release 2013.11.13

Suggested in #1743

.gitignore

@@ -25,3 +25,4 @@ updates_key.pem
 *.mp4
 *.part
 test/testdata
+.tox

Makefile

@@ -71,6 +71,7 @@ youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-
 		--exclude '*~' \
 		--exclude '__pycache' \
 		--exclude '.git' \
+		--exclude 'testdata' \
 		-- \
 		bin devscripts test youtube_dl \
 		CHANGELOG LICENSE README.md README.txt \

README.md

@@ -21,6 +21,8 @@ which means you can modify it, redistribute it or use it however you like.
                                sudo if needed)
     -i, --ignore-errors        continue on download errors, for example to to
                                skip unavailable videos in a playlist
+    --abort-on-error           Abort downloading of further videos (in the
+                               playlist or the command line) if an error occurs
     --dump-user-agent          display the current browser identification
     --user-agent UA            specify a custom user agent
     --referer REF              specify a custom referer, use if the video access
@@ -30,9 +32,10 @@ which means you can modify it, redistribute it or use it however you like.
     --extractor-descriptions   Output descriptions of all supported extractors
     --proxy URL                Use the specified HTTP/HTTPS proxy
     --no-check-certificate     Suppress HTTPS certificate validation.
-    --cache-dir None           Location in the filesystem where youtube-dl can
-                               store downloaded information permanently.
-                               ~/.youtube-dl/cache by default
+    --cache-dir DIR            Location in the filesystem where youtube-dl can
+                               store downloaded information permanently. By
+                               default $XDG_CACHE_HOME/youtube-dl or ~/.cache
+                               /youtube-dl .
     --no-cache-dir             Disable filesystem caching
 
 ## Video Selection:
@@ -50,11 +53,16 @@ which means you can modify it, redistribute it or use it however you like.
     --date DATE                download only videos uploaded in this date
     --datebefore DATE          download only videos uploaded before this date
     --dateafter DATE           download only videos uploaded after this date
+    --no-playlist              download only the currently playing video
+    --age-limit YEARS          download only videos suitable for the given age
+    --download-archive FILE    Download only videos not present in the archive
+                               file. Record all downloaded videos in it.
 
 ## Download Options:
-    -r, --rate-limit LIMIT     maximum download rate (e.g. 50k or 44.6m)
+    -r, --rate-limit LIMIT     maximum download rate in bytes per second (e.g.
+                               50K or 4.2M)
     -R, --retries RETRIES      number of retries (default is 10)
-    --buffer-size SIZE         size of download buffer (e.g. 1024 or 16k)
+    --buffer-size SIZE         size of download buffer (e.g. 1024 or 16K)
                                (default is 1024)
     --no-resize-buffer         do not automatically adjust the buffer size. By
                                default, the buffer size is automatically resized
@@ -70,7 +78,10 @@ which means you can modify it, redistribute it or use it however you like.
                                %(uploader_id)s for the uploader nickname if
                                different, %(autonumber)s to get an automatically
                                incremented number, %(ext)s for the filename
-                               extension, %(upload_date)s for the upload date
+                               extension, %(format)s for the format description
+                               (like "22 - 1280x720" or "HD"),%(format_id)s for
+                               the unique id of the format (like Youtube's
+                               itags: "137"),%(upload_date)s for the upload date
                                (YYYYMMDD), %(extractor)s for the provider
                                (youtube, metacafe, etc), %(id)s for the video id
                                , %(playlist)s for the playlist the video is in,
@@ -81,12 +92,14 @@ which means you can modify it, redistribute it or use it however you like.
                                ownloads/%(uploader)s/%(title)s-%(id)s.%(ext)s' .
     --autonumber-size NUMBER   Specifies the number of digits in %(autonumber)s
                                when it is present in output filename template or
-                               --autonumber option is given
+                               --auto-number option is given
     --restrict-filenames       Restrict filenames to only ASCII characters, and
                                avoid "&" and spaces in filenames
     -a, --batch-file FILE      file containing URLs to download ('-' for stdin)
     -w, --no-overwrites        do not overwrite files
-    -c, --continue             resume partially downloaded files
+    -c, --continue             force resume of partially downloaded files. By
+                               default, youtube-dl will resume downloads if
+                               possible.
     --no-continue              do not resume partially downloaded files (restart
                                from beginning)
     --cookies FILE             file to read cookies from and dump cookie jar in
@@ -95,6 +108,7 @@ which means you can modify it, redistribute it or use it however you like.
                                file modification time
     --write-description        write video description to a .description file
     --write-info-json          write video metadata to a .info.json file
+    --write-annotations        write video annotations to a .annotation file
     --write-thumbnail          write thumbnail image to disk
 
 ## Verbosity / Simulation Options:
@@ -115,6 +129,8 @@ which means you can modify it, redistribute it or use it however you like.
     -v, --verbose              print various debugging information
     --dump-intermediate-pages  print downloaded pages to debug problems(very
                                verbose)
+    --write-pages              Write downloaded pages to files in the current
+                               directory
 
 ## Video Format Options:
     -f, --format FORMAT        video format code, specifiy the order of
@@ -161,6 +177,7 @@ which means you can modify it, redistribute it or use it however you like.
                                processed files are overwritten by default
     --embed-subs               embed subtitles in the video (only for mp4
                                videos)
+    --add-metadata             add metadata to the files
 
 # CONFIGURATION
 

devscripts/bash-completion.in

@@ -1,4 +1,4 @@
-__youtube-dl()
+__youtube_dl()
 {
     local cur prev opts
     COMPREPLY=()
@@ -15,4 +15,4 @@ __youtube-dl()
     fi
 }
 
-complete -F __youtube-dl youtube-dl
+complete -F __youtube_dl youtube-dl

devscripts/check-porn.py

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

devscripts/gh-pages/update-sites.py

@@ -16,10 +16,11 @@ def main():
     ie_htmls = []
     for ie in sorted(youtube_dl.gen_extractors(), key=lambda i: i.IE_NAME.lower()):
         ie_html = '<b>{}</b>'.format(ie.IE_NAME)
-        try:
+        ie_desc = getattr(ie, 'IE_DESC', None)
+        if ie_desc is False:
+            continue
+        elif ie_desc is not None:
             ie_html += ': {}'.format(ie.IE_DESC)
-        except AttributeError:
-            pass
         if ie.working() == False:
             ie_html += ' (Currently broken)'
         ie_htmls.append('<li>{}</li>'.format(ie_html))

devscripts/release.sh

@@ -88,10 +88,6 @@ ROOT=$(pwd)
     "$ROOT/devscripts/gh-pages/update-sites.py"
     git add *.html *.html.in update
     git commit -m "release $version"
-    git show HEAD
-    read -p "Is it good, can I push? (y/n) " -n 1
-    if [[ ! $REPLY =~ ^[Yy]$ ]]; then exit 1; fi
-    echo
     git push "$ROOT" gh-pages
     git push "$ORIGIN_URL" gh-pages
 )

devscripts/youtube_genalgo.py

@@ -1,116 +0,0 @@
-#!/usr/bin/env python
-# encoding: utf-8
-
-# Generate youtube signature algorithm from test cases
-
-import sys
-
-tests = [
-    # 93 - vfl79wBKW 2013/07/20
-    (u"qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`~\"€",
-     u".>/?;:|}][{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWQ098765'321mnbvcxzasdfghjklpoiu"),
-    # 92 - vflQw-fB4 2013/07/17
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`~\"",
-     "mrtyuioplkjhgfdsazxcvbnq1234567890QWERTY}IOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]\"|:;"),
-    # 91 - vfl79wBKW 2013/07/20 (sporadic)
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`~",
-     "/?;:|}][{=+-_)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYTREWQ09876543.1mnbvcxzasdfghjklpoiu"),
-    # 90
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'`",
-     "mrtyuioplkjhgfdsazxcvbne1234567890QWER[YUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={`]}|"),
-    # 89 
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<'",
-     "/?;:|}<[{=+-_)(*&^%$#@!MqBVCXZASDFGHJKLPOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuyt"),
-    # 88 - vflapUV9V 2013/08/28
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[]}|:;?/>.<",
-     "ioplkjhgfdsazxcvbnm12<4567890QWERTYUIOZLKJHGFDSAeXCVBNM!@#$%^&*()_-+={[]}|:;?/>.3"),
-    # 87
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$^&*()_-+={[]}|:;?/>.<",
-     "uioplkjhgfdsazxcvbnm1t34567890QWE2TYUIOPLKJHGFDSAZXCVeNM!@#$^&*()_-+={[]}|:;?/>.<"),
-    # 86 - vflHql6Pr 2013/09/24
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<",
-     ";}|[{=+-d)(*&^%$#@!MNBVCXZASDFGHJKLPOIUYT_EWQ0987654321mnbvcxzas/fghjklpoiuytrewq"),
-    # 85 - vflkuzxcs 2013/09/11
-    ('0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ!"#$%&\'()*+,-./:;<=>?@[',
-     '3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@'),
-    # 84 - vflHql6Pr 2013/09/24 (sporadic)
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[};?>.<",
-     "}[{=+-_)g*&^%$#@!MNBVCXZASDFGHJKLPOIUYTRE(Q0987654321mnbvcxzasdf?hjklpoiuytrewq"),
-    # 83
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!#$%^&*()_+={[};?/>.<",
-     ".>/?;}[{=+_)(*&^%<#!MNBVCXZASPFGHJKLwOIUYTREWQ0987654321mnbvcxzasdfghjklpoiuytreq"),
-    # 82 - vflGNjMhJ 2013/09/12
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.<",
-     ".>/?;}[<=+-(*&^%$#@!MNBVCXeASDFGHKLPOqUYTREWQ0987654321mnbvcxzasdfghjklpoiuytrIwZ"),
-    # 81 - vflLC8JvQ 2013/07/25
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>.",
-     "C>/?;}[{=+-(*&^%$#@!MNBVYXZASDFGHKLPOIU.TREWQ0q87659321mnbvcxzasdfghjkl4oiuytrewp"),
-    # 80 - vflZK4ZYR 2013/08/23 (sporadic)
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/>",
-     "wertyuioplkjhgfdsaqxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&z(-+={[};?/>"),
-    # 79 - vflLC8JvQ 2013/07/25 (sporadic)
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKHGFDSAZXCVBNM!@#$%^&*(-+={[};?/",
-     "Z?;}[{=+-(*&^%$#@!MNBVCXRASDFGHKLPOIUYT/EWQ0q87659321mnbvcxzasdfghjkl4oiuytrewp"),
-]
-
-tests_age_gate = [
-    # 86 - vflqinMWD
-    ("qwertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!@#$%^&*()_-+={[|};?/>.<",
-     "ertyuioplkjhgfdsazxcvbnm1234567890QWERTYUIOPLKJHGFDSAZXCVBNM!/#$%^&*()_-+={[|};?@"),
-]
-
-def find_matching(wrong, right):
-    idxs = [wrong.index(c) for c in right]
-    return compress(idxs)
-    return ('s[%d]' % i for i in idxs)
-
-def compress(idxs):
-    def _genslice(start, end, step):
-        starts = '' if start == 0 else str(start)
-        ends = ':%d' % (end+step)
-        steps = '' if step == 1 else (':%d' % step)
-        return 's[%s%s%s]' % (starts, ends, steps)
-
-    step = None
-    for i, prev in zip(idxs[1:], idxs[:-1]):
-        if step is not None:
-            if i - prev == step:
-                continue
-            yield _genslice(start, prev, step)
-            step = None
-            continue
-        if i - prev in [-1, 1]:
-            step = i - prev
-            start = prev
-            continue
-        else:
-            yield 's[%d]' % prev
-    if step is None:
-        yield 's[%d]' % i
-    else:
-        yield _genslice(start, i, step)
-
-def _assert_compress(inp, exp):
-    res = list(compress(inp))
-    if res != exp:
-        print('Got %r, expected %r' % (res, exp))
-        assert res == exp
-_assert_compress([0,2,4,6], ['s[0]', 's[2]', 's[4]', 's[6]'])
-_assert_compress([0,1,2,4,6,7], ['s[:3]', 's[4]', 's[6:8]'])
-_assert_compress([8,0,1,2,4,7,6,9], ['s[8]', 's[:3]', 's[4]', 's[7:5:-1]', 's[9]'])
-
-def gen(wrong, right, indent):
-    code = ' + '.join(find_matching(wrong, right))
-    return 'if len(s) == %d:\n%s    return %s\n' % (len(wrong), indent, code)
-
-def genall(tests):
-    indent = ' ' * 8
-    return indent + (indent + 'el').join(gen(wrong, right, indent) for wrong,right in tests)
-
-def main():
-    print(genall(tests))
-    print(u'    Age gate:')
-    print(genall(tests_age_gate))
-
-if __name__ == '__main__':
-    main()

