal quality
 
-        playlist = config['playlist']
+        return {
-        videoUrl = playlist[1]['url']
-
-        info = {
             'id': videoId,
-            'url': videoUrl,
+            'formats': formats,
             'uploader': showName,
-            'upload_date': None,
             'title': title,
-            'ext': 'mp4',
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': videoDesc,
             'player_url': playerUrl,
         }
-
-        return [info]

youtube_dl/extractor/facebook.py

@@ -1,5 +1,4 @@
 import json
-import netrc
 import re
 import socket
 

youtube_dl/extractor/faz.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -28,9 +27,8 @@ class FazIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         config_xml_url = self._search_regex(r'writeFLV\(\'(.+?)\',', webpage,
             u'config xml url')
-        config_xml = self._download_webpage(config_xml_url, video_id,
+        config = self._download_xml(config_xml_url, video_id,
             u'Downloading config xml')
-        config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
 
         encodings = config.find('ENCODINGS')
         formats = []

youtube_dl/extractor/fktv.py

@@ -39,7 +39,6 @@ class FKTVIE(InfoExtractor):
         for i, _ in enumerate(files, 1):
             video_id = '%04d%d' % (episode, i)
             video_url = 'http://dl%d.fernsehkritik.tv/fernsehkritik%d%s.flv' % (server, episode, '' if i == 1 else '-%d' % i)
-            video_title = 'Fernsehkritik %d.%d' % (episode, i)
             videos.append({
                 'id': video_id,
                 'url': video_url,

youtube_dl/extractor/francetv.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 import json
 
 from .common import InfoExtractor
@@ -11,11 +10,10 @@ from ..utils import (
 
 class FranceTVBaseInfoExtractor(InfoExtractor):
     def _extract_video(self, video_id):
-        xml_desc = self._download_webpage(
+        info = self._download_xml(
             'http://www.francetvinfo.fr/appftv/webservices/video/'
             'getInfosOeuvre.php?id-diffusion='
             + video_id, video_id, 'Downloading XML config')
-        info = xml.etree.ElementTree.fromstring(xml_desc.encode('utf-8'))
 
         manifest_url = info.find('videos/video/url').text
         video_url = manifest_url.replace('manifest.f4m', 'index_2_av.m3u8')

youtube_dl/extractor/gamekings.py

@@ -1,9 +1,6 @@
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    determine_ext,
-)
 
 
 class GamekingsIE(InfoExtractor):
@@ -11,7 +8,8 @@ class GamekingsIE(InfoExtractor):
     _TEST = {
         u"url": u"http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/",
         u'file': u'20130811.mp4',
-        u'md5': u'2f32b1f7b80fdc5cb616efb4f387f8a3',
+        # MD5 is flaky, seems to change regularly
+        #u'md5': u'2f32b1f7b80fdc5cb616efb4f387f8a3',
         u'info_dict': {
             u"title": u"Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review",
             u"description": u"Melle en Steven hebben voor de review een week in de rechtbank doorbracht met Phoenix Wright: Ace Attorney - Dual Destinies.",

youtube_dl/extractor/gamespot.py

@@ -24,7 +24,7 @@ class GameSpotIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        page_id = video_id = mobj.group('page_id')
+        page_id = mobj.group('page_id')
         webpage = self._download_webpage(url, page_id)
         data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video')
         data_video = json.loads(unescapeHTML(data_video_json))

youtube_dl/extractor/generic.py

@@ -162,6 +162,16 @@ class GenericIE(InfoExtractor):
             raise ExtractorError(u'Failed to download URL: %s' % url)
 
         self.report_extraction(video_id)
+
+        # it's tempting to parse this further, but you would
+        # have to take into account all the variations like
+        #   Video Title - Site Name
+        #   Site Name | Video Title
+        #   Video Title - Tagline | Site Name
+        # and so on and so forth; it's just not practical
+        video_title = self._html_search_regex(r'<title>(.*)</title>',
+            webpage, u'video title', default=u'video', flags=re.DOTALL)
+
         # Look for BrightCove:
         bc_url = BrightcoveIE._extract_brightcove_url(webpage)
         if bc_url is not None:
@@ -177,17 +187,29 @@ class GenericIE(InfoExtractor):
             return self.url_result(surl, 'Vimeo')
 
         # Look for embedded YouTube player
-        mobj = re.search(
+        matches = re.findall(
-            r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?youtube.com/embed/.+?)\1', webpage)
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube.com/embed/.+?)\1', webpage)
-        if mobj:
+        if matches:
-            surl = unescapeHTML(mobj.group(u'url'))
+            urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube')
-            return self.url_result(surl, 'Youtube')
+                     for tuppl in matches]
+            return self.playlist_result(
+                urlrs, playlist_id=video_id, playlist_title=video_title)
+
+        # Look for embedded Dailymotion player
+        matches = re.findall(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion.com/embed/video/.+?)\1', webpage)
+        if matches:
+            urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion')
+                     for tuppl in matches]
+            return self.playlist_result(
+                urlrs, playlist_id=video_id, playlist_title=video_title)
 
         # Look for Bandcamp pages with custom domain
         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
         if mobj is not None:
             burl = unescapeHTML(mobj.group(1))
-            return self.url_result(burl, 'Bandcamp')
+            # Don't set the extractor because it can be a track url or an album
+            return self.url_result(burl)
 
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
@@ -196,7 +218,7 @@ class GenericIE(InfoExtractor):
             mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
         if mobj is None:
             # Broaden the search a little bit: JWPlayer JS loader
-            mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"&]*)', webpage)
+            mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"]*)', webpage)
         if mobj is None:
             # Try to find twitter cards info
             mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
@@ -223,27 +245,16 @@ class GenericIE(InfoExtractor):
         video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
 
         # here's a fun little line of code for you:
-        video_extension = os.path.splitext(video_id)[1][1:]
         video_id = os.path.splitext(video_id)[0]
 
-        # it's tempting to parse this further, but you would
-        # have to take into account all the variations like
-        #   Video Title - Site Name
-        #   Site Name | Video Title
-        #   Video Title - Tagline | Site Name
-        # and so on and so forth; it's just not practical
-        video_title = self._html_search_regex(r'<title>(.*)</title>',
-            webpage, u'video title', default=u'video', flags=re.DOTALL)
-
         # video uploader is domain name
         video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*',
             url, u'video uploader')
 
-        return [{
+        return {
             'id':       video_id,
             'url':      video_url,
             'uploader': video_uploader,
             'upload_date':  None,
             'title':    video_title,
-            'ext':      video_extension,
+        }
-        }]

youtube_dl/extractor/howcast.py

@@ -8,7 +8,7 @@ class HowcastIE(InfoExtractor):
     _TEST = {
         u'url': u'http://www.howcast.com/videos/390161-How-to-Tie-a-Square-Knot-Properly',
         u'file': u'390161.mp4',
-        u'md5': u'1d7ba54e2c9d7dc6935ef39e00529138',
+        u'md5': u'8b743df908c42f60cf6496586c7f12c3',
         u'info_dict': {
             u"description": u"The square knot, also known as the reef knot, is one of the oldest, most basic knots to tie, and can be used in many different ways. Here's the proper way to tie a square knot.", 
             u"title": u"How to Tie a Square Knot Properly"

youtube_dl/extractor/imdb.py

@@ -0,0 +1,59 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    get_element_by_attribute,
+)
+
+
+class ImdbIE(InfoExtractor):
+    IE_NAME = u'imdb'
+    IE_DESC = u'Internet Movie Database trailers'
+    _VALID_URL = r'http://www\.imdb\.com/video/imdb/vi(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.imdb.com/video/imdb/vi2524815897',
+        u'md5': u'9f34fa777ade3a6e57a054fdbcb3a068',
+        u'info_dict': {
+            u'id': u'2524815897',
+            u'ext': u'mp4',
+            u'title': u'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
+            u'description': u'md5:9061c2219254e5d14e03c25c98e96a81',
+            u'duration': 151,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url,video_id)
+        descr = get_element_by_attribute('itemprop', 'description', webpage)
+        available_formats = re.findall(
+            r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage,
+            flags=re.MULTILINE)
+        formats = []
+        for f_id, f_path in available_formats:
+            format_page = self._download_webpage(
+                compat_urlparse.urljoin(url, f_path),
+                u'Downloading info for %s format' % f_id)
+            json_data = self._search_regex(
+                r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>',
+                format_page, u'json data', flags=re.DOTALL)
+            info = json.loads(json_data)
+            format_info = info['videoPlayerObject']['video']
+            formats.append({
+                'format_id': f_id,
+                'url': format_info['url'],
+                'height': int(info['titleObject']['encoding']['selected'][:-1]),
+            })
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'description': descr,
+            'thumbnail': format_info['slate'],
+            'duration': int(info['titleObject']['title']['duration_seconds']),
+        }

youtube_dl/extractor/internetvideoarchive.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -43,9 +42,8 @@ class InternetVideoArchiveIE(InfoExtractor):
         video_id = query_dic['publishedid'][0]
         url = self._build_url(query)
 
-        flashconfiguration_xml = self._download_webpage(url, video_id,
+        flashconfiguration = self._download_xml(url, video_id,
             u'Downloading flash configuration')
-        flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8'))
         file_url = flashconfiguration.find('file').text
         file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
         # Replace some of the parameters in the query to get the best quality
@@ -53,9 +51,8 @@ class InternetVideoArchiveIE(InfoExtractor):
         file_url = re.sub(r'(?<=\?)(.+)$',
             lambda m: self._clean_query(m.group()),
             file_url)
-        info_xml = self._download_webpage(file_url, video_id,
+        info = self._download_xml(file_url, video_id,
             u'Downloading video info')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
         item = info.find('channel/item')
 
         def _bp(p):

youtube_dl/extractor/jeuxvideo.py

@@ -2,7 +2,6 @@
 
 import json
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 
@@ -22,7 +21,7 @@ class JeuxVideoIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        title = re.match(self._VALID_URL, url).group(1)
+        title = mobj.group(1)
         webpage = self._download_webpage(url, title)
         xml_link = self._html_search_regex(
             r'<param name="flashvars" value="config=(.*?)" />',
@@ -32,12 +31,9 @@ class JeuxVideoIE(InfoExtractor):
             r'http://www\.jeuxvideo\.com/config/\w+/\d+/(.*?)/\d+_player\.xml',
             xml_link, u'video ID')
 
-        xml_config = self._download_webpage(
+        config = self._download_xml(
             xml_link, title, u'Downloading XML config')
-        config = xml.etree.ElementTree.fromstring(xml_config.encode('utf-8'))
+        info_json = config.find('format.json').text
-        info_json = self._search_regex(
-            r'(?sm)<format\.json>(.*?)</format\.json>',
-            xml_config, u'JSON information')
         info = json.loads(info_json)['versions'][0]
         
         video_url = 'http://video720.jeuxvideo.com/' + info['file']

youtube_dl/extractor/justintv.py

@@ -1,7 +1,6 @@
 import json
 import os
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -94,10 +93,9 @@ class JustinTVIE(InfoExtractor):
             archive_id = m.group(1)
 
             api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id
-            chapter_info_xml = self._download_webpage(api, chapter_id,
+            doc = self._download_xml(api, chapter_id,
                                              note=u'Downloading chapter information',
                                              errnote=u'Chapter information download failed')
-            doc = xml.etree.ElementTree.fromstring(chapter_info_xml)
             for a in doc.findall('.//archive'):
                 if archive_id == a.find('./id').text:
                     break

youtube_dl/extractor/livestream.py

@@ -1,14 +1,11 @@
 import re
 import json
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse_urlparse,
     compat_urlparse,
-    get_meta_content,
     xpath_with_ns,
-    ExtractorError,
 )
 
 
@@ -82,8 +79,7 @@ class LivestreamOriginalIE(InfoExtractor):
         user = mobj.group('user')
         api_url = 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user, video_id)
 
-        api_response = self._download_webpage(api_url, video_id)
+        info = self._download_xml(api_url, video_id)
-        info = xml.etree.ElementTree.fromstring(api_response.encode('utf-8'))
         item = info.find('channel').find('item')
         ns = {'media': 'http://search.yahoo.com/mrss'}
         thumbnail_url = item.find(xpath_with_ns('media:thumbnail', ns)).attrib['url']

youtube_dl/extractor/mixcloud.py

@@ -60,7 +60,7 @@ class MixcloudIE(InfoExtractor):
             'title': info['name'],
             'url': final_song_url,
             'ext': 'mp3',
-            'description': info['description'],
+            'description': info.get('description'),
             'thumbnail': info['pictures'].get('extra_large'),
             'uploader': info['user']['name'],
             'uploader_id': info['user']['username'],

youtube_dl/extractor/mtv.py

@@ -48,7 +48,7 @@ class MTVIE(InfoExtractor):
     def _transform_rtmp_url(rtmp_video_url):
         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
         if not m:
-            raise ExtractorError(u'Cannot transform RTMP url')
+            return rtmp_video_url
         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
         return base + m.group('finalid')
 
@@ -59,7 +59,6 @@ class MTVIE(InfoExtractor):
         if '/error_country_block.swf' in metadataXml:
             raise ExtractorError(u'This video is not available from your country.', expected=True)
         mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8'))
-        renditions = mdoc.findall('.//rendition')
 
         formats = []
         for rendition in mdoc.findall('.//rendition'):
@@ -110,9 +109,8 @@ class MTVIE(InfoExtractor):
     def _get_videos_info(self, uri):
         video_id = self._id_from_uri(uri)
         data = compat_urllib_parse.urlencode({'uri': uri})
-        infoXml = self._download_webpage(self._FEED_URL +'?' + data, video_id,
+        idoc = self._download_xml(self._FEED_URL +'?' + data, video_id,
                                          u'Downloading info')
-        idoc = xml.etree.ElementTree.fromstring(infoXml.encode('utf-8'))
         return [self._get_video_info(item) for item in idoc.findall('.//item')]
 
     def _real_extract(self, url):

youtube_dl/extractor/myspass.py

@@ -1,5 +1,4 @@
 import os.path
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -33,8 +32,7 @@ class MySpassIE(InfoExtractor):
 
         # get metadata
         metadata_url = META_DATA_URL_TEMPLATE % video_id
-        metadata_text = self._download_webpage(metadata_url, video_id)
+        metadata = self._download_xml(metadata_url, video_id)
-        metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8'))
 
         # extract values from metadata
         url_flv_el = metadata.find('url_flv')

youtube_dl/extractor/naver.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -38,14 +37,12 @@ class NaverIE(InfoExtractor):
             'protocol': 'p2p',
             'inKey': key,
         })
-        info_xml = self._download_webpage(
+        info = self._download_xml(
             'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
             video_id, u'Downloading video info')
-        urls_xml = self._download_webpage(
+        urls = self._download_xml(
             'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
             video_id, u'Downloading video formats info')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
-        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
 
         formats = []
         for format_el in urls.findall('EncodingOptions/EncodingOption'):

youtube_dl/extractor/nbc.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import find_xpath_attr, compat_str
@@ -21,8 +20,8 @@ class NBCNewsIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        info_xml = self._download_webpage('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
+        all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8')).find('video')
+        info = all_info.find('video')
 
         return {'id': video_id,
                 'title': info.find('headline').text,

youtube_dl/extractor/nhl.py

@@ -1,6 +1,5 @@
 import re
 import json
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -26,9 +25,8 @@ class NHLBaseInfoExtractor(InfoExtractor):
             'path': initial_video_url.replace('.mp4', '_sd.mp4'),
         })
         path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
-        path_response = self._download_webpage(path_url, video_id,
+        path_doc = self._download_xml(path_url, video_id,
             u'Downloading final video url')
-        path_doc = xml.etree.ElementTree.fromstring(path_response)
         video_url = path_doc.find('path').text
 
         join = compat_urlparse.urljoin
@@ -72,7 +70,7 @@ class NHLIE(NHLBaseInfoExtractor):
 
 class NHLVideocenterIE(NHLBaseInfoExtractor):
     IE_NAME = u'nhl.com:videocenter'
-    IE_DESC = u'Download the first 12 videos from a videocenter category'
+    IE_DESC = u'NHL videocenter category'
     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?.*?catid=(?P<catid>[^&]+))?'
 
     @classmethod

youtube_dl/extractor/niconico.py

@@ -0,0 +1,127 @@
+# encoding: utf-8
+
+import re
+import socket
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_http_client,
+    compat_urllib_error,
+    compat_urllib_parse,
+    compat_urllib_request,
+    compat_urlparse,
+    compat_str,
+
+    ExtractorError,
+    unified_strdate,
+)
+
+
+class NiconicoIE(InfoExtractor):
+    IE_NAME = u'niconico'
+    IE_DESC = u'ニコニコ動画'
+
+    _TEST = {
+        u'url': u'http://www.nicovideo.jp/watch/sm22312215',
+        u'file': u'sm22312215.mp4',
+        u'md5': u'd1a75c0823e2f629128c43e1212760f9',
+        u'info_dict': {
+            u'title': u'Big Buck Bunny',
+            u'uploader': u'takuya0301',
+            u'uploader_id': u'2698420',
+            u'upload_date': u'20131123',
+            u'description': u'(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
+        },
+        u'params': {
+            u'username': u'ydl.niconico@gmail.com',
+            u'password': u'youtube-dl',
+        },
+    }
+
+    _VALID_URL = r'^https?://(?:www\.|secure\.)?nicovideo\.jp/watch/([a-z][a-z][0-9]+)(?:.*)$'
+    _NETRC_MACHINE = 'niconico'
+    # If True it will raise an error if no login info is provided
+    _LOGIN_REQUIRED = True
+
+    def _real_initialize(self):
+        self._login()
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        # No authentication to be performed
+        if username is None:
+            if self._LOGIN_REQUIRED:
+                raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
+            return False
+
+        # Log in
+        login_form_strs = {
+            u'mail': username,
+            u'password': password,
+        }
+        # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
+        # chokes on unicode
+        login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
+        login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
+        request = compat_urllib_request.Request(
+            u'https://secure.nicovideo.jp/secure/login', login_data)
+        login_results = self._download_webpage(
+            request, u'', note=u'Logging in', errnote=u'Unable to log in')
+        if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
+            self._downloader.report_warning(u'unable to log in: bad username or password')
+            return False
+        return True
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+
+        # Get video webpage. We are not actually interested in it, but need
+        # the cookies in order to be able to download the info webpage
+        self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
+
+        video_info = self._download_xml(
+            'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
+            note=u'Downloading video info page')
+
+        # Get flv info
+        flv_info_webpage = self._download_webpage(
+            u'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
+            video_id, u'Downloading flv info')
+        video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
+
+        # Start extracting information
+        video_title = video_info.find('.//title').text
+        video_extension = video_info.find('.//movie_type').text
+        video_format = video_extension.upper()
+        video_thumbnail = video_info.find('.//thumbnail_url').text
+        video_description = video_info.find('.//description').text
+        video_uploader_id = video_info.find('.//user_id').text
+        video_upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
+        video_view_count = video_info.find('.//view_counter').text
+        video_webpage_url = video_info.find('.//watch_url').text
+
+        # uploader
+        video_uploader = video_uploader_id
+        url = 'http://seiga.nicovideo.jp/api/user/info?id=' + video_uploader_id
+        try:
+            user_info = self._download_xml(
+                url, video_id, note=u'Downloading user information')
+            video_uploader = user_info.find('.//nickname').text
+        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+            self._downloader.report_warning(u'Unable to download user info webpage: %s' % compat_str(err))
+
+        return {
+            'id':          video_id,
+            'url':         video_real_url,
+            'title':       video_title,
+            'ext':         video_extension,
+            'format':      video_format,
+            'thumbnail':   video_thumbnail,
+            'description': video_description,
+            'uploader':    video_uploader,
+            'upload_date': video_upload_date,
+            'uploader_id': video_uploader_id,
+            'view_count':  video_view_count,
+            'webpage_url': video_webpage_url,
+        }