setup.py

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

test/helper.py

@@ -1,38 +1,80 @@
+import errno
 import io
+import hashlib
 import json
 import os.path
+import re
+import types
+import sys
 
 import youtube_dl.extractor
-from youtube_dl import YoutubeDL, YoutubeDLHandler
-from youtube_dl.utils import (
-    compat_cookiejar,
-    compat_urllib_request,
-)
+from youtube_dl import YoutubeDL
+from youtube_dl.utils import preferredencoding
 
-# General configuration (from __init__, not very elegant...)
-jar = compat_cookiejar.CookieJar()
-cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
-proxy_handler = compat_urllib_request.ProxyHandler()
-opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
-compat_urllib_request.install_opener(opener)
 
-PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
+def global_setup():
+    youtube_dl._setup_opener(timeout=10)
+
+
+def get_params(override=None):
+    PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
+                                   "parameters.json")
     with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
         parameters = json.load(pf)
+    if override:
+        parameters.update(override)
+    return parameters
+
+
+def try_rm(filename):
+    """ Remove a file if it exists """
+    try:
+        os.remove(filename)
+    except OSError as ose:
+        if ose.errno != errno.ENOENT:
+            raise
+
+
+def report_warning(message):
+    '''
+    Print the message to stderr, it will be prefixed with 'WARNING:'
+    If stderr is a tty file the 'WARNING:' will be colored
+    '''
+    if sys.stderr.isatty() and os.name != 'nt':
+        _msg_header = u'\033[0;33mWARNING:\033[0m'
+    else:
+        _msg_header = u'WARNING:'
+    output = u'%s %s\n' % (_msg_header, message)
+    if 'b' in getattr(sys.stderr, 'mode', '') or sys.version_info[0] < 3:
+        output = output.encode(preferredencoding())
+    sys.stderr.write(output)
+
 
 class FakeYDL(YoutubeDL):
-    def __init__(self):
-        self.result = []
+    def __init__(self, override=None):
         # Different instances of the downloader can't share the same dictionary
         # some test set the "sublang" parameter, which would break the md5 checks.
-        self.params = dict(parameters)
-    def to_screen(self, s):
+        params = get_params(override=override)
+        super(FakeYDL, self).__init__(params)
+        self.result = []
+        
+    def to_screen(self, s, skip_eol=None):
         print(s)
+
     def trouble(self, s, tb=None):
         raise Exception(s)
+
     def download(self, x):
         self.result.append(x)
 
+    def expect_warning(self, regex):
+        # Silence an expected warning matching a regex
+        old_report_warning = self.report_warning
+        def report_warning(self, message):
+            if re.match(regex, message): return
+            old_report_warning(message)
+        self.report_warning = types.MethodType(report_warning, self)
+
 def get_testcases():
     for ie in youtube_dl.extractor.gen_extractors():
         t = getattr(ie, '_TEST', None)
@@ -42,3 +84,6 @@ def get_testcases():
         for t in getattr(ie, '_TESTS', []):
             t['name'] = type(ie).__name__[:-len('IE')]
             yield t
+
+
+md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()

test/test_YoutubeDL.py

@@ -0,0 +1,145 @@
+#!/usr/bin/env python
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL
+
+
+class YDL(FakeYDL):
+    def __init__(self, *args, **kwargs):
+        super(YDL, self).__init__(*args, **kwargs)
+        self.downloaded_info_dicts = []
+        self.msgs = []
+
+    def process_info(self, info_dict):
+        self.downloaded_info_dicts.append(info_dict)
+
+    def to_screen(self, msg):
+        self.msgs.append(msg)
+
+
+class TestFormatSelection(unittest.TestCase):
+    def test_prefer_free_formats(self):
+        # Same resolution => download webm
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = True
+        formats = [
+            {u'ext': u'webm', u'height': 460},
+            {u'ext': u'mp4',  u'height': 460},
+        ]
+        info_dict = {u'formats': formats, u'extractor': u'test'}
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'ext'], u'webm')
+
+        # Different resolution => download best quality (mp4)
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = True
+        formats = [
+            {u'ext': u'webm', u'height': 720},
+            {u'ext': u'mp4', u'height': 1080},
+        ]
+        info_dict[u'formats'] = formats
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'ext'], u'mp4')
+
+        # No prefer_free_formats => keep original formats order
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = False
+        formats = [
+            {u'ext': u'webm', u'height': 720},
+            {u'ext': u'flv', u'height': 720},
+        ]
+        info_dict[u'formats'] = formats
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'ext'], u'flv')
+
+    def test_format_limit(self):
+        formats = [
+            {u'format_id': u'meh', u'url': u'http://example.com/meh'},
+            {u'format_id': u'good', u'url': u'http://example.com/good'},
+            {u'format_id': u'great', u'url': u'http://example.com/great'},
+            {u'format_id': u'excellent', u'url': u'http://example.com/exc'},
+        ]
+        info_dict = {
+            u'formats': formats, u'extractor': u'test', 'id': 'testvid'}
+
+        ydl = YDL()
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'format_id'], u'excellent')
+
+        ydl = YDL({'format_limit': 'good'})
+        assert ydl.params['format_limit'] == 'good'
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'format_id'], u'good')
+
+        ydl = YDL({'format_limit': 'great', 'format': 'all'})
+        ydl.process_ie_result(info_dict)
+        self.assertEqual(ydl.downloaded_info_dicts[0][u'format_id'], u'meh')
+        self.assertEqual(ydl.downloaded_info_dicts[1][u'format_id'], u'good')
+        self.assertEqual(ydl.downloaded_info_dicts[2][u'format_id'], u'great')
+        self.assertTrue('3' in ydl.msgs[0])
+
+        ydl = YDL()
+        ydl.params['format_limit'] = 'excellent'
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded[u'format_id'], u'excellent')
+
+    def test_format_selection(self):
+        formats = [
+            {u'format_id': u'35', u'ext': u'mp4'},
+            {u'format_id': u'45', u'ext': u'webm'},
+            {u'format_id': u'47', u'ext': u'webm'},
+            {u'format_id': u'2', u'ext': u'flv'},
+        ]
+        info_dict = {u'formats': formats, u'extractor': u'test'}
+
+        ydl = YDL({'format': u'20/47'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'47')
+
+        ydl = YDL({'format': u'20/71/worst'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'35')
+
+        ydl = YDL()
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'2')
+
+        ydl = YDL({'format': u'webm/mp4'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'47')
+
+        ydl = YDL({'format': u'3gp/40/mp4'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], u'35')
+
+    def test_add_extra_info(self):
+        test_dict = {
+            'extractor': 'Foo',
+        }
+        extra_info = {
+            'extractor': 'Bar',
+            'playlist': 'funny videos',
+        }
+        YDL.add_extra_info(test_dict, extra_info)
+        self.assertEqual(test_dict['extractor'], 'Foo')
+        self.assertEqual(test_dict['playlist'], 'funny videos')
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_age_restriction.py

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

test/test_all_urls.py

@@ -1,14 +1,20 @@
 #!/usr/bin/env python
 
-import sys
-import unittest
-
 # Allow direct execution
 import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+from test.helper import get_testcases
+
+from youtube_dl.extractor import (
+    gen_extractors,
+    JustinTVIE,
+    YoutubeIE,
+)
 