youtube_dl/extractor/podomatic.py

@@ -0,0 +1,49 @@
+import json
+import re
+
+from .common import InfoExtractor
+
+
+class PodomaticIE(InfoExtractor):
+    IE_NAME = 'podomatic'
+    _VALID_URL = r'^(?P<proto>https?)://(?P<channel>[^.]+)\.podomatic\.com/entry/(?P<id>[^?]+)'
+
+    _TEST = {
+        u"url": u"http://scienceteachingtips.podomatic.com/entry/2009-01-02T16_03_35-08_00",
+        u"file": u"2009-01-02T16_03_35-08_00.mp3",
+        u"md5": u"84bb855fcf3429e6bf72460e1eed782d",
+        u"info_dict": {
+            u"uploader": u"Science Teaching Tips",
+            u"uploader_id": u"scienceteachingtips",
+            u"title": u"64.  When the Moon Hits Your Eye",
+            u"duration": 446,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        channel = mobj.group('channel')
+
+        json_url = (('%s://%s.podomatic.com/entry/embed_params/%s' +
+                     '?permalink=true&rtmp=0') %
+                    (mobj.group('proto'), channel, video_id))
+        data_json = self._download_webpage(
+            json_url, video_id, note=u'Downloading video info')
+        data = json.loads(data_json)
+
+        video_url = data['downloadLink']
+        uploader = data['podcast']
+        title = data['title']
+        thumbnail = data['imageLocation']
+        duration = int(data['length'] / 1000.0)
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'uploader': uploader,
+            'uploader_id': channel,
+            'thumbnail': thumbnail,
+            'duration': duration,
+        }

youtube_dl/extractor/pornhub.py

@@ -6,7 +6,6 @@ from ..utils import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
     compat_urllib_parse,
-    unescapeHTML,
 )
 from ..aes import (
     aes_decrypt_text

youtube_dl/extractor/redtube.py

@@ -8,7 +8,9 @@ class RedTubeIE(InfoExtractor):
     _TEST = {
         u'url': u'http://www.redtube.com/66418',
         u'file': u'66418.mp4',
-        u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d',
+        # md5 varies from time to time, as in
+        # https://travis-ci.org/rg3/youtube-dl/jobs/14052463#L295
+        #u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d',
         u'info_dict': {
             u"title": u"Sucked on a toilet",
             u"age_limit": 18,

youtube_dl/extractor/rtlnow.py

@@ -62,18 +62,6 @@ class RTLnowIE(InfoExtractor):
             u'skip_download': True,
         },
     },
-    {
-        u'url': u'http://www.rtlnitronow.de/recht-ordnung/stadtpolizei-frankfurt-gerichtsvollzieher-leipzig.php?film_id=129679&player=1&season=1',
-        u'file': u'129679.flv',
-        u'info_dict': {
-            u'upload_date': u'20131016', 
-            u'title': u'Recht & Ordnung - Stadtpolizei Frankfurt/ Gerichtsvollzieher...',
-            u'description': u'Stadtpolizei Frankfurt/ Gerichtsvollzieher Leipzig',
-        },
-        u'params': {
-            u'skip_download': True,
-        },
-    },
     {
         u'url': u'http://www.n-tvnow.de/top-gear/episode-1-2013-01-01-00-00-00.php?film_id=124903&player=1&season=10',
         u'file': u'124903.flv',

youtube_dl/extractor/sina.py

@@ -1,7 +1,6 @@
 # coding: utf-8
 
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -35,12 +34,11 @@ class SinaIE(InfoExtractor):
 
     def _extract_video(self, video_id):
         data = compat_urllib_parse.urlencode({'vid': video_id})
-        url_page = self._download_webpage('http://v.iask.com/v_play.php?%s' % data,
+        url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data,
             video_id, u'Downloading video url')
         image_page = self._download_webpage(
             'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data,
             video_id, u'Downloading thumbnail info')
-        url_doc = xml.etree.ElementTree.fromstring(url_page.encode('utf-8'))
 
         return {'id': video_id,
                 'url': url_doc.find('./durl/url').text,

youtube_dl/extractor/soundcloud.py

@@ -59,6 +59,7 @@ class SoundcloudIE(InfoExtractor):
     ]
 
     _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28'
+    _IPHONE_CLIENT_ID = '376f225bf427445fc4bfb6b99b72e0bf'
 
     @classmethod
     def suitable(cls, url):
@@ -75,36 +76,79 @@ class SoundcloudIE(InfoExtractor):
     def _extract_info_dict(self, info, full_title=None, quiet=False):
         track_id = compat_str(info['id'])
         name = full_title or track_id
-        if quiet == False:
+        if quiet:
             self.report_extraction(name)
 
         thumbnail = info['artwork_url']
         if thumbnail is not None:
             thumbnail = thumbnail.replace('-large', '-t500x500')
+        ext = info.get('original_format', u'mp3')
         result = {
-            'id':       track_id,
+            'id': track_id,
-            'url':      info['stream_url'] + '?client_id=' + self._CLIENT_ID,
             'uploader': info['user']['username'],
             'upload_date': unified_strdate(info['created_at']),
-            'title':    info['title'],
+            'title': info['title'],
-            'ext':      info.get('original_format', u'mp3'),
             'description': info['description'],
             'thumbnail': thumbnail,
         }
         if info.get('downloadable', False):
-            result['url'] = 'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(track_id, self._CLIENT_ID)
+            # We can build a direct link to the song
-        if not info.get('streamable', False):
+            format_url = (
-            # We have to get the rtmp url
+                u'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(
+                    track_id, self._CLIENT_ID))
+            result['formats'] = [{
+                'format_id': 'download',
+                'ext': ext,
+                'url': format_url,
+                'vcodec': 'none',
+            }]
+        else:
+            # We have to retrieve the url
             stream_json = self._download_webpage(
-                'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._CLIENT_ID),
+                'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._IPHONE_CLIENT_ID),
                 track_id, u'Downloading track url')
-            rtmp_url = json.loads(stream_json)['rtmp_mp3_128_url']
+
-            # The url doesn't have an rtmp app, we have to extract the playpath
+            formats = []
-            url, path = rtmp_url.split('mp3:', 1)
+            format_dict = json.loads(stream_json)
-            result.update({
+            for key, stream_url in format_dict.items():
-                'url': url,
+                if key.startswith(u'http'):
-                'play_path': 'mp3:' + path,
+                    formats.append({
-            })
+                        'format_id': key,
+                        'ext': ext,
+                        'url': stream_url,
+                        'vcodec': 'none',
+                    })
+                elif key.startswith(u'rtmp'):
+                    # The url doesn't have an rtmp app, we have to extract the playpath
+                    url, path = stream_url.split('mp3:', 1)
+                    formats.append({
+                        'format_id': key,
+                        'url': url,
+                        'play_path': 'mp3:' + path,
+                        'ext': ext,
+                        'vcodec': 'none',
+                    })
+
+            if not formats:
+                # We fallback to the stream_url in the original info, this
+                # cannot be always used, sometimes it can give an HTTP 404 error
+                formats.append({
+                    'format_id': u'fallback',
+                    'url': info['stream_url'] + '?client_id=' + self._CLIENT_ID,
+                    'ext': ext,
+                    'vcodec': 'none',
+                })
+
+            def format_pref(f):
+                if f['format_id'].startswith('http'):
+                    return 2
+                if f['format_id'].startswith('rtmp'):
+                    return 1
+                return 0
+
+            formats.sort(key=format_pref)
+            result['formats'] = formats
+
         return result
 
     def _real_extract(self, url):
@@ -158,7 +202,6 @@ class SoundcloudSetIE(SoundcloudIE):
         resolv_url = self._resolv_url(url)
         info_json = self._download_webpage(resolv_url, full_title)
 
-        videos = []
         info = json.loads(info_json)
         if 'errors' in info:
             for err in info['errors']:

youtube_dl/extractor/southparkstudios.py

@@ -5,21 +5,19 @@ from .mtv import MTVIE, _media_xml_tag
 
 class SouthParkStudiosIE(MTVIE):
     IE_NAME = u'southparkstudios.com'
-    _VALID_URL = r'https?://www\.southparkstudios\.com/(clips|full-episodes)/(?P<id>.+?)(\?|#|$)'
+    _VALID_URL = r'(https?://)?(www\.)?(?P<url>southparkstudios\.com/(clips|full-episodes)/(?P<id>.+?)(\?|#|$))'
 
     _FEED_URL = 'http://www.southparkstudios.com/feeds/video-player/mrss'
 
-    _TEST = {
+    # Overwrite MTVIE properties we don't want
+    _TESTS = [{
         u'url': u'http://www.southparkstudios.com/clips/104437/bat-daded#tab=featured',
         u'file': u'a7bff6c2-ed00-11e0-aca6-0026b9414f30.mp4',
         u'info_dict': {
             u'title': u'Bat Daded',
             u'description': u'Randy disqualifies South Park by getting into a fight with Bat Dad.',
         },
-    }
+    }]
-
-    # Overwrite MTVIE properties we don't want
-    _TESTS = []
 
     def _get_thumbnail_url(self, uri, itemdoc):
         search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
@@ -31,8 +29,23 @@ class SouthParkStudiosIE(MTVIE):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
+        url = u'http://www.' + mobj.group(u'url')
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
         mgid = self._search_regex(r'swfobject.embedSWF\(".*?(mgid:.*?)"',
                                   webpage, u'mgid')
         return self._get_videos_info(mgid)
+
+class SouthparkDeIE(SouthParkStudiosIE):
+    IE_NAME = u'southpark.de'
+    _VALID_URL = r'(https?://)?(www\.)?(?P<url>southpark\.de/(clips|alle-episoden)/(?P<id>.+?)(\?|#|$))'
+    _FEED_URL = 'http://www.southpark.de/feeds/video-player/mrss/'
+
+    _TESTS = [{
+        u'url': u'http://www.southpark.de/clips/uygssh/the-government-wont-respect-my-privacy#tab=featured',
+        u'file': u'85487c96-b3b9-4e39-9127-ad88583d9bf2.mp4',
+        u'info_dict': {
+            u'title': u'The Government Won\'t Respect My Privacy',
+            u'description': u'Cartman explains the benefits of "Shitter" to Stan, Kyle and Craig.',
+        },
+    }]

youtube_dl/extractor/spankwire.py

@@ -6,7 +6,6 @@ from ..utils import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
     compat_urllib_parse,
-    unescapeHTML,
 )
 from ..aes import (
     aes_decrypt_text
@@ -36,11 +35,12 @@ class SpankwireIE(InfoExtractor):
         webpage = self._download_webpage(req, video_id)
 
         video_title = self._html_search_regex(r'<h1>([^<]+)', webpage, u'title')
-        video_uploader = self._html_search_regex(r'by:\s*<a [^>]*>(.+?)</a>', webpage, u'uploader', fatal=False)
+        video_uploader = self._html_search_regex(
-        thumbnail = self._html_search_regex(r'flashvars\.image_url = "([^"]+)', webpage, u'thumbnail', fatal=False)
+            r'by:\s*<a [^>]*>(.+?)</a>', webpage, u'uploader', fatal=False)
-        description = self._html_search_regex(r'>\s*Description:</div>\s*<[^>]*>([^<]+)', webpage, u'description', fatal=False)
+        thumbnail = self._html_search_regex(
-        if len(description) == 0:
+            r'flashvars\.image_url = "([^"]+)', webpage, u'thumbnail', fatal=False)
-            description = None
+        description = self._html_search_regex(
+            r'<div\s+id="descriptionContent">([^<]+)<', webpage, u'description', fatal=False)
 
         video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'flashvars\.quality_[0-9]{3}p = "([^"]+)', webpage)))
         if webpage.find('flashvars\.encrypted = "true"') != -1:

youtube_dl/extractor/spiegel.py

@@ -1,19 +1,26 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 
 
 class SpiegelIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$'
-    _TEST = {
+    _TESTS = [{
         u'url': u'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
         u'file': u'1259285.mp4',
         u'md5': u'2c2754212136f35fb4b19767d242f66e',
         u'info_dict': {
             u"title": u"Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv"
         }
-    }
+    },
+    {
+        u'url': u'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
+        u'file': u'1309159.mp4',
+        u'md5': u'f2cdf638d7aa47654e251e1aee360af1',
+        u'info_dict': {
+            u'title': u'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers'
+        }
+    }]
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
@@ -21,25 +28,36 @@ class SpiegelIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        video_title = self._html_search_regex(r'<div class="module-title">(.*?)</div>',
+        video_title = self._html_search_regex(
-            webpage, u'title')
+            r'<div class="module-title">(.*?)</div>', webpage, u'title')
 
         xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml'
-        xml_code = self._download_webpage(xml_url, video_id,
+        idoc = self._download_xml(
-                    note=u'Downloading XML', errnote=u'Failed to download XML')
+            xml_url, video_id,
+            note=u'Downloading XML', errnote=u'Failed to download XML')
 
-        idoc = xml.etree.ElementTree.fromstring(xml_code)
+        formats = [
-        last_type = idoc[-1]
+            {
-        filename = last_type.findall('./filename')[0].text
+                'format_id': n.tag.rpartition('type')[2],
-        duration = float(last_type.findall('./duration')[0].text)
+                'url': u'http://video2.spiegel.de/flash/' + n.find('./filename').text,
+                'width': int(n.find('./width').text),
+                'height': int(n.find('./height').text),
+                'abr': int(n.find('./audiobitrate').text),
+                'vbr': int(n.find('./videobitrate').text),
+                'vcodec': n.find('./codec').text,
+                'acodec': 'MP4A',
+            }
+            for n in list(idoc)
+            # Blacklist type 6, it's extremely LQ and not available on the same server
+            if n.tag.startswith('type') and n.tag != 'type6'
+        ]
+        formats.sort(key=lambda f: f['vbr'])
+        duration = float(idoc[0].findall('./duration')[0].text)
 
-        video_url = 'http://video2.spiegel.de/flash/' + filename
-        video_ext = filename.rpartition('.')[2]
         info = {
             'id': video_id,
-            'url': video_url,
-            'ext': video_ext,
             'title': video_title,
             'duration': duration,
+            'formats': formats,
         }
-        return [info]
+        return info

youtube_dl/extractor/streamcloud.py

@@ -0,0 +1,66 @@
+# coding: utf-8
+import re
+import time
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    compat_urllib_request,
+)
+
+
+class StreamcloudIE(InfoExtractor):
+    IE_NAME = u'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,
+        },
+        u'skip': u'Only available from the EU'
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        orig_webpage = self._download_webpage(url, video_id)
+
+        fields = re.findall(r'''(?x)<input\s+
+            type="(?:hidden|submit)"\s+
+            name="([^"]+)"\s+
+            (?:id="[^"]+"\s+)?
+            value="([^"]*)"
+            ''', orig_webpage)
+        post = compat_urllib_parse.urlencode(fields)
+
+        self.to_screen('%s: Waiting for timeout' % video_id)
+        time.sleep(12)
+        headers = {
+            b'Content-Type': b'application/x-www-form-urlencoded',
+        }
+        req = compat_urllib_request.Request(url, post, headers)
+
+        webpage = self._download_webpage(
+            req, video_id, note=u'Downloading video page ...')
+        title = self._html_search_regex(
+            r'<h1[^>]*>([^<]+)<', webpage, u'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)
+        thumbnail = self._search_regex(
+            r'image:\s*"([^"]+)"', webpage, u'thumbnail URL', fatal=False)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'duration': duration,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/sztvhu.py

@@ -15,7 +15,8 @@ class SztvHuIE(InfoExtractor):
         u'info_dict': {
             u"title": u"Cserkészek népszerűsítették a környezettudatos életmódot a Savaria téren",
             u"description": u'A zöld nap játékos ismeretterjesztő programjait a Magyar Cserkész Szövetség szervezte, akik az ország nyolc városában adják át tudásukat az érdeklődőknek. A PET...',
-        }
+        },
+        u'skip': u'Service temporarily disabled as of 2013-11-20'
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/teamcoco.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -32,8 +31,7 @@ class TeamcocoIE(InfoExtractor):
         self.report_extraction(video_id)
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
-        data_xml = self._download_webpage(data_url, video_id, 'Downloading data webpage')
+        data = self._download_xml(data_url, video_id, 'Downloading data webpage')
-        data = xml.etree.ElementTree.fromstring(data_xml.encode('utf-8'))
 
 
         qualities = ['500k', '480p', '1000k', '720p', '1080p']
@@ -60,7 +58,7 @@ class TeamcocoIE(InfoExtractor):
                 return -1
         formats.sort(key=sort_key)
         if not formats:
-            raise RegexNotFoundError(u'Unable to extract video URL')
+            raise ExtractorError(u'Unable to extract video URL')
 
         return {
             'id':          video_id,

youtube_dl/extractor/ted.py

@@ -4,7 +4,6 @@ import re
 from .subtitles import SubtitlesInfoExtractor
 
 from ..utils import (
-    compat_str,
     RegexNotFoundError,
 )
 
@@ -43,26 +42,25 @@ class TEDIE(SubtitlesInfoExtractor):
             self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name))
             return [self._playlist_videos_info(url,name,playlist_id)]
 
-    def _playlist_videos_info(self,url,name,playlist_id=0):
+
+    def _playlist_videos_info(self, url, name, playlist_id):
         '''Returns the videos of the playlist'''
-        video_RE=r'''
+
-                     <li\ id="talk_(\d+)"([.\s]*?)data-id="(?P<video_id>\d+)"
+        webpage = self._download_webpage(
-                     ([.\s]*?)data-playlist_item_id="(\d+)"
+            url, playlist_id, u'Downloading playlist webpage')
-                     ([.\s]*?)data-mediaslug="(?P<mediaSlug>.+?)"
+        matches = re.finditer(
-                     '''
+            r'<p\s+class="talk-title[^"]*"><a\s+href="(?P<talk_url>/talks/[^"]+\.html)">[^<]*</a></p>',
-        video_name_RE=r'<p\ class="talk-title"><a href="(?P<talk_url>/talks/(.+).html)">(?P<fullname>.+?)</a></p>'
+            webpage)
-        webpage=self._download_webpage(url, playlist_id, 'Downloading playlist webpage')
-        m_videos=re.finditer(video_RE,webpage,re.VERBOSE)
-        m_names=re.finditer(video_name_RE,webpage)
 
         playlist_title = self._html_search_regex(r'div class="headline">\s*?<h1>\s*?<span>(.*?)</span>',
                                                  webpage, 'playlist title')
 
-        playlist_entries = []
+        playlist_entries = [
-        for m_video, m_name in zip(m_videos,m_names):
+            self.url_result(u'http://www.ted.com' + m.group('talk_url'), 'TED')
-            talk_url='http://www.ted.com%s' % m_name.group('talk_url')
+            for m in matches
-            playlist_entries.append(self.url_result(talk_url, 'TED'))
+        ]
-        return self.playlist_result(playlist_entries, playlist_id = playlist_id, playlist_title = playlist_title)
+        return self.playlist_result(
+            playlist_entries, playlist_id=playlist_id, playlist_title=playlist_title)
 
     def _talk_info(self, url, video_id=0):
         """Return the video for the talk in the url"""
@@ -85,7 +83,7 @@ class TEDIE(SubtitlesInfoExtractor):
             'ext': 'mp4',
             'url': stream['file'],
             'format': stream['id']
-            } for stream in info['htmlStreams']]
+        } for stream in info['htmlStreams']]
 
         video_id = info['id']
 
@@ -95,7 +93,7 @@ class TEDIE(SubtitlesInfoExtractor):
             self._list_available_subtitles(video_id, webpage)
             return
 
-        info = {
+        return {
             'id': video_id,
             'title': title,
             'thumbnail': thumbnail,
@@ -104,11 +102,6 @@ class TEDIE(SubtitlesInfoExtractor):
             'formats': formats,
         }
 