-from youtube_dl.extractor import YoutubeIE, YoutubePlaylistIE, YoutubeChannelIE, JustinTVIE, gen_extractors
-from helper import get_testcases
 
 class TestAllURLsMatching(unittest.TestCase):
     def setUp(self):

test/test_dailymotion_subtitles.py

@@ -1,71 +0,0 @@
-#!/usr/bin/env python
-
-import sys
-import unittest
-import json
-import io
-import hashlib
-
-# Allow direct execution
-import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-
-from youtube_dl.extractor import DailymotionIE
-from youtube_dl.utils import *
-from helper import FakeYDL
-
-md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
-
-class TestDailymotionSubtitles(unittest.TestCase):
-    def setUp(self):
-        self.DL = FakeYDL()
-        self.url = 'http://www.dailymotion.com/video/xczg00'
-    def getInfoDict(self):
-        IE = DailymotionIE(self.DL)
-        info_dict = IE.extract(self.url)
-        return info_dict
-    def getSubtitles(self):
-        info_dict = self.getInfoDict()
-        return info_dict[0]['subtitles']
-    def test_no_writesubtitles(self):
-        subtitles = self.getSubtitles()
-        self.assertEqual(subtitles, None)
-    def test_subtitles(self):
-        self.DL.params['writesubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '976553874490cba125086bbfea3ff76f')
-    def test_subtitles_lang(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitleslangs'] = ['fr']
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['fr']), '594564ec7d588942e384e920e5341792')
-    def test_allsubtitles(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['allsubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles.keys()), 5)
-    def test_list_subtitles(self):
-        self.DL.params['listsubtitles'] = True
-        info_dict = self.getInfoDict()
-        self.assertEqual(info_dict, None)
-    def test_automatic_captions(self):
-        self.DL.params['writeautomaticsub'] = True
-        self.DL.params['subtitleslang'] = ['en']
-        subtitles = self.getSubtitles()
-        self.assertTrue(len(subtitles.keys()) == 0)
-    def test_nosubtitles(self):
-        self.url = 'http://www.dailymotion.com/video/x12u166_le-zapping-tele-star-du-08-aout-2013_tv'
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['allsubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles), 0)
-    def test_multiple_langs(self):
-        self.DL.params['writesubtitles'] = True
-        langs = ['es', 'fr', 'de']
-        self.DL.params['subtitleslangs'] = langs
-        subtitles = self.getSubtitles()
-        for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
-
-if __name__ == '__main__':
-    unittest.main()

test/test_download.py

@@ -1,43 +1,40 @@
 #!/usr/bin/env python
 
-import errno
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import (
+    get_params,
+    get_testcases,
+    global_setup,
+    try_rm,
+    md5,
+    report_warning
+)
+global_setup()
+
+
 import hashlib
 import io
-import os
 import json
-import unittest
-import sys
 import socket
-import binascii
-
-# Allow direct execution
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 import youtube_dl.YoutubeDL
-from youtube_dl.utils import *
-
-PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
+from youtube_dl.utils import (
+    compat_str,
+    compat_urllib_error,
+    compat_HTTPError,
+    DownloadError,
+    ExtractorError,
+    UnavailableVideoError,
+)
+from youtube_dl.extractor import get_info_extractor
 
 RETRIES = 3
 
-# General configuration (from __init__, not very elegant...)
-jar = compat_cookiejar.CookieJar()
-cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
-proxy_handler = compat_urllib_request.ProxyHandler()
-opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
-compat_urllib_request.install_opener(opener)
-socket.setdefaulttimeout(10)
-
-def _try_rm(filename):
-    """ Remove a file if it exists """
-    try:
-        os.remove(filename)
-    except OSError as ose:
-        if ose.errno != errno.ENOENT:
-            raise
-
-md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
-
 class YoutubeDL(youtube_dl.YoutubeDL):
     def __init__(self, *args, **kwargs):
         self.to_stderr = self.to_screen
@@ -54,17 +51,12 @@ def _file_md5(fn):
     with open(fn, 'rb') as f:
         return hashlib.md5(f.read()).hexdigest()
 
-from helper import get_testcases
 defs = get_testcases()
 
-with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
-    parameters = json.load(pf)
-
 
 class TestDownload(unittest.TestCase):
     maxDiff = None
     def setUp(self):
-        self.parameters = parameters
         self.defs = defs
 
 ### Dynamically generate tests
@@ -72,20 +64,27 @@ def generator(test_case):
 
     def test_template(self):
         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
+        other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])]
         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))
-        if not ie._WORKING:
+        if not ie.working():
             print_skipping('IE marked as not _WORKING')
             return
-        if 'playlist' not in test_case and not test_case['file']:
-            print_skipping('No output file specified')
+        if 'playlist' not in test_case:
+            info_dict = test_case.get('info_dict', {})
+            if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
+                print_skipping('The output file cannot be know, the "file" '
+                    'key is missing or the info_dict is incomplete')
                 return
         if 'skip' in test_case:
             print_skipping(test_case['skip'])
             return
+        for other_ie in other_ies:
+            if not other_ie.working():
+                print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key())
+                return
 
-        params = self.parameters.copy()
-        params.update(test_case.get('params', {}))
+        params = get_params(test_case.get('params', {}))
 
         ydl = YoutubeDL(params)
         ydl.add_default_info_extractors()
@@ -95,35 +94,47 @@ def generator(test_case):
                 finished_hook_called.add(status['filename'])
         ydl.fd.add_progress_hook(_hook)
 
+        def get_tc_filename(tc):
+            return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
+
         test_cases = test_case.get('playlist', [test_case])
+        def try_rm_tcs_files():
             for tc in test_cases:
-            _try_rm(tc['file'])
-            _try_rm(tc['file'] + '.part')
-            _try_rm(tc['file'] + '.info.json')
+                tc_filename = get_tc_filename(tc)
+                try_rm(tc_filename)
+                try_rm(tc_filename + '.part')
+                try_rm(tc_filename + '.info.json')
+        try_rm_tcs_files()
         try:
-            for retry in range(1, RETRIES + 1):
+            try_num = 1
+            while True:
                 try:
                     ydl.download([test_case['url']])
                 except (DownloadError, ExtractorError) as err:
-                    if retry == RETRIES: raise
-
                     # Check if the exception is not a network related one
-                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
+                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
                         raise
 
-                    print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
+                    if try_num == RETRIES:
+                        report_warning(u'Failed due to network errors, skipping...')
+                        return
+
+                    print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
+
+                    try_num += 1
                 else:
                     break
 
             for tc in test_cases:
+                tc_filename = get_tc_filename(tc)
                 if not test_case.get('params', {}).get('skip_download', False):
-                    self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
-                    self.assertTrue(tc['file'] in finished_hook_called)
-                self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
+                    self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
+                    self.assertTrue(tc_filename in finished_hook_called)
+                self.assertTrue(os.path.exists(tc_filename + '.info.json'))
                 if 'md5' in tc:
-                    md5_for_file = _file_md5(tc['file'])
+                    md5_for_file = _file_md5(tc_filename)
                     self.assertEqual(md5_for_file, tc['md5'])
-                with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
+                with io.open(tc_filename + '.info.json', encoding='utf-8') as infof:
                     info_dict = json.load(infof)
                 for (info_field, expected) in tc.get('info_dict', {}).items():
                     if isinstance(expected, compat_str) and expected.startswith('md5:'):
@@ -143,11 +154,11 @@ def generator(test_case):
                 # Check for the presence of mandatory fields
                 for key in ('id', 'url', 'title', 'ext'):
                     self.assertTrue(key in info_dict.keys() and info_dict[key])
+                # Check for mandatory fields that are automatically set by YoutubeDL
+                for key in ['webpage_url', 'extractor', 'extractor_key']:
+                    self.assertTrue(info_dict.get(key), u'Missing field: %s' % key)
         finally:
-            for tc in test_cases:
-                _try_rm(tc['file'])
-                _try_rm(tc['file'] + '.part')
-                _try_rm(tc['file'] + '.info.json')
+            try_rm_tcs_files()
 
     return test_template
 

test/test_playlists.py

@@ -1,25 +1,29 @@
 #!/usr/bin/env python
 # encoding: utf-8
 
-import sys
-import unittest
-import json
 
 # Allow direct execution
 import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL, global_setup
+global_setup()
+
 
 from youtube_dl.extractor import (
     DailymotionPlaylistIE,
     DailymotionUserIE,
     VimeoChannelIE,
     UstreamChannelIE,
+    SoundcloudSetIE,
     SoundcloudUserIE,
     LivestreamIE,
+    NHLVideocenterIE,
+    BambuserChannelIE,
 )
-from youtube_dl.utils import *
 
-from helper import FakeYDL
 
 class TestPlaylists(unittest.TestCase):
     def assertIsPlaylist(self, info):
@@ -58,6 +62,14 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['id'], u'5124905')
         self.assertTrue(len(result['entries']) >= 11)
 
+    def test_soundcloud_set(self):
+        dl = FakeYDL()
+        ie = SoundcloudSetIE(dl)
+        result = ie.extract('https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], u'The Royal Concept EP')
+        self.assertTrue(len(result['entries']) >= 6)
+
     def test_soundcloud_user(self):
         dl = FakeYDL()
         ie = SoundcloudUserIE(dl)
@@ -74,5 +86,22 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['title'], u'TEDCity2.0 (English)')
         self.assertTrue(len(result['entries']) >= 4)
 
+    def test_nhl_videocenter(self):
+        dl = FakeYDL()
+        ie = NHLVideocenterIE(dl)
+        result = ie.extract('http://video.canucks.nhl.com/videocenter/console?catid=999')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], u'999')
+        self.assertEqual(result['title'], u'Highlights')
+        self.assertEqual(len(result['entries']), 12)
+
+    def test_bambuser_channel(self):
+        dl = FakeYDL()
+        ie = BambuserChannelIE(dl)
+        result = ie.extract('http://bambuser.com/channel/pixelversity')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], u'pixelversity')
+        self.assertTrue(len(result['entries']) >= 66)
+
 if __name__ == '__main__':
     unittest.main()

test/test_subtitles.py

@@ -0,0 +1,211 @@
+#!/usr/bin/env python
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL, global_setup, md5
+global_setup()
+
+
+from youtube_dl.extractor import (
+    YoutubeIE,
+    DailymotionIE,
+    TEDIE,
+)
+
+
+class BaseTestSubtitles(unittest.TestCase):
+    url = None
+    IE = None
+    def setUp(self):
+        self.DL = FakeYDL()
+        self.ie = self.IE(self.DL)
+
+    def getInfoDict(self):
+        info_dict = self.ie.extract(self.url)
+        return info_dict
+
+    def getSubtitles(self):
+        info_dict = self.getInfoDict()
+        return info_dict['subtitles']
+
+
+class TestYoutubeSubtitles(BaseTestSubtitles):
+    url = 'QRS8MkLhQmM'
+    IE = YoutubeIE
+
+    def getSubtitles(self):
+        info_dict = self.getInfoDict()
+        return info_dict[0]['subtitles']
+
+    def test_youtube_no_writesubtitles(self):
+        self.DL.params['writesubtitles'] = False
+        subtitles = self.getSubtitles()
+        self.assertEqual(subtitles, None)
+
+    def test_youtube_subtitles(self):
+        self.DL.params['writesubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260')
+
+    def test_youtube_subtitles_lang(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitleslangs'] = ['it']
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d')
+
+    def test_youtube_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles.keys()), 13)
+
+    def test_youtube_subtitles_sbv_format(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitlesformat'] = 'sbv'
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b')
+
+    def test_youtube_subtitles_vtt_format(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitlesformat'] = 'vtt'
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '356cdc577fde0c6783b9b822e7206ff7')
+
+    def test_youtube_list_subtitles(self):
+        self.DL.expect_warning(u'Video doesn\'t have automatic captions')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
+        self.assertEqual(info_dict, None)
+
+    def test_youtube_automatic_captions(self):
+        self.url = '8YoUxe5ncPo'
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslangs'] = ['it']
+        subtitles = self.getSubtitles()
+        self.assertTrue(subtitles['it'] is not None)
+
+    def test_youtube_nosubtitles(self):
+        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.url = 'sAjKT8FhjI8'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles), 0)
+
+    def test_youtube_multiple_langs(self):
+        self.url = 'QRS8MkLhQmM'
+        self.DL.params['writesubtitles'] = True
+        langs = ['it', 'fr', 'de']
+        self.DL.params['subtitleslangs'] = langs
+        subtitles = self.getSubtitles()
+        for lang in langs:
+            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+
+
+class TestDailymotionSubtitles(BaseTestSubtitles):
+    url = 'http://www.dailymotion.com/video/xczg00'
+    IE = DailymotionIE
+
+    def test_no_writesubtitles(self):
+        subtitles = self.getSubtitles()
+        self.assertEqual(subtitles, None)
+
+    def test_subtitles(self):
+        self.DL.params['writesubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '976553874490cba125086bbfea3ff76f')
+
+    def test_subtitles_lang(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitleslangs'] = ['fr']
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['fr']), '594564ec7d588942e384e920e5341792')
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles.keys()), 5)
+
+    def test_list_subtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
+        self.assertEqual(info_dict, None)
+
+    def test_automatic_captions(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslang'] = ['en']
+        subtitles = self.getSubtitles()
+        self.assertTrue(len(subtitles.keys()) == 0)
+
+    def test_nosubtitles(self):
+        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.url = 'http://www.dailymotion.com/video/x12u166_le-zapping-tele-star-du-08-aout-2013_tv'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles), 0)
+
+    def test_multiple_langs(self):
+        self.DL.params['writesubtitles'] = True
+        langs = ['es', 'fr', 'de']
+        self.DL.params['subtitleslangs'] = langs
+        subtitles = self.getSubtitles()
+        for lang in langs:
+            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+
+
+class TestTedSubtitles(BaseTestSubtitles):
+    url = 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html'
+    IE = TEDIE
+
+    def test_no_writesubtitles(self):
+        subtitles = self.getSubtitles()
+        self.assertEqual(subtitles, None)
+
+    def test_subtitles(self):
+        self.DL.params['writesubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '2154f31ff9b9f89a0aa671537559c21d')
+
+    def test_subtitles_lang(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitleslangs'] = ['fr']
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['fr']), '7616cbc6df20ec2c1204083c83871cf6')
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles.keys()), 28)
+
+    def test_list_subtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
+        self.assertEqual(info_dict, None)
+
+    def test_automatic_captions(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslang'] = ['en']
+        subtitles = self.getSubtitles()
+        self.assertTrue(len(subtitles.keys()) == 0)
+
+    def test_multiple_langs(self):
+        self.DL.params['writesubtitles'] = True
+        langs = ['es', 'fr', 'de']
+        self.DL.params['subtitleslangs'] = langs
+        subtitles = self.getSubtitles()
+        for lang in langs:
+            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -1,14 +1,15 @@
 #!/usr/bin/env python
-
-# Various small unit tests
-
-import sys
-import unittest
-import xml.etree.ElementTree
+# coding: utf-8
 
 # Allow direct execution
 import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+# Various small unit tests
+import xml.etree.ElementTree
 
 #from youtube_dl.utils import htmlentity_transform
 from youtube_dl.utils import (
@@ -20,6 +21,9 @@ from youtube_dl.utils import (
     unified_strdate,
     find_xpath_attr,
     get_meta_content,
+    xpath_with_ns,
+    smuggle_url,
+    unsmuggle_url,
 )
 
 if sys.version_info < (3, 0):
@@ -141,5 +145,31 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(get_meta('description'), u'foo & bar')
         self.assertEqual(get_meta('author'), 'Plato')
 
+    def test_xpath_with_ns(self):
+        testxml = u'''<root xmlns:media="http://example.com/">
+            <media:song>
+                <media:author>The Author</media:author>
+                <url>http://server.com/download.mp3</url>
+            </media:song>
+        </root>'''
+        doc = xml.etree.ElementTree.fromstring(testxml)
+        find = lambda p: doc.find(xpath_with_ns(p, {'media': 'http://example.com/'}))
+        self.assertTrue(find('media:song') is not None)
+        self.assertEqual(find('media:song/media:author').text, u'The Author')
+        self.assertEqual(find('media:song/url').text, u'http://server.com/download.mp3')
+
+    def test_smuggle_url(self):
+        data = {u"ö": u"ö", u"abc": [3]}
+        url = 'https://foo.bar/baz?x=y#a'
+        smug_url = smuggle_url(url, data)
+        unsmug_url, unsmug_data = unsmuggle_url(smug_url)
+        self.assertEqual(url, unsmug_url)
+        self.assertEqual(data, unsmug_data)
+
+        res_url, res_data = unsmuggle_url(url)
+        self.assertEqual(res_url, url)
+        self.assertEqual(res_data, None)
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_write_annotations.py

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

test/test_write_info_json.py

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

test/test_youtube_lists.py

@@ -1,17 +1,23 @@
 #!/usr/bin/env python
 
-import sys
-import unittest
-import json
-
 # Allow direct execution
 import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from youtube_dl.extractor import YoutubeUserIE, YoutubePlaylistIE, YoutubeIE, YoutubeChannelIE, YoutubeShowIE
-from youtube_dl.utils import *
+from test.helper import FakeYDL, global_setup
+global_setup()
+
+
+from youtube_dl.extractor import (
+    YoutubeUserIE,
+    YoutubePlaylistIE,
+    YoutubeIE,
+    YoutubeChannelIE,
+    YoutubeShowIE,
+)
 
-from helper import FakeYDL
 
 class TestYoutubeLists(unittest.TestCase):
     def assertIsPlaylist(self, info):
@@ -27,6 +33,14 @@ class TestYoutubeLists(unittest.TestCase):
         ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
         self.assertEqual(ytie_results, [ 'bV9L5Ht9LgY', 'FXxLjLQi3Fg', 'tU3Bgo5qJZE'])
 
+    def test_youtube_playlist_noplaylist(self):
+        dl = FakeYDL()
+        dl.params['noplaylist'] = True
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('https://www.youtube.com/watch?v=FXxLjLQi3Fg&list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
+        self.assertEqual(result['_type'], 'url')
+        self.assertEqual(YoutubeIE()._extract_id(result['url']), 'FXxLjLQi3Fg')
+
     def test_issue_673(self):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
@@ -92,7 +106,7 @@ class TestYoutubeLists(unittest.TestCase):
         dl = FakeYDL()
         ie = YoutubeShowIE(dl)
         result = ie.extract('http://www.youtube.com/show/airdisasters')
-        self.assertTrue(len(result) >= 4)
+        self.assertTrue(len(result) >= 3)
 