-        # TODO: Remove when #980 has been merged
-        info.update(info['formats'][-1])
-
-        return info
-
     def _get_available_subtitles(self, video_id, webpage):
         try:
             options = self._search_regex(r'(?:<select name="subtitles_language_select" id="subtitles_language_select">)(.*?)(?:</select>)', webpage, 'subtitles_language_select', flags=re.DOTALL)
@@ -119,6 +112,6 @@ class TEDIE(SubtitlesInfoExtractor):
                     url = 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/srt' % (video_id, l)
                     sub_lang_list[l] = url
                 return sub_lang_list
-        except RegexNotFoundError as err:
+        except RegexNotFoundError:
             self._downloader.report_warning(u'video doesn\'t have subtitles')
         return {}

youtube_dl/extractor/toutv.py

@@ -0,0 +1,71 @@
+# coding: utf-8
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    unified_strdate,
+)
+
+
+class TouTvIE(InfoExtractor):
+    IE_NAME = u'tou.tv'
+    _VALID_URL = r'https?://www\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/(?P<episode>S[0-9]+E[0-9]+)))'
+
+    _TEST = {
+        u'url': u'http://www.tou.tv/30-vies/S04E41',
+        u'file': u'30-vies_S04E41.mp4',
+        u'info_dict': {
+            u'title': u'30 vies Saison 4 / Épisode 41',
+            u'description': u'md5:da363002db82ccbe4dafeb9cab039b09',
+            u'age_limit': 8,
+            u'uploader': u'Groupe des Nouveaux Médias',
+            u'duration': 1296,
+            u'upload_date': u'20131118',
+            u'thumbnail': u'http://static.tou.tv/medias/images/2013-11-18_19_00_00_30VIES_0341_01_L.jpeg',
+        },
+        u'params': {
+            u'skip_download': True,  # Requires rtmpdump
+        },
+        u'skip': 'Only available in Canada'
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+
+        mediaId = self._search_regex(
+            r'"idMedia":\s*"([^"]+)"', webpage, u'media ID')
+
+        streams_url = u'http://release.theplatform.com/content.select?pid=' + mediaId
+        streams_doc = self._download_xml(
+            streams_url, video_id, note=u'Downloading stream list')
+
+        video_url = next(n.text
+                         for n in streams_doc.findall('.//choice/url')
+                         if u'//ad.doubleclick' not in n.text)
+        if video_url.endswith('/Unavailable.flv'):
+            raise ExtractorError(
+                u'Access to this video is blocked from outside of Canada',
+                expected=True)
+
+        duration_str = self._html_search_meta(
+            'video:duration', webpage, u'duration')
+        duration = int(duration_str) if duration_str else None
+        upload_date_str = self._html_search_meta(
+            'video:release_date', webpage, u'upload date')
+        upload_date = unified_strdate(upload_date_str) if upload_date_str else None
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'url': video_url,
+            'description': self._og_search_description(webpage),
+            'uploader': self._dc_search_uploader(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'age_limit': self._media_rating_search(webpage),
+            'duration': duration,
+            'upload_date': upload_date,
+            'ext': 'mp4',
+        }

youtube_dl/extractor/trilulilu.py

@@ -1,6 +1,5 @@
 import json
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 
@@ -36,12 +35,10 @@ class TriluliluIE(InfoExtractor):
 
         format_url = (u'http://fs%(server)s.trilulilu.ro/%(hash)s/'
                       u'video-formats2' % log)
-        format_str = self._download_webpage(
+        format_doc = self._download_xml(
             format_url, video_id,
             note=u'Downloading formats',
             errnote=u'Error while downloading formats')
-
-        format_doc = xml.etree.ElementTree.fromstring(format_str)
  
         video_url_template = (
             u'http://fs%(server)s.trilulilu.ro/stream.php?type=video'

youtube_dl/extractor/tube8.py

@@ -5,8 +5,6 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
-    compat_urllib_parse,
-    unescapeHTML,
 )
 from ..aes import (
     aes_decrypt_text

youtube_dl/extractor/tvp.py

@@ -13,9 +13,10 @@ class TvpIE(InfoExtractor):
         u'md5': u'148408967a6a468953c0a75cbdaf0d7a',
         u'file': u'12878238.wmv',
         u'info_dict': {
-            u'title': u'31.10.2013',
+            u'title': u'31.10.2013 - Odcinek 2',
-            u'description': u'31.10.2013',
+            u'description': u'31.10.2013 - Odcinek 2',
         },
+        u'skip': u'Download has to use same server IP as extraction. Therefore, a good (load-balancing) DNS resolver will make the download fail.'
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/vevo.py

@@ -78,12 +78,13 @@ class VevoIE(InfoExtractor):
                 continue
 
             format_url = self._SMIL_BASE_URL + m.group('path')
-            format_note = ('%(vcodec)s@%(vbr)4sk, %(acodec)s@%(abr)3sk' %
-                           m.groupdict())
             formats.append({
                 'url': format_url,
                 'format_id': u'SMIL_' + m.group('cbr'),
-                'format_note': format_note,
+                'vcodec': m.group('vcodec'),
+                'acodec': m.group('acodec'),
+                'vbr': int(m.group('vbr')),
+                'abr': int(m.group('abr')),
                 'ext': m.group('ext'),
                 'width': int(m.group('width')),
                 'height': int(m.group('height')),

youtube_dl/extractor/videofyme.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -27,9 +26,8 @@ class VideofyMeIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        config_xml = self._download_webpage('http://sunshine.videofy.me/?videoId=%s' % video_id,
+        config = self._download_xml('http://sunshine.videofy.me/?videoId=%s' % video_id,
                                             video_id)
-        config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
         video = config.find('video')
         sources = video.find('sources')
         url_node = next(node for node in [find_xpath_attr(sources, 'source', 'id', 'HQ %s' % key) 

youtube_dl/extractor/videopremium.py

@@ -5,7 +5,7 @@ from .common import InfoExtractor
 
 
 class VideoPremiumIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?videopremium\.tv/(?P<id>\w+)(?:/.*)?'
+    _VALID_URL = r'(?:https?://)?(?:www\.)?videopremium\.(?:tv|me)/(?P<id>\w+)(?:/.*)?'
     _TEST = {
         u'url': u'http://videopremium.tv/4w7oadjsf156',
         u'file': u'4w7oadjsf156.f4v',
@@ -24,12 +24,16 @@ class VideoPremiumIE(InfoExtractor):
         webpage_url = 'http://videopremium.tv/' + video_id
         webpage = self._download_webpage(webpage_url, video_id)
 
-        self.report_extraction(video_id)
+        if re.match(r"^<html><head><script[^>]*>window.location\s*=", webpage):
+            # Download again, we need a cookie
+            webpage = self._download_webpage(
+                webpage_url, video_id,
+                note=u'Downloading webpage again (with cookie)')
 
-        video_title = self._html_search_regex(r'<h2(?:.*?)>\s*(.+?)\s*<',
+        video_title = self._html_search_regex(
-            webpage, u'video title')
+            r'<h2(?:.*?)>\s*(.+?)\s*<', webpage, u'video title')
 
-        return [{
+        return {
             'id':          video_id,
             'url':         "rtmp://e%d.md.iplay.md/play" % random.randint(1, 16),
             'play_path':   "mp4:%s.f4v" % video_id,
@@ -37,4 +41,4 @@ class VideoPremiumIE(InfoExtractor):
             'player_url':  "http://videopremium.tv/uplayer/uppod.swf",
             'ext':         'f4v',
             'title':       video_title,
-        }]
+        }

youtube_dl/extractor/viki.py

@@ -0,0 +1,101 @@
+import re
+
+from ..utils import (
+    ExtractorError,
+    unescapeHTML,
+    unified_strdate,
+)
+from .subtitles import SubtitlesInfoExtractor
+
+
+class VikiIE(SubtitlesInfoExtractor):
+    IE_NAME = u'viki'
+
+    _VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
+    _TEST = {
+        u'url': u'http://www.viki.com/videos/1023585v-heirs-episode-14',
+        u'file': u'1023585v.mp4',
+        u'md5': u'a21454021c2646f5433514177e2caa5f',
+        u'info_dict': {
+            u'title': u'Heirs Episode 14',
+            u'uploader': u'SBS',
+            u'description': u'md5:c4b17b9626dd4b143dcc4d855ba3474e',
+            u'upload_date': u'20131121',
+            u'age_limit': 13,
+        },
+        u'skip': u'Blocked in the US',
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+
+        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_m = re.search(
+            r'<strong>Broadcast Network: </strong>\s*([^<]*)<', webpage)
+        if uploader_m is None:
+            uploader = None
+        else:
+            uploader = uploader_m.group(1).strip()
+
+        rating_str = self._html_search_regex(
+            r'<strong>Rating: </strong>\s*([^<]*)<', webpage,
+            u'rating information', default='').strip()
+        RATINGS = {
+            'G': 0,
+            'PG': 10,
+            'PG-13': 13,
+            'R': 16,
+            'NC': 18,
+        }
+        age_limit = RATINGS.get(rating_str)
+
+        info_url = 'http://www.viki.com/player5_fragment/%s?action=show&controller=videos' % video_id
+        info_webpage = self._download_webpage(
+            info_url, video_id, note=u'Downloading info page')
+        if re.match(r'\s*<div\s+class="video-error', info_webpage):
+            raise ExtractorError(
+                u'Video %s is blocked from your location.' % video_id,
+                expected=True)
+        video_url = self._html_search_regex(
+            r'<source[^>]+src="([^"]+)"', info_webpage, u'video URL')
+
+        upload_date_str = self._html_search_regex(
+            r'"created_at":"([^"]+)"', info_webpage, u'upload date')
+        upload_date = (
+            unified_strdate(upload_date_str)
+            if upload_date_str is not None
+            else None
+        )
+
+        # subtitles
+        video_subtitles = self.extract_subtitles(video_id, info_webpage)
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id, info_webpage)
+            return
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'description': description,
+            'thumbnail': thumbnail,
+            'age_limit': age_limit,
+            'uploader': uploader,
+            'subtitles': video_subtitles,
+            'upload_date': upload_date,
+        }
+
+    def _get_available_subtitles(self, video_id, info_webpage):
+        res = {}
+        for sturl_html in re.findall(r'<track src="([^"]+)"/>', info_webpage):
+            sturl = unescapeHTML(sturl_html)
+            m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
+            if not m:
+                continue
+            res[m.group('lang')] = sturl
+        return res

youtube_dl/extractor/vimeo.py

@@ -151,7 +151,7 @@ class VimeoIE(InfoExtractor):
                 config = json.loads(config_json)
             except RegexNotFoundError:
                 # For pro videos or player.vimeo.com urls
-                config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'],
+                config = self._search_regex([r' = {config:({.+?}),assets:', r'(?:c|b)=({.+?});'],
                     webpage, u'info section', flags=re.DOTALL)
                 config = json.loads(config)
         except Exception as e:
@@ -249,25 +249,46 @@ class VimeoChannelIE(InfoExtractor):
     IE_NAME = u'vimeo:channel'
     _VALID_URL = r'(?:https?://)?vimeo.\com/channels/(?P<id>[^/]+)'
     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
+    _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 
-    def _real_extract(self, url):
+    def _extract_videos(self, list_id, base_url):
-        mobj = re.match(self._VALID_URL, url)
-        channel_id =  mobj.group('id')
         video_ids = []
-
         for pagenum in itertools.count(1):
-            webpage = self._download_webpage('http://vimeo.com/channels/%s/videos/page:%d' % (channel_id, pagenum),
+            webpage = self._download_webpage(
-                                             channel_id, u'Downloading page %s' % pagenum)
+                '%s/videos/page:%d/' % (base_url, pagenum),list_id,
+                u'Downloading page %s' % pagenum)
             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
                 break
 