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_signature.py

@@ -1,14 +1,18 @@
 #!/usr/bin/env python
 
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import global_setup
+global_setup()
+
+
 import io
 import re
 import string
-import sys
-import unittest
-
-# Allow direct execution
-import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from youtube_dl.extractor import YoutubeIE
 from youtube_dl.utils import compat_str, compat_urlretrieve

test/test_youtube_subtitles.py

@@ -1,84 +0,0 @@
-#!/usr/bin/env python
-
-import sys
-import unittest
-import json
-import io
-import hashlib
-
-# Allow direct execution
-import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-
-from youtube_dl.extractor import YoutubeIE
-from youtube_dl.utils import *
-from helper import FakeYDL
-
-md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
-
-class TestYoutubeSubtitles(unittest.TestCase):
-    def setUp(self):
-        self.DL = FakeYDL()
-        self.url = 'QRS8MkLhQmM'
-    def getInfoDict(self):
-        IE = YoutubeIE(self.DL)
-        info_dict = IE.extract(self.url)
-        return info_dict
-    def getSubtitles(self):
-        info_dict = self.getInfoDict()
-        return info_dict[0]['subtitles']        
-    def test_youtube_no_writesubtitles(self):
-        self.DL.params['writesubtitles'] = False
-        subtitles = self.getSubtitles()
-        self.assertEqual(subtitles, None)
-    def test_youtube_subtitles(self):
-        self.DL.params['writesubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260')
-    def test_youtube_subtitles_lang(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitleslangs'] = ['it']
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d')
-    def test_youtube_allsubtitles(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['allsubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles.keys()), 13)
-    def test_youtube_subtitles_sbv_format(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitlesformat'] = 'sbv'
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b')
-    def test_youtube_subtitles_vtt_format(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitlesformat'] = 'vtt'
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '356cdc577fde0c6783b9b822e7206ff7')
-    def test_youtube_list_subtitles(self):
-        self.DL.params['listsubtitles'] = True
-        info_dict = self.getInfoDict()
-        self.assertEqual(info_dict, None)
-    def test_youtube_automatic_captions(self):
-        self.url = '8YoUxe5ncPo'
-        self.DL.params['writeautomaticsub'] = True
-        self.DL.params['subtitleslangs'] = ['it']
-        subtitles = self.getSubtitles()
-        self.assertTrue(subtitles['it'] is not None)
-    def test_youtube_nosubtitles(self):
-        self.url = 'sAjKT8FhjI8'
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['allsubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles), 0)
-    def test_youtube_multiple_langs(self):
-        self.url = 'QRS8MkLhQmM'
-        self.DL.params['writesubtitles'] = True
-        langs = ['it', 'fr', 'de']
-        self.DL.params['subtitleslangs'] = langs
-        subtitles = self.getSubtitles()
-        for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
-
-if __name__ == '__main__':
-    unittest.main()

tox.ini

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

youtube_dl/FileDownloader.py

@@ -4,12 +4,19 @@ import re
 import subprocess
 import sys
 import time
-import traceback
 
 if os.name == 'nt':
     import ctypes
 
-from .utils import *
+from .utils import (
+    compat_urllib_error,
+    compat_urllib_request,
+    ContentTooShortError,
+    determine_ext,
+    encodeFilename,
+    sanitize_open,
+    timeconvert,
+)
 
 
 class FileDownloader(object):
@@ -194,7 +201,7 @@ class FileDownloader(object):
             if old_filename == new_filename:
                 return
             os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
-        except (IOError, OSError) as err:
+        except (IOError, OSError):
             self.report_error(u'unable to rename file')
 
     def try_utime(self, filename, last_modified_hdr):
@@ -227,8 +234,14 @@ class FileDownloader(object):
         if self.params.get('noprogress', False):
             return
         clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'')
+        if eta is not None:
             eta_str = self.format_eta(eta)
+        else:
+            eta_str = 'Unknown ETA'
+        if percent is not None:
             percent_str = self.format_percent(percent)
+        else:
+            percent_str = 'Unknown %'
         speed_str = self.format_speed(speed)
         if self.params.get('progress_with_newline', False):
             self.to_screen(u'[download] %s of %s at %s ETA %s' %
@@ -251,7 +264,7 @@ class FileDownloader(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 report_unable_to_resume(self):
@@ -267,9 +280,10 @@ class FileDownloader(object):
             self.to_screen(u'\r%s[download] 100%% of %s in %s' %
                 (clear_line, data_len_str, self.format_seconds(tot_time)))
 
-    def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url):
+    def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url, live):
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
+        test = self.params.get('test', False)
 
         # Check for rtmpdump first
         try:
@@ -291,6 +305,10 @@ class FileDownloader(object):
             basic_args += ['--playpath', play_path]
         if tc_url is not None:
             basic_args += ['--tcUrl', url]
+        if test:
+            basic_args += ['--stop', '1']
+        if live:
+            basic_args += ['--live']
         args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)]
         if self.params.get('verbose', False):
             try:
@@ -300,7 +318,7 @@ class FileDownloader(object):
                 shell_quote = repr
             self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(args))
         retval = subprocess.call(args)
-        while retval == 2 or retval == 1:
+        while (retval == 2 or retval == 1) and not test:
             prevsize = os.path.getsize(encodeFilename(tmpfilename))
             self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True)
             time.sleep(5.0) # This seems to be needed
@@ -313,7 +331,7 @@ class FileDownloader(object):
                 self.to_screen(u'\r[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
                 retval = 0
                 break
-        if retval == 0:
+        if retval == 0 or (test and retval == 2):
             fsize = os.path.getsize(encodeFilename(tmpfilename))
             self.to_screen(u'\r[rtmpdump] %s bytes' % fsize)
             self.try_rename(tmpfilename, filename)
@@ -363,15 +381,20 @@ class FileDownloader(object):
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
 
-        args = ['ffmpeg', '-y', '-i', url, '-f', 'mp4', tmpfilename]
-        # Check for ffmpeg first
-        try:
-            subprocess.call(['ffmpeg', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
-        except (OSError, IOError):
-            self.report_error(u'm3u8 download detected but "%s" could not be run' % args[0] )
-            return False
+        args = ['-y', '-i', url, '-f', 'mp4', '-c', 'copy',
+            '-bsf:a', 'aac_adtstoasc', tmpfilename]
 
-        retval = subprocess.call(args)
+        for program in ['avconv', 'ffmpeg']:
+            try:
+                subprocess.call([program, '-version'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+                break
+            except (OSError, IOError):
+                pass
+        else:
+            self.report_error(u'm3u8 download detected but ffmpeg or avconv could not be found')
+        cmd = [program] + args
+
+        retval = subprocess.call(cmd)
         if retval == 0:
             fsize = os.path.getsize(encodeFilename(tmpfilename))
             self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
@@ -408,7 +431,8 @@ class FileDownloader(object):
                                                 info_dict.get('player_url', None),
                                                 info_dict.get('page_url', None),
                                                 info_dict.get('play_path', None),
-                                                info_dict.get('tc_url', None))
+                                                info_dict.get('tc_url', None),
+                                                info_dict.get('rtmp_live', False))
 
         # Attempt to download using mplayer
         if url.startswith('mms') or url.startswith('rtsp'):
@@ -547,8 +571,7 @@ class FileDownloader(object):
             # Progress message
             speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
             if data_len is None:
-                self.report_progress('Unknown %', data_len_str, speed_str, 'Unknown ETA')
-                eta = None
+                eta = percent = None
             else:
                 percent = self.calc_percent(byte_counter, data_len)
                 eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)

youtube_dl/PostProcessor.py

@@ -3,7 +3,14 @@ import subprocess
 import sys
 import time
 
-from .utils import *
+
+from .utils import (
+    compat_subprocess_get_DEVNULL,
+    encodeFilename,
+    PostProcessingError,
+    shell_quote,
+    subtitles_filename,
+)
 
 
 class PostProcessor(object):
@@ -82,6 +89,8 @@ class FFmpegPostProcessor(PostProcessor):
                + opts +
                [encodeFilename(self._ffmpeg_filename_argument(out_path))])
 
+        if self._downloader.params.get('verbose', False):
+            self._downloader.to_screen(u'[debug] ffmpeg command line: %s' % shell_quote(cmd))
         p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
         stdout,stderr = p.communicate()
         if p.returncode != 0:
@@ -177,7 +186,8 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
             extension = self._preferredcodec
             more_opts = []
             if self._preferredquality is not None:
-                if int(self._preferredquality) < 10:
+                # The opus codec doesn't support the -aq option
+                if int(self._preferredquality) < 10 and extension != 'opus':
                     more_opts += [self._exes['avconv'] and '-q:a' or '-aq', self._preferredquality]
                 else:
                     more_opts += [self._exes['avconv'] and '-b:a' or '-ab', self._preferredquality + 'k']
@@ -444,8 +454,11 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
         if information['ext'] != u'mp4':
             self._downloader.to_screen(u'[ffmpeg] Subtitles can only be embedded in mp4 files')
             return True, information
-        sub_langs = [key for key in information['subtitles']]
+        if not information.get('subtitles'):
+            self._downloader.to_screen(u'[ffmpeg] There aren\'t any subtitles to embed') 
+            return True, information
 
+        sub_langs = [key for key in information['subtitles']]
         filename = information['filepath']
         input_files = [filename] + [subtitles_filename(filename, lang, self._subformat) for lang in sub_langs]
 
@@ -464,3 +477,35 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
         os.rename(encodeFilename(temp_filename), encodeFilename(filename))
 
         return True, information
+
+
+class FFmpegMetadataPP(FFmpegPostProcessor):
+    def run(self, info):
+        metadata = {}
+        if info.get('title') is not None:
+            metadata['title'] = info['title']
+        if info.get('upload_date') is not None:
+            metadata['date'] = info['upload_date']
+        if info.get('uploader') is not None:
+            metadata['artist'] = info['uploader']
+        elif info.get('uploader_id') is not None:
+            metadata['artist'] = info['uploader_id']
+
+        if not metadata:
+            self._downloader.to_screen(u'[ffmpeg] There isn\'t any metadata to add')
+            return True, info
+
+        filename = info['filepath']
+        ext = os.path.splitext(filename)[1][1:]
+        temp_filename = filename + u'.temp'
+
+        options = ['-c', 'copy']
+        for (name, value) in metadata.items():
+            options.extend(['-metadata', '%s=%s' % (name, value)])
+        options.extend(['-f', ext])
+
+        self._downloader.to_screen(u'[ffmpeg] Adding metadata to \'%s\'' % filename)
+        self.run_ffmpeg(filename, temp_filename, options)
+        os.remove(encodeFilename(filename))
+        os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+        return True, info

youtube_dl/YoutubeDL.py

@@ -3,6 +3,7 @@
 
 from __future__ import absolute_import
 
+import errno
 import io
 import os
 import re
@@ -70,6 +71,7 @@ class YoutubeDL(object):
     logtostderr:       Log messages to stderr instead of stdout.
     writedescription:  Write the video description to a .description file
     writeinfojson:     Write the video description to a .info.json file
+    writeannotations:  Write the video annotations to a .annotations.xml file
     writethumbnail:    Write the thumbnail image to a file
     writesubtitles:    Write the video subtitles to a file
     writeautomaticsub: Write the automatic subtitles to a file
@@ -83,6 +85,12 @@ class YoutubeDL(object):
     skip_download:     Skip the actual download of the video file
     cachedir:          Location of the cache files in the filesystem.
                        None to disable filesystem cache.
+    noplaylist:        Download single video instead of a playlist if in doubt.
+    age_limit:         An integer representing the user's age in years.
+                       Unsuitable videos for the given age are skipped.
+    downloadarchive:   File name of a file where all downloads are recorded.
+                       Videos already present in the file are not downloaded
+                       again.
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
@@ -112,7 +120,7 @@ class YoutubeDL(object):
                 and not params['restrictfilenames']):
             # On Python 3, the Unicode filesystem API will throw errors (#1474)
             self.report_warning(
-                u'Assuming --restrict-filenames isnce file system encoding '
+                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
@@ -226,19 +234,6 @@ class YoutubeDL(object):
         error_message = u'%s %s' % (_msg_header, message)
         self.trouble(error_message, tb)
 
-    def slow_down(self, start_time, byte_counter):
-        """Sleep if the download speed is over the rate limit."""
-        rate_limit = self.params.get('ratelimit', None)
-        if rate_limit is None or byte_counter == 0:
-            return
-        now = time.time()
-        elapsed = now - start_time
-        if elapsed <= 0.0:
-            return
-        speed = float(byte_counter) / elapsed
-        if speed > rate_limit:
-            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
-
     def report_writedescription(self, descfn):
         """ Report that the description file is being written """
         self.to_screen(u'[info] Writing video description to: ' + descfn)
@@ -251,6 +246,10 @@ class YoutubeDL(object):
         """ Report that the metadata file has been written """
         self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn)
 
+    def report_writeannotations(self, annofn):
+        """ Report that the annotations file has been written. """
+        self.to_screen(u'[info] Writing video annotations to: ' + annofn)
+
     def report_file_already_downloaded(self, file_name):
         """Report file has already been fully downloaded."""
         try:
@@ -273,16 +272,18 @@ class YoutubeDL(object):
                 autonumber_size = 5
             autonumber_templ = u'%0' + str(autonumber_size) + u'd'
             template_dict['autonumber'] = autonumber_templ % self._num_downloads
-            if template_dict['playlist_index'] is not None:
+            if template_dict.get('playlist_index') is not None:
                 template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index']
 
             sanitize = lambda k, v: sanitize_filename(
                 u'NA' if v is None else compat_str(v),
                 restricted=self.params.get('restrictfilenames'),
                 is_id=(k == u'id'))
-            template_dict = dict((k, sanitize(k, v)) for k,v in template_dict.items())
+            template_dict = dict((k, sanitize(k, v))
+                                 for k, v in template_dict.items())
 
-            filename = self.params['outtmpl'] % template_dict
+            tmpl = os.path.expanduser(self.params['outtmpl'])
+            filename = tmpl % template_dict
             return filename
         except KeyError as err:
             self.report_error(u'Erroneous output template')
@@ -308,8 +309,21 @@ class YoutubeDL(object):
             dateRange = self.params.get('daterange', DateRange())
             if date not in dateRange:
                 return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
+        age_limit = self.params.get('age_limit')
+        if age_limit is not None:
+            if age_limit < info_dict.get('age_limit', 0):
+                return u'Skipping "' + title + '" because it is age restricted'
+        if self.in_download_archive(info_dict):
+            return (u'%(title)s has already been recorded in archive'
+                    % info_dict)
         return None
 
+    @staticmethod
+    def add_extra_info(info_dict, extra_info):
+        '''Set the keys from extra_info in info dict if they are missing'''
+        for key, value in extra_info.items():
+            info_dict.setdefault(key, value)
+
     def extract_info(self, url, download=True, ie_key=None, extra_info={}):
         '''
         Returns a list with a dictionary for each video we find.
@@ -336,17 +350,17 @@ class YoutubeDL(object):
                     break
                 if isinstance(ie_result, list):
                     # Backwards compatibility: old IE result format
-                    for result in ie_result:
-                        result.update(extra_info)
                     ie_result = {
                         '_type': 'compat_list',
                         'entries': ie_result,
                     }
-                else:
-                    ie_result.update(extra_info)
-                if 'extractor' not in ie_result:
-                    ie_result['extractor'] = ie.IE_NAME
-                return self.process_ie_result(ie_result, download=download)
+                self.add_extra_info(ie_result,
+                    {
+                        'extractor': ie.IE_NAME,
+                        'webpage_url': url,
+                        'extractor_key': ie.ie_key(),
+                    })
+                return self.process_ie_result(ie_result, download, extra_info)
             except ExtractorError as de: # An error we somewhat expected
                 self.report_error(compat_str(de), de.format_traceback())
                 break
@@ -370,14 +384,8 @@ class YoutubeDL(object):
 
         result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
         if result_type == 'video':
-            ie_result.update(extra_info)
-            if 'playlist' not in ie_result:
-                # It isn't part of a playlist
-                ie_result['playlist'] = None
-                ie_result['playlist_index'] = None
-            if download:
-                self.process_info(ie_result)
-            return ie_result
+            self.add_extra_info(ie_result, extra_info)
+            return self.process_video_result(ie_result)
         elif result_type == 'url':
             # We have to add extra_info to the results because it may be
             # contained in a playlist
@@ -386,6 +394,7 @@ 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)
@@ -411,12 +420,10 @@ class YoutubeDL(object):
                 extra = {
                     'playlist': playlist,
                     'playlist_index': i + playliststart,
+                    'extractor': ie_result['extractor'],
+                    'webpage_url': ie_result['webpage_url'],
+                    'extractor_key': ie_result['extractor_key'],
                 }
-                if not 'extractor' in entry:
-                    # We set the extractor, if it's an url it will be set then to
-                    # the new extractor, but if it's already a video we must make
-                    # sure it's present: see issue #877
-                    entry['extractor'] = ie_result['extractor']
                 entry_result = self.process_ie_result(entry,
                                                       download=download,
                                                       extra_info=extra)
@@ -425,16 +432,122 @@ class YoutubeDL(object):
             return ie_result
         elif result_type == 'compat_list':
             def _fixup(r):
-                r.setdefault('extractor', ie_result['extractor'])
+                self.add_extra_info(r,
+                    {
+                        'extractor': ie_result['extractor'],
+                        'webpage_url': ie_result['webpage_url'],
+                        'extractor_key': ie_result['extractor_key'],
+                    })
                 return r
             ie_result['entries'] = [
-                self.process_ie_result(_fixup(r), download=download)
+                self.process_ie_result(_fixup(r), download, extra_info)
                 for r in ie_result['entries']
             ]
             return ie_result
         else:
             raise Exception('Invalid result type: %s' % result_type)
 
+    def select_format(self, format_spec, available_formats):
+        if format_spec == 'best' or format_spec is None:
+            return available_formats[-1]
+        elif format_spec == 'worst':
+            return available_formats[0]
+        else:
+            extensions = [u'mp4', u'flv', u'webm', u'3gp']
+            if format_spec in extensions:
+                filter_f = lambda f: f['ext'] == format_spec
+            else:
+                filter_f = lambda f: f['format_id'] == format_spec
+            matches = list(filter(filter_f, available_formats))
+            if matches:
+                return matches[-1]
+        return None
+
+    def process_video_result(self, info_dict, download=True):
+        assert info_dict.get('_type', 'video') == 'video'
+
+        if 'playlist' not in info_dict:
+            # It isn't part of a playlist
+            info_dict['playlist'] = None
+            info_dict['playlist_index'] = None
+
+        # This extractors handle format selection themselves
+        if info_dict['extractor'] in [u'youtube', u'Youku']:
+            if download:
+                self.process_info(info_dict)
+            return info_dict
+
+        # We now pick which formats have to be downloaded
+        if info_dict.get('formats') is None:
+            # There's only one format available
+            formats = [info_dict]
+        else:
+            formats = info_dict['formats']
+
+        # We check that all the formats have the format and format_id fields
+        for (i, format) in enumerate(formats):
+            if format.get('format_id') is None:
+                format['format_id'] = compat_str(i)
+            if format.get('format') is None:
+                format['format'] = u'{id} - {res}{note}'.format(
+                    id=format['format_id'],
+                    res=self.format_resolution(format),
+                    note=u' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
+                )
+            # Automatically determine file extension if missing
+            if 'ext' not in format:
+                format['ext'] = determine_ext(format['url'])
+
+        if self.params.get('listformats', None):
+            self.list_formats(info_dict)
+            return
+
+        format_limit = self.params.get('format_limit', None)
+        if format_limit:
+            formats = list(takewhile_inclusive(
+                lambda f: f['format_id'] != format_limit, formats
+            ))
+        if self.params.get('prefer_free_formats'):
+            def _free_formats_key(f):
+                try:
+                    ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext'])
+                except ValueError:
+                    ext_ord = -1
+                # We only compare the extension if they have the same height and width
+                return (f.get('height'), f.get('width'), ext_ord)
+            formats = sorted(formats, key=_free_formats_key)
+
+        req_format = self.params.get('format', 'best')
+        if req_format is None:
+            req_format = 'best'
+        formats_to_download = []
+        # The -1 is for supporting YoutubeIE
+        if req_format in ('-1', 'all'):
+            formats_to_download = formats
+        else:
+            # We can accept formats requestd in the format: 34/5/best, we pick
+            # the first that is available, starting from left
+            req_formats = req_format.split('/')
+            for rf in req_formats:
+                selected_format = self.select_format(rf, formats)
+                if selected_format is not None:
+                    formats_to_download = [selected_format]
+                    break
+        if not formats_to_download:
+            raise ExtractorError(u'requested format not available',
+                                 expected=True)
+
+        if download:
+            if len(formats_to_download) > 1:
+                self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
+            for format in formats_to_download:
+                new_info = dict(info_dict)
+                new_info.update(format)
+                self.process_info(new_info)
+        # We update the info dict with the best quality format (backwards compatibility)
+        info_dict.update(formats_to_download[-1])
+        return info_dict
+
     def process_info(self, info_dict):
         """Process a single resolved IE result."""
 