         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
                    for video_id in video_ids]
-        channel_title = self._html_search_regex(r'<a href="/channels/%s">(.*?)</a>' % channel_id,
+        list_title = self._html_search_regex(self._TITLE_RE, webpage,
-                                                webpage, u'channel title')
+            u'list title')
         return {'_type': 'playlist',
-                'id': channel_id,
+                'id': list_id,
-                'title': channel_title,
+                'title': list_title,
                 'entries': entries,
                 }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        channel_id =  mobj.group('id')
+        return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
+
+
+class VimeoUserIE(VimeoChannelIE):
+    IE_NAME = u'vimeo:user'
+    _VALID_URL = r'(?:https?://)?vimeo.\com/(?P<name>[^/]+)'
+    _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
+
+    @classmethod
+    def suitable(cls, url):
+        if VimeoChannelIE.suitable(url) or VimeoIE.suitable(url):
+            return False
+        return super(VimeoUserIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        return self._extract_videos(name, 'http://vimeo.com/%s' % name)

youtube_dl/extractor/xtube.py

@@ -5,7 +5,6 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
-    compat_urllib_parse,
 )
 
 class XTubeIE(InfoExtractor):

youtube_dl/extractor/yahoo.py

@@ -17,27 +17,21 @@ class YahooIE(InfoExtractor):
     _TESTS = [
         {
             u'url': u'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
-            u'file': u'214727115.flv',
+            u'file': u'214727115.mp4',
+            u'md5': u'4962b075c08be8690a922ee026d05e69',
             u'info_dict': {
                 u'title': u'Julian Smith & Travis Legg Watch Julian Smith',
                 u'description': u'Julian and Travis watch Julian Smith',
             },
-            u'params': {
-                # Requires rtmpdump
-                u'skip_download': True,
-            },
         },
         {
             u'url': u'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
-            u'file': u'103000935.flv',
+            u'file': u'103000935.mp4',
+            u'md5': u'd6e6fc6e1313c608f316ddad7b82b306',
             u'info_dict': {
                 u'title': u'Codefellas - The Cougar Lies with Spanish Moss',
                 u'description': u'Agent Topple\'s mustache does its dirty work, and Nicole brokers a deal for peace. But why is the NSA collecting millions of Instagram brunch photos? And if your waffles have nothing to hide, what are they so worried about?',
             },
-            u'params': {
-                # Requires rtmpdump
-                u'skip_download': True,
-            },
         },
     ]
 
@@ -46,15 +40,19 @@ class YahooIE(InfoExtractor):
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
 
-        items_json = self._search_regex(r'YVIDEO_INIT_ITEMS = ({.*?});$',
+        items_json = self._search_regex(r'mediaItems: ({.*?})$',
             webpage, u'items', flags=re.MULTILINE)
         items = json.loads(items_json)
         info = items['mediaItems']['query']['results']['mediaObj'][0]
         # The 'meta' field is not always in the video webpage, we request it
         # from another page
         long_id = info['id']
+        return self._get_info(info['id'], video_id)
+
+    def _get_info(self, long_id, video_id):
         query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"'
-                 ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2"' % long_id)
+                 ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="US"'
+                 ' AND protocol="http"' % long_id)
         data = compat_urllib_parse.urlencode({
             'q': query,
             'env': 'prod',
@@ -91,17 +89,39 @@ class YahooIE(InfoExtractor):
             formats.append(format_info)
         formats = sorted(formats, key=lambda f:(f['height'], f['width']))
 
-        info = {
+        return {
             'id': video_id,
             'title': meta['title'],
             'formats': formats,
             'description': clean_html(meta['description']),
             'thumbnail': meta['thumbnail'],
         }
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
 
-        return info
+
+class YahooNewsIE(YahooIE):
+    IE_NAME = 'yahoo:news'
+    _VALID_URL = r'http://news\.yahoo\.com/video/.*?-(?P<id>\d*?)\.html'
+
+    _TEST = {
+        u'url': u'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
+        u'md5': u'67010fdf3a08d290e060a4dd96baa07b',
+        u'info_dict': {
+            u'id': u'104538833',
+            u'ext': u'mp4',
+            u'title': u'China Moses Is Crazy About the Blues',
+            u'description': u'md5:9900ab8cd5808175c7b3fe55b979bed0',
+        },
+    }
+
+    # Overwrite YahooIE properties we don't want
+    _TESTS = []
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        long_id = self._search_regex(r'contentId: \'(.+?)\',', webpage, u'long id')
+        return self._get_info(long_id, video_id)
 
 
 class YahooSearchIE(SearchInfoExtractor):

youtube_dl/extractor/youtube.py

@@ -11,7 +11,6 @@ import socket
 import string
 import struct
 import traceback
-import xml.etree.ElementTree
 import zlib
 
 from .common import InfoExtractor, SearchInfoExtractor
@@ -29,6 +28,7 @@ from ..utils import (
     clean_html,
     get_cachedir,
     get_element_by_id,
+    get_element_by_attribute,
     ExtractorError,
     unescapeHTML,
     unified_strdate,
@@ -139,10 +139,10 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
 
 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
     IE_DESC = u'YouTube.com'
-    _VALID_URL = r"""^
+    _VALID_URL = r"""(?x)^
                      (
-                         (?:https?://)?                                       # http(s):// (optional)
+                         (?:https?://|//)?                                    # http(s):// or protocol-independent URL (optional)
-                         (?:(?:(?:(?:\w+\.)?youtube(?:-nocookie)?\.com/|
+                         (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
                             tube\.majestyc\.net/|
                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains
                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
@@ -248,21 +248,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '248': 'webm',
     }
     _video_dimensions = {
-        '5': '240x400',
+        '5': '400x240',
         '6': '???',
         '13': '???',
-        '17': '144x176',
+        '17': '176x144',
-        '18': '360x640',
+        '18': '640x360',
-        '22': '720x1280',
+        '22': '1280x720',
-        '34': '360x640',
+        '34': '640x360',
-        '35': '480x854',
+        '35': '854x480',
-        '36': '240x320',
+        '36': '320x240',
-        '37': '1080x1920',
+        '37': '1920x1080',
-        '38': '3072x4096',
+        '38': '4096x3072',
-        '43': '360x640',
+        '43': '640x360',
-        '44': '480x854',
+        '44': '854x480',
-        '45': '720x1280',
+        '45': '1280x720',
-        '46': '1080x1920',
+        '46': '1920x1080',
         '82': '360p',
         '83': '480p',
         '84': '720p',
@@ -363,6 +363,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 u"uploader_id": u"justintimberlakeVEVO"
             }
         },
+        {
+            u"url":  u"//www.YouTube.com/watch?v=yZIXLfi8CZQ",
+            u"file":  u"yZIXLfi8CZQ.mp4",
+            u"note": u"Embed-only video (#1746)",
+            u"info_dict": {
+                u"upload_date": u"20120608",
+                u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012",
+                u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7",
+                u"uploader": u"SET India",
+                u"uploader_id": u"setindia"
+            }
+        },
     ]
 
 
@@ -370,7 +382,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
     def suitable(cls, url):
         """Receives a URL and returns True if suitable for this IE."""
         if YoutubePlaylistIE.suitable(url): return False
-        return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
+        return re.match(cls._VALID_URL, url) is not None
 
     def __init__(self, *args, **kwargs):
         super(YoutubeIE, self).__init__(*args, **kwargs)
@@ -1132,8 +1144,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'asrs': 1,
             })
             list_url = caption_url + '&' + list_params
-            list_page = self._download_webpage(list_url, video_id)
+            caption_list = self._download_xml(list_url, video_id)
-            caption_list = xml.etree.ElementTree.fromstring(list_page.encode('utf-8'))
             original_lang_node = caption_list.find('track')
             if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' :
                 self._downloader.report_warning(u'Video doesn\'t have automatic captions')
@@ -1272,7 +1283,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             # 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': 'embedded',
+                                                  'el': 'player_embedded',
                                                   'gl': 'US',
                                                   'hl': 'en',
                                                   'eurl': 'https://youtube.googleapis.com/v/' + video_id,
@@ -1301,6 +1312,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             else:
                 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
 
+        if 'view_count' in video_info:
+            view_count = int(video_info['view_count'][0])
+        else:
+            view_count = None
+
         # Check for "rental" videos
         if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
             raise ExtractorError(u'"rental" videos not supported')
@@ -1489,10 +1505,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'age_limit':    18 if age_gate else 0,
                 'annotations':  video_annotations,
                 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id,
+                'view_count': view_count,
             })
         return results
 
-class YoutubePlaylistIE(InfoExtractor):
+class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
     IE_DESC = u'YouTube.com playlists'
     _VALID_URL = r"""(?:
                         (?:https?://)?
@@ -1508,8 +1525,9 @@ class YoutubePlaylistIE(InfoExtractor):
                      |
                         ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
                      )"""
-    _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
+    _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s'
-    _MAX_RESULTS = 50
+    _MORE_PAGES_INDICATOR = r'data-link-type="next"'
+    _VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
     IE_NAME = u'youtube:playlist'
 