@@ -472,9 +585,9 @@ class YoutubeDL(object):
         if self.params.get('forceurl', False):
             # For RTMP URLs, also include the playpath
             compat_print(info_dict['url'] + info_dict.get('play_path', u''))
-        if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict:
+        if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
             compat_print(info_dict['thumbnail'])
-        if self.params.get('forcedescription', False) and 'description' in info_dict:
+        if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
             compat_print(info_dict['description'])
         if self.params.get('forcefilename', False) and filename is not None:
             compat_print(filename)
@@ -508,6 +621,18 @@ class YoutubeDL(object):
                 self.report_error(u'Cannot write description file ' + descfn)
                 return
 
+        if self.params.get('writeannotations', False):
+            try:
+                annofn = filename + u'.annotations.xml'
+                self.report_writeannotations(annofn)
+                with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
+                    annofile.write(info_dict['annotations'])
+            except (KeyError, TypeError):
+                self.report_warning(u'There are no annotations to write.')
+            except (OSError, IOError):
+                self.report_error(u'Cannot write annotations file: ' + annofn)
+                return
+
         subtitles_are_requested = any([self.params.get('writesubtitles', False),
                                        self.params.get('writeautomaticsub')])
 
@@ -545,11 +670,15 @@ class YoutubeDL(object):
                 thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format
                 self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
                                (info_dict['extractor'], info_dict['id']))
+                try:
                     uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
                     with open(thumb_filename, 'wb') as thumbf:
                         shutil.copyfileobj(uf, thumbf)
                     self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
                         (info_dict['extractor'], info_dict['id'], thumb_filename))
+                except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+                    self.report_warning(u'Unable to download thumbnail "%s": %s' %
+                        (info_dict['thumbnail'], compat_str(err)))
 
         if not self.params.get('skip_download', False):
             if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)):
@@ -573,6 +702,8 @@ class YoutubeDL(object):
                     self.report_error(u'postprocessing: %s' % str(err))
                     return
 
+        self.record_download_archive(info_dict)
+
     def download(self, url_list):
         """Download a given list of URLs."""
         if len(url_list) > 1 and self.fixed_template():
@@ -612,3 +743,61 @@ class YoutubeDL(object):
                 os.remove(encodeFilename(filename))
             except (IOError, OSError):
                 self.report_warning(u'Unable to remove downloaded video file')
+
+    def in_download_archive(self, info_dict):
+        fn = self.params.get('download_archive')
+        if fn is None:
+            return False
+        vid_id = info_dict['extractor'] + u' ' + info_dict['id']
+        try:
+            with locked_file(fn, 'r', encoding='utf-8') as archive_file:
+                for line in archive_file:
+                    if line.strip() == vid_id:
+                        return True
+        except IOError as ioe:
+            if ioe.errno != errno.ENOENT:
+                raise
+        return False
+
+    def record_download_archive(self, info_dict):
+        fn = self.params.get('download_archive')
+        if fn is None:
+            return
+        vid_id = info_dict['extractor'] + u' ' + info_dict['id']
+        with locked_file(fn, 'a', encoding='utf-8') as archive_file:
+            archive_file.write(vid_id + u'\n')
+
+    @staticmethod
+    def format_resolution(format, default='unknown'):
+        if format.get('_resolution') is not None:
+            return format['_resolution']
+        if format.get('height') is not None:
+            if format.get('width') is not None:
+                res = u'%sx%s' % (format['width'], format['height'])
+            else:
+                res = u'%sp' % format['height']
+        else:
+            res = default
+        return res
+
+    def list_formats(self, info_dict):
+        def line(format):
+            return (u'%-20s%-10s%-12s%s' % (
+                format['format_id'],
+                format['ext'],
+                self.format_resolution(format),
+                format.get('format_note', ''),
+                )
+            )
+
+        formats = info_dict.get('formats', [info_dict])
+        formats_s = list(map(line, formats))
+        if len(formats) > 1:
+            formats_s[0] += (' ' if formats[0].get('format_note') else '') + '(worst)'
+            formats_s[-1] += (' ' if formats[-1].get('format_note') else '') + '(best)'
+
+        header_line = line({
+            'format_id': u'format code', 'ext': u'extension',
+            '_resolution': u'resolution', 'format_note': u'note'})
+        self.to_screen(u'[info] Available formats for %s:\n%s\n%s' %
+                       (info_dict['id'], header_line, u"\n".join(formats_s)))

youtube_dl/__init__.py

@@ -31,11 +31,15 @@ __authors__  = (
     'Huarong Huo',
     'Ismael Mejía',
     'Steffan \'Ruirize\' James',
+    'Andras Elso',
+    'Jelle van der Waa',
+    'Marcin Cieślak',
 )
 
 __license__ = 'Public Domain'
 
 import codecs
+import collections
 import getpass
 import optparse
 import os
@@ -45,17 +49,43 @@ import shlex
 import socket
 import subprocess
 import sys
-import warnings
+import traceback
 import platform
 
 
-from .utils import *
+from .utils import (
+    compat_cookiejar,
+    compat_print,
+    compat_str,
+    compat_urllib_request,
+    DateRange,
+    decodeOption,
+    determine_ext,
+    DownloadError,
+    get_cachedir,
+    make_HTTPS_handler,
+    MaxDownloadsReached,
+    platform_name,
+    preferredencoding,
+    SameFileError,
+    std_headers,
+    write_string,
+    YoutubeDLHandler,
+)
 from .update import update_self
 from .version import __version__
-from .FileDownloader import *
+from .FileDownloader import (
+    FileDownloader,
+)
 from .extractor import gen_extractors
 from .YoutubeDL import YoutubeDL
-from .PostProcessor import *
+from .PostProcessor import (
+    FFmpegMetadataPP,
+    FFmpegVideoConvertor,
+    FFmpegExtractAudioPP,
+    FFmpegEmbedSubtitlePP,
+)
+
 
 def parseOpts(overrideArguments=None):
     def _readOptions(filename_bytes):
@@ -105,7 +135,7 @@ def parseOpts(overrideArguments=None):
 
     def _hide_login_info(opts):
         opts = list(opts)
-        for private_opt in ['-p', '--password', '-u', '--username']:
+        for private_opt in ['-p', '--password', '-u', '--username', '--video-password']:
             try:
                 i = opts.index(private_opt)
                 opts[i+1] = '<PRIVATE>'
@@ -151,6 +181,9 @@ def parseOpts(overrideArguments=None):
             action='store_true', dest='update_self', help='update this program to latest version. Make sure that you have sufficient permissions (run with sudo if needed)')
     general.add_option('-i', '--ignore-errors',
             action='store_true', dest='ignoreerrors', help='continue on download errors, for example to to skip unavailable videos in a playlist', default=False)
+    general.add_option('--abort-on-error',
+            action='store_false', dest='ignoreerrors',
+            help='Abort downloading of further videos (in the playlist or the command line) if an error occurs')
     general.add_option('--dump-user-agent',
             action='store_true', dest='dump_user_agent',
             help='display the current browser identification', default=False)
@@ -168,8 +201,8 @@ def parseOpts(overrideArguments=None):
     general.add_option('--proxy', dest='proxy', default=None, help='Use the specified HTTP/HTTPS proxy', metavar='URL')
     general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.')
     general.add_option(
-        '--cache-dir', dest='cachedir', default=u'~/.youtube-dl/cache',
-        help='Location in the filesystem where youtube-dl can store downloaded information permanently. %default by default')
+        '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR',
+        help='Location in the filesystem where youtube-dl can store downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl .')
     general.add_option(
         '--no-cache-dir', action='store_const', const=None, dest='cachedir',
         help='Disable filesystem caching')
@@ -187,6 +220,13 @@ def parseOpts(overrideArguments=None):
     selection.add_option('--date', metavar='DATE', dest='date', help='download only videos uploaded in this date', default=None)
     selection.add_option('--datebefore', metavar='DATE', dest='datebefore', help='download only videos uploaded before this date', default=None)
     selection.add_option('--dateafter', metavar='DATE', dest='dateafter', help='download only videos uploaded after this date', default=None)
+    selection.add_option('--no-playlist', action='store_true', dest='noplaylist', help='download only the currently playing video', default=False)
+    selection.add_option('--age-limit', metavar='YEARS', dest='age_limit',
+                         help='download only videos suitable for the given age',
+                         default=None, type=int)
+    selection.add_option('--download-archive', metavar='FILE',
+                         dest='download_archive',
+                         help='Download only videos not present in the archive file. Record all downloaded videos in it.')
 
 
     authentication.add_option('-u', '--username',
@@ -200,7 +240,7 @@ def parseOpts(overrideArguments=None):
 
 
     video_format.add_option('-f', '--format',
-            action='store', dest='format', metavar='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')
     video_format.add_option('--all-formats',
             action='store_const', dest='format', help='download all available video formats', const='all')
@@ -232,11 +272,11 @@ def parseOpts(overrideArguments=None):
             help='languages of the subtitles to download (optional) separated by commas, use IETF language tags like \'en,pt\'')
 