     @classmethod
@@ -1517,6 +1535,27 @@ class YoutubePlaylistIE(InfoExtractor):
         """Receives a URL and returns True if suitable for this IE."""
         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
 
+    def _real_initialize(self):
+        self._login()
+
+    def _ids_to_results(self, ids):
+        return [self.url_result(vid_id, 'Youtube', video_id=vid_id)
+                       for vid_id in ids]
+
+    def _extract_mix(self, playlist_id):
+        # The mixes are generated from a a single video
+        # the id of the playlist is just 'RD' + video_id
+        url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[2:], playlist_id)
+        webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix')
+        title_span = (get_element_by_attribute('class', 'title long-title', webpage) or
+            get_element_by_attribute('class', 'title ', webpage))
+        title = clean_html(title_span)
+        video_re = r'data-index="\d+".*?href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s' % re.escape(playlist_id)
+        ids = orderedSet(re.findall(video_re, webpage))
+        url_results = self._ids_to_results(ids)
+
+        return self.playlist_result(url_results, playlist_id, title)
+
     def _real_extract(self, url):
         # Extract playlist id
         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
@@ -1530,45 +1569,33 @@ class YoutubePlaylistIE(InfoExtractor):
             video_id = query_dict['v'][0]
             if self._downloader.params.get('noplaylist'):
                 self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id)
-                return self.url_result('https://www.youtube.com/watch?v=' + video_id, 'Youtube')
+                return self.url_result(video_id, 'Youtube', video_id=video_id)
             else:
                 self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id))
 
-        # Download playlist videos from API
+        if len(playlist_id) == 13:  # 'RD' + 11 characters for the video id
-        videos = []
+            # Mixes require a custom extraction process
+            return self._extract_mix(playlist_id)
+
+        # Extract the video ids from the playlist pages
+        ids = []
 
         for page_num in itertools.count(1):
-            start_index = self._MAX_RESULTS * (page_num - 1) + 1
+            url = self._TEMPLATE_URL % (playlist_id, page_num)
-            if start_index >= 1000:
-                self._downloader.report_warning(u'Max number of results reached')
-                break
-            url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index)
             page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
+            matches = re.finditer(self._VIDEO_RE, page)
+            # We remove the duplicates and the link with index 0
+            # (it's not the first video of the playlist)
+            new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
+            ids.extend(new_ids)
 
-            try:
+            if re.search(self._MORE_PAGES_INDICATOR, page) is None:
-                response = json.loads(page)
-            except ValueError as err:
-                raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
-
-            if 'feed' not in response:
-                raise ExtractorError(u'Got a malformed response from YouTube API')
-            playlist_title = response['feed']['title']['$t']
-            if 'entry' not in response['feed']:
-                # Number of videos is a multiple of self._MAX_RESULTS
                 break
 
-            for entry in response['feed']['entry']:
+        playlist_title = self._og_search_title(page)
-                index = entry['yt$position']['$t']
-                if 'media$group' in entry and 'yt$videoid' in entry['media$group']:
-                    videos.append((
-                        index,
-                        'https://www.youtube.com/watch?v=' + entry['media$group']['yt$videoid']['$t']
-                    ))
 
-        videos = [v[1] for v in sorted(videos)]
+        url_results = self._ids_to_results(ids)
-
+        return self.playlist_result(url_results, playlist_id, playlist_title)
-        url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
-        return [self.playlist_result(url_results, playlist_id, playlist_title)]
 
 
 class YoutubeChannelIE(InfoExtractor):
@@ -1594,26 +1621,37 @@ class YoutubeChannelIE(InfoExtractor):
         # Download channel page
         channel_id = mobj.group(1)
         video_ids = []
+        url = 'https://www.youtube.com/channel/%s/videos' % channel_id
+        channel_page = self._download_webpage(url, channel_id)
+        if re.search(r'channel-header-autogenerated-label', channel_page) is not None:
+            autogenerated = True
+        else:
+            autogenerated = False
 
-        # Download all channel pages using the json-based channel_ajax query
+        if autogenerated:
-        for pagenum in itertools.count(1):
+            # The videos are contained in a single page
-            url = self._MORE_PAGES_URL % (pagenum, channel_id)
+            # the ajax pages can't be used, they are empty
-            page = self._download_webpage(url, channel_id,
+            video_ids = self.extract_videos_from_page(channel_page)
-                                          u'Downloading page #%s' % pagenum)
+        else:
-
+            # Download all channel pages using the json-based channel_ajax query
-            page = json.loads(page)
+            for pagenum in itertools.count(1):
-
+                url = self._MORE_PAGES_URL % (pagenum, channel_id)
-            ids_in_page = self.extract_videos_from_page(page['content_html'])
+                page = self._download_webpage(url, channel_id,
-            video_ids.extend(ids_in_page)
+                                              u'Downloading page #%s' % pagenum)
-
+    
-            if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
+                page = json.loads(page)
-                break
+    
+                ids_in_page = self.extract_videos_from_page(page['content_html'])
+                video_ids.extend(ids_in_page)
+    
+                if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
+                    break
 
         self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
 
-        urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
+        url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id)
-        url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
+                       for video_id in video_ids]
-        return [self.playlist_result(url_entries, channel_id)]
+        return self.playlist_result(url_entries, channel_id)
 
 
 class YoutubeUserIE(InfoExtractor):
@@ -1677,9 +1715,11 @@ class YoutubeUserIE(InfoExtractor):
             if len(ids_in_page) < self._GDATA_PAGE_SIZE:
                 break
 
-        urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
+        url_results = [
-        url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
+            self.url_result(video_id, 'Youtube', video_id=video_id)
-        return [self.playlist_result(url_results, playlist_title = username)]
+            for video_id in video_ids]
+        return self.playlist_result(url_results, playlist_title=username)
+
 
 class YoutubeSearchIE(SearchInfoExtractor):
     IE_DESC = u'YouTube.com searches'
@@ -1720,7 +1760,8 @@ class YoutubeSearchIE(SearchInfoExtractor):
 
         if len(video_ids) > n:
             video_ids = video_ids[:n]
-        videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
+        videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
+                  for video_id in video_ids]
         return self.playlist_result(videos, query)
 
 class YoutubeSearchDateIE(YoutubeSearchIE):
@@ -1750,7 +1791,6 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
     Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
     """
     _LOGIN_REQUIRED = True
-    _PAGING_STEP = 30
     # use action_load_personal_feed instead of action_load_system_feed
     _PERSONAL_FEED = False
 
@@ -1770,9 +1810,8 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
 
     def _real_extract(self, url):
         feed_entries = []
-        # The step argument is available only in 2.7 or higher
+        paging = 0
-        for i in itertools.count(0):
+        for i in itertools.count(1):
-            paging = i*self._PAGING_STEP
             info = self._download_webpage(self._FEED_TEMPLATE % paging,
                                           u'%s feed' % self._FEED_NAME,
                                           u'Downloading page %s' % i)
@@ -1780,9 +1819,12 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
             feed_html = info['feed_html']
             m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
             ids = orderedSet(m.group(1) for m in m_ids)
-            feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
+            feed_entries.extend(
+                self.url_result(video_id, 'Youtube', video_id=video_id)
+                for video_id in ids)
             if info['paging'] is None:
                 break
+            paging = info['paging']
         return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
 
 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
@@ -1802,9 +1844,15 @@ class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
     _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
     _FEED_NAME = 'watch_later'
     _PLAYLIST_TITLE = u'Youtube Watch Later'
-    _PAGING_STEP = 100
     _PERSONAL_FEED = True
 
+class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
+    IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)'
+    _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory'
+    _FEED_NAME = 'history'
+    _PERSONAL_FEED = True
+    _PLAYLIST_TITLE = u'Youtube Watch History'
+
 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
     IE_NAME = u'youtube:favorites'
     IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'

youtube_dl/extractor/zdf.py

@@ -1,75 +1,125 @@
+# coding: utf-8
+
+import operator
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
+    unified_strdate,
-    ExtractorError,
 )
 
 
 class ZDFIE(InfoExtractor):
-    _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek(?P<hash>#)?\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
+    _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P<hash>#)?/(.*beitrag/(?:video/)?)(?P<video_id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
-    _MEDIA_STREAM = r'<a href="(?P<video_url>.+(?P<media_type>.streaming).+/zdf/(?P<quality>[^\/]+)/[^"]*)".+class="play".+>'
+
+    _TEST = {
+        u"url": u"http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt",
+        u"file": u"2037704.webm",
+        u"info_dict": {
+            u"upload_date": u"20131127",
+            u"description": u"Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial \"Ende des Machtpokers - Große Koalition für Deutschland\".",
+            u"uploader": u"spezial",
+            u"title": u"ZDFspezial - Ende des Machtpokers"
+        },
+        u"skip": u"Videos on ZDF.de are depublicised in short order",
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
         video_id = mobj.group('video_id')
 
-        if mobj.group('hash'):
+        xml_url = u'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
-            url = url.replace(u'#', u'', 1)
+        doc = self._download_xml(
+            xml_url, video_id,
+            note=u'Downloading video info',
+            errnote=u'Failed to download video info')
 
-        html = self._download_webpage(url, video_id)
+        title = doc.find('.//information/title').text
-        streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
+        description = doc.find('.//information/detail').text
-        if streams is None:
+        uploader_node = doc.find('.//details/originChannelTitle')
-            raise ExtractorError(u'No media url found.')
+        uploader = None if uploader_node is None else uploader_node.text
+        duration_str = doc.find('.//details/length').text
+        duration_m = re.match(r'''(?x)^
+            (?P<hours>[0-9]{2})
+            :(?P<minutes>[0-9]{2})
+            :(?P<seconds>[0-9]{2})
+            (?:\.(?P<ms>[0-9]+)?)
+            ''', duration_str)
+        duration = (
+            (
+                (int(duration_m.group('hours')) * 60 * 60) +
+                (int(duration_m.group('minutes')) * 60) +
+                int(duration_m.group('seconds'))
+            )
+            if duration_m
+            else None
+        )
+        upload_date = unified_strdate(doc.find('.//details/airtime').text)
 
-        # s['media_type'] == 'wstreaming' -> use 'Windows Media Player' and mms url
+        def xml_to_format(fnode):
-        # s['media_type'] == 'hstreaming' -> use 'Quicktime' and rtsp url
+            video_url = fnode.find('url').text
-        # choose first/default media type and highest quality for now
+            is_available = u'http://www.metafilegenerator' not in video_url
-        def stream_pref(s):
+
-            TYPE_ORDER = ['ostreaming', 'hstreaming', 'wstreaming']
+            format_id = fnode.attrib['basetype']
+            format_m = re.match(r'''(?x)
+                (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
+                (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
+            ''', format_id)
+
+            ext = format_m.group('container')
+            is_supported = ext != 'f4f'
+
+            PROTO_ORDER = ['http', 'rtmp', 'rtsp']
             try:
-                type_pref = TYPE_ORDER.index(s['media_type'])
+                proto_pref = -PROTO_ORDER.index(format_m.group('proto'))
             except ValueError:
-                type_pref = 999
+                proto_pref = 999
 
-            QUALITY_ORDER = ['veryhigh', '300']
+            quality = fnode.find('./quality').text
+            QUALITY_ORDER = ['veryhigh', '300', 'high', 'med', 'low']
             try:
-                quality_pref = QUALITY_ORDER.index(s['quality'])
+                quality_pref = -QUALITY_ORDER.index(quality)
             except ValueError:
                 quality_pref = 999
 
-            return (type_pref, quality_pref)
+            abr = int(fnode.find('./audioBitrate').text) // 1000
+            vbr = int(fnode.find('./videoBitrate').text) // 1000
+            pref = (is_available, is_supported,
+                    proto_pref, quality_pref, vbr, abr)
 
-        sorted_streams = sorted(streams, key=stream_pref)
+            format_note = u''
-        if not sorted_streams:
+            if not is_supported:
-            raise ExtractorError(u'No stream found.')
+                format_note += u'(unsupported)'
-        stream = sorted_streams[0]
+            if not format_note:
+                format_note = None
 
-        media_link = self._download_webpage(
+            return {
-            stream['video_url'],
+                'format_id': format_id + u'-' + quality,
-            video_id,
+                'url': video_url,
-            u'Get stream URL')
+                'ext': ext,
+                'acodec': format_m.group('acodec'),
+                'vcodec': format_m.group('vcodec'),
+                'abr': abr,
+                'vbr': vbr,
+                'width': int(fnode.find('./width').text),
+                'height': int(fnode.find('./height').text),
+                'filesize': int(fnode.find('./filesize').text),
+                'format_note': format_note,
+                '_pref': pref,
+                '_available': is_available,
+            }
 
-        MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"'
+        format_nodes = doc.findall('.//formitaeten/formitaet')
-        RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)'
+        formats = sorted(filter(lambda f: f['_available'],
-
+                                map(xml_to_format, format_nodes)),
-        mobj = re.search(self._MEDIA_STREAM, media_link)
+                         key=operator.itemgetter('_pref'))
-        if mobj is None:
-            mobj = re.search(RTSP_STREAM, media_link)
-            if mobj is None:
-                raise ExtractorError(u'Cannot extract mms:// or rtsp:// URL')
-        video_url = mobj.group('video_url')
-
-        title = self._html_search_regex(
-            r'<h1(?: class="beitragHeadline")?>(.*?)</h1>',
-            html, u'title')
 
         return {
             'id': video_id,
-            'url': video_url,
             'title': title,
-            'ext': determine_ext(video_url)
+            'formats': formats,
+            'description': description,
+            'uploader': uploader,
+            'duration': duration,
+            'upload_date': upload_date,
         }

youtube_dl/update.py

@@ -2,11 +2,15 @@ import io
 import json
 import traceback
 import hashlib
+import os
 import subprocess
 import sys
 from zipimport import zipimporter
 
-from .utils import *
+from .utils import (
+    compat_str,
+    compat_urllib_request,
+)
 from .version import __version__
 
 def rsa_verify(message, signature, key):
@@ -37,6 +41,7 @@ def rsa_verify(message, signature, key):
     if signature != sha256(message).digest(): return False
     return True
 
+
 def update_self(to_screen, verbose):
     """Update the program file with the latest version from the repository"""
 
@@ -78,6 +83,13 @@ def update_self(to_screen, verbose):
         return
 
     version_id = versions_info['latest']
+
+    def version_tuple(version_str):
+        return tuple(map(int, version_str.split('.')))
+    if version_tuple(__version__) >= version_tuple(version_id):
+        to_screen(u'youtube-dl is up to date (%s)' % __version__)
+        return
+
     to_screen(u'Updating to version ' + version_id + '...')
     version = versions_info['versions'][version_id]
 
@@ -105,7 +117,7 @@ def update_self(to_screen, verbose):
             urlh = compat_urllib_request.urlopen(version['exe'][0])
             newcontent = urlh.read()
             urlh.close()
-        except (IOError, OSError) as err:
+        except (IOError, OSError):
             if verbose: to_screen(compat_str(traceback.format_exc()))
             to_screen(u'ERROR: unable to download latest version')
             return
@@ -118,7 +130,7 @@ def update_self(to_screen, verbose):
         try:
             with open(exe + '.new', 'wb') as outf:
                 outf.write(newcontent)
-        except (IOError, OSError) as err:
+        except (IOError, OSError):
             if verbose: to_screen(compat_str(traceback.format_exc()))
             to_screen(u'ERROR: unable to write the new version')
             return
@@ -137,7 +149,7 @@ start /b "" cmd /c del "%%~f0"&exit /b"
 
             subprocess.Popen([bat])  # Continues to run in the background
             return  # Do not show premature success messages
-        except (IOError, OSError) as err:
+        except (IOError, OSError):
             if verbose: to_screen(compat_str(traceback.format_exc()))
             to_screen(u'ERROR: unable to overwrite current version')
             return
@@ -148,7 +160,7 @@ start /b "" cmd /c del "%%~f0"&exit /b"
             urlh = compat_urllib_request.urlopen(version['bin'][0])
             newcontent = urlh.read()
             urlh.close()
-        except (IOError, OSError) as err:
+        except (IOError, OSError):
             if verbose: to_screen(compat_str(traceback.format_exc()))
             to_screen(u'ERROR: unable to download latest version')
             return
@@ -161,7 +173,7 @@ start /b "" cmd /c del "%%~f0"&exit /b"
         try:
             with open(filename, 'wb') as outf:
                 outf.write(newcontent)
-        except (IOError, OSError) as err:
+        except (IOError, OSError):
             if verbose: to_screen(compat_str(traceback.format_exc()))
             to_screen(u'ERROR: unable to overwrite current version')
             return

youtube_dl/utils.py

@@ -8,13 +8,16 @@ import gzip
 import io
 import json
 import locale
+import math
 import os
 import pipes
 import platform
 import re
+import ssl
 import socket
 import sys
 import traceback
+import xml.etree.ElementTree
 import zlib
 
 try:
@@ -535,17 +538,34 @@ def formatSeconds(secs):
     else:
         return '%d' % secs
 
-def make_HTTPS_handler(opts):
+def make_HTTPS_handler(opts_no_check_certificate):
-    if sys.version_info < (3,2):
+    if sys.version_info < (3, 2):
-        # Python's 2.x handler is very simplistic
+        import httplib
-        return compat_urllib_request.HTTPSHandler()
+
+        class HTTPSConnectionV3(httplib.HTTPSConnection):
+            def __init__(self, *args, **kwargs):
+                httplib.HTTPSConnection.__init__(self, *args, **kwargs)
+
+            def connect(self):
+                sock = socket.create_connection((self.host, self.port), self.timeout)
+                if self._tunnel_host:
+                    self.sock = sock
+                    self._tunnel()
+                try:
+                    self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv3)
+                except ssl.SSLError:
+                    self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23)
+
+        class HTTPSHandlerV3(compat_urllib_request.HTTPSHandler):
+            def https_open(self, req):
+                return self.do_open(HTTPSConnectionV3, req)
+        return HTTPSHandlerV3()
     else:
-        import ssl
+        context = ssl.SSLContext(ssl.PROTOCOL_SSLv3)
-        context = ssl.SSLContext(ssl.PROTOCOL_SSLv23)
         context.set_default_verify_paths()
         
         context.verify_mode = (ssl.CERT_NONE
-                               if opts.no_check_certificate
+                               if opts_no_check_certificate
                                else ssl.CERT_REQUIRED)
         return compat_urllib_request.HTTPSHandler(context=context)
 
@@ -734,6 +754,8 @@ def unified_strdate(date_str):
         '%Y/%m/%d %H:%M:%S',
         '%d.%m.%Y %H:%M',
         '%Y-%m-%dT%H:%M:%SZ',
+        '%Y-%m-%dT%H:%M:%S.%fZ',
+        '%Y-%m-%dT%H:%M:%S.%f0Z',
         '%Y-%m-%dT%H:%M:%S',
     ]
     for expression in format_expressions:
@@ -949,7 +971,16 @@ class locked_file(object):
 
 
 def shell_quote(args):
-    return ' '.join(map(pipes.quote, args))
+    quoted_args = []
+    encoding = sys.getfilesystemencoding()
+    if encoding is None:
+        encoding = 'utf-8'
+    for a in args:
+        if isinstance(a, bytes):
+            # We may get a filename encoded with 'encodeFilename'
+            a = a.decode(encoding)
+        quoted_args.append(pipes.quote(a))
+    return u' '.join(quoted_args)
 
 
 def takewhile_inclusive(pred, seq):
@@ -976,3 +1007,17 @@ def unsmuggle_url(smug_url):
     jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0]
     data = json.loads(jsond)
     return url, data
+
+
+def format_bytes(bytes):
+    if bytes is None:
+        return u'N/A'
+    if type(bytes) is str:
+        bytes = float(bytes)
+    if bytes == 0.0:
+        exponent = 0
+    else:
+        exponent = int(math.log(bytes, 1024.0))
+    suffix = [u'B', u'KiB', u'MiB', u'GiB', u'TiB', u'PiB', u'EiB', u'ZiB', u'YiB'][exponent]
+    converted = float(bytes) / float(1024 ** exponent)
+    return u'%.2f%s' % (converted, suffix)

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2013.11.15'
+__version__ = '2013.12.02'