     downloader.add_option('-r', '--rate-limit',
-            dest='ratelimit', metavar='LIMIT', help='maximum download rate (e.g. 50k or 44.6m)')
+            dest='ratelimit', metavar='LIMIT', help='maximum download rate in bytes per second (e.g. 50K or 4.2M)')
     downloader.add_option('-R', '--retries',
             dest='retries', metavar='RETRIES', help='number of retries (default is %default)', default=10)
     downloader.add_option('--buffer-size',
-            dest='buffersize', metavar='SIZE', help='size of download buffer (e.g. 1024 or 16k) (default is %default)', default="1024")
+            dest='buffersize', metavar='SIZE', help='size of download buffer (e.g. 1024 or 16K) (default is %default)', default="1024")
     downloader.add_option('--no-resize-buffer',
             action='store_true', dest='noresizebuffer',
             help='do not automatically adjust the buffer size. By default, the buffer size is automatically resized from an initial value of SIZE.', default=False)
@@ -278,6 +318,9 @@ def parseOpts(overrideArguments=None):
     verbosity.add_option('--dump-intermediate-pages',
             action='store_true', dest='dump_intermediate_pages', default=False,
             help='print downloaded pages to debug problems(very verbose)')
+    verbosity.add_option('--write-pages',
+            action='store_true', dest='write_pages', default=False,
+            help='Write downloaded pages to files in the current directory')
     verbosity.add_option('--youtube-print-sig-code',
             action='store_true', dest='youtube_print_sig_code', default=False,
             help=optparse.SUPPRESS_HELP)
@@ -297,7 +340,10 @@ def parseOpts(overrideArguments=None):
             help=('output filename template. Use %(title)s to get the title, '
                   '%(uploader)s for the uploader name, %(uploader_id)s for the uploader nickname if different, '
                   '%(autonumber)s to get an automatically incremented number, '
-                  '%(ext)s for the filename extension, %(upload_date)s for the upload date (YYYYMMDD), '
+                  '%(ext)s for the filename extension, '
+                  '%(format)s for the format description (like "22 - 1280x720" or "HD"),'
+                  '%(format_id)s for the unique id of the format (like Youtube\'s itags: "137"),'
+                  '%(upload_date)s for the upload date (YYYYMMDD), '
                   '%(extractor)s for the provider (youtube, metacafe, etc), '
                   '%(id)s for the video id , %(playlist)s for the playlist the video is in, '
                   '%(playlist_index)s for the position in the playlist and %% for a literal percent. '
@@ -305,7 +351,7 @@ def parseOpts(overrideArguments=None):
                   'for example with -o \'/my/downloads/%(uploader)s/%(title)s-%(id)s.%(ext)s\' .'))
     filesystem.add_option('--autonumber-size',
             dest='autonumber_size', metavar='NUMBER',
-            help='Specifies the number of digits in %(autonumber)s when it is present in output filename template or --autonumber option is given')
+            help='Specifies the number of digits in %(autonumber)s when it is present in output filename template or --auto-number option is given')
     filesystem.add_option('--restrict-filenames',
             action='store_true', dest='restrictfilenames',
             help='Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames', default=False)
@@ -314,7 +360,7 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option('-w', '--no-overwrites',
             action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
     filesystem.add_option('-c', '--continue',
-            action='store_true', dest='continue_dl', help='resume partially downloaded files', default=True)
+            action='store_true', dest='continue_dl', help='force resume of partially downloaded files. By default, youtube-dl will resume downloads if possible.', default=True)
     filesystem.add_option('--no-continue',
             action='store_false', dest='continue_dl',
             help='do not resume partially downloaded files (restart from beginning)')
@@ -331,6 +377,9 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option('--write-info-json',
             action='store_true', dest='writeinfojson',
             help='write video metadata to a .info.json file', default=False)
+    filesystem.add_option('--write-annotations',
+            action='store_true', dest='writeannotations',
+            help='write video annotations to a .annotation file', default=False)
     filesystem.add_option('--write-thumbnail',
             action='store_true', dest='writethumbnail',
             help='write thumbnail image to disk', default=False)
@@ -350,6 +399,8 @@ def parseOpts(overrideArguments=None):
             help='do not overwrite post-processed files; the post-processed files are overwritten by default')
     postproc.add_option('--embed-subs', action='store_true', dest='embedsubtitles', default=False,
             help='embed subtitles in the video (only for mp4 videos)')
+    postproc.add_option('--add-metadata', action='store_true', dest='addmetadata', default=False,
+            help='add metadata to the files')
 
 
     parser.add_option_group(general)
@@ -369,8 +420,12 @@ def parseOpts(overrideArguments=None):
     else:
         xdg_config_home = os.environ.get('XDG_CONFIG_HOME')
         if xdg_config_home:
+            userConfFile = os.path.join(xdg_config_home, 'youtube-dl', 'config')
+            if not os.path.isfile(userConfFile):
                 userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
         else:
+            userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl', 'config')
+            if not os.path.isfile(userConfFile):
                 userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
         systemConf = _readOptions('/etc/youtube-dl.conf')
         userConf = _readOptions(userConfFile)
@@ -436,27 +491,7 @@ def _real_main(argv=None):
     all_urls = batchurls + args
     all_urls = [url.strip() for url in all_urls]
 
-    # General configuration
-    cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
-    if opts.proxy is not None:
-        if opts.proxy == '':
-            proxies = {}
-        else:
-            proxies = {'http': opts.proxy, 'https': opts.proxy}
-    else:
-        proxies = compat_urllib_request.getproxies()
-        # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
-        if 'http' in proxies and 'https' not in proxies:
-            proxies['https'] = proxies['http']
-    proxy_handler = compat_urllib_request.ProxyHandler(proxies)
-    https_handler = make_HTTPS_handler(opts)
-    opener = compat_urllib_request.build_opener(https_handler, proxy_handler, cookie_processor, YoutubeDLHandler())
-    # 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(300) # 5 minutes should be enough (famous last words)
+    opener = _setup_opener(jar=jar, opts=opts)
 
     extractors = gen_extractors()
 
@@ -473,6 +508,8 @@ def _real_main(argv=None):
             if not ie._WORKING:
                 continue
             desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
+            if desc is False:
+                continue
             if hasattr(ie, 'SEARCH_KEY'):
                 _SEARCHES = (u'cute kittens', u'slithering pythons', u'falling cat', u'angry poodle', u'purple fish', u'running tortoise')
                 _COUNTS = (u'', u'5', u'10', u'all')
@@ -599,11 +636,13 @@ def _real_main(argv=None):
         'progress_with_newline': opts.progress_with_newline,
         'playliststart': opts.playliststart,
         'playlistend': opts.playlistend,
+        'noplaylist': opts.noplaylist,
         'logtostderr': opts.outtmpl == '-',
         'consoletitle': opts.consoletitle,
         'nopart': opts.nopart,
         'updatetime': opts.updatetime,
         'writedescription': opts.writedescription,
+        'writeannotations': opts.writeannotations,
         'writeinfojson': opts.writeinfojson,
         'writethumbnail': opts.writethumbnail,
         'writesubtitles': opts.writesubtitles,
@@ -618,6 +657,7 @@ def _real_main(argv=None):
         'prefer_free_formats': opts.prefer_free_formats,
         'verbose': opts.verbose,
         'dump_intermediate_pages': opts.dump_intermediate_pages,
+        'write_pages': opts.write_pages,
         'test': opts.test,
         'keepvideo': opts.keepvideo,
         'min_filesize': opts.min_filesize,
@@ -625,6 +665,8 @@ def _real_main(argv=None):
         'daterange': date,
         'cachedir': opts.cachedir,
         'youtube_print_sig_code': opts.youtube_print_sig_code,
+        'age_limit': opts.age_limit,
+        'download_archive': opts.download_archive,
         })
 
     if opts.verbose:
@@ -644,11 +686,19 @@ def _real_main(argv=None):
             except:
                 pass
         write_string(u'[debug] Python version %s - %s' %(platform.python_version(), platform_name()) + u'\n')
-        write_string(u'[debug] Proxy map: ' + str(proxy_handler.proxies) + 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:
@@ -658,7 +708,7 @@ def _real_main(argv=None):
 
     # Update version
     if opts.update_self:
-        update_self(ydl.to_screen, opts.verbose, sys.argv[0])
+        update_self(ydl.to_screen, opts.verbose)
 
     # Maybe do nothing
     if len(all_urls) < 1:
@@ -677,11 +727,42 @@ def _real_main(argv=None):
     if opts.cookiefile is not None:
         try:
             jar.save()
-        except (IOError, OSError) as err:
+        except (IOError, OSError):
             sys.exit(u'ERROR: unable to save cookie jar')
 
     sys.exit(retcode)
 
+
+def _setup_opener(jar=None, opts=None, timeout=300):
+    if opts is None:
+        FakeOptions = collections.namedtuple(
+            'FakeOptions', ['proxy', 'no_check_certificate'])
+        opts = FakeOptions(proxy=None, no_check_certificate=False)
+
+    cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
+    if opts.proxy is not None:
+        if opts.proxy == '':
+            proxies = {}
+        else:
+            proxies = {'http': opts.proxy, 'https': opts.proxy}
+    else:
+        proxies = compat_urllib_request.getproxies()
+        # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
+        if 'http' in proxies and 'https' not in proxies:
+            proxies['https'] = proxies['http']
+    proxy_handler = compat_urllib_request.ProxyHandler(proxies)
+    https_handler = make_HTTPS_handler(opts)
+    opener = compat_urllib_request.build_opener(
+        https_handler, proxy_handler, cookie_processor, YoutubeDLHandler())
+    # Delete the default user-agent header, which would otherwise apply in
+    # cases where our custom HTTP handler doesn't come into play
+    # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
+    opener.addheaders = []
+    compat_urllib_request.install_opener(opener)
+    socket.setdefaulttimeout(timeout)
+    return opener
+
+
 def main(argv=None):
     try:
         _real_main(argv)

youtube_dl/extractor/__init__.py

@@ -2,8 +2,14 @@ from .appletrailers import AppleTrailersIE
 from .addanime import AddAnimeIE
 from .archiveorg import ArchiveOrgIE
 from .ard import ARDIE
-from .arte import ArteTvIE
+from .arte import (
+    ArteTvIE,
+    ArteTVPlus7IE,
+    ArteTVCreativeIE,
+    ArteTVFutureIE,
+)
 from .auengine import AUEngineIE
+from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE
 from .bliptv import BlipTVIE, BlipTVUserIE
 from .bloomberg import BloombergIE
@@ -12,6 +18,7 @@ from .brightcove import BrightcoveIE
 from .c56 import C56IE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
+from .cinemassacre import CinemassacreIE
 from .cnn import CNNIE
 from .collegehumor import CollegeHumorIE
 from .comedycentral import ComedyCentralIE
@@ -31,9 +38,12 @@ from .defense import DefenseGouvFrIE
 from .ebaumsworld import EbaumsWorldIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
+from .eitb import EitbIE
 from .escapist import EscapistIE
 from .exfm import ExfmIE
+from .extremetube import ExtremeTubeIE
 from .facebook import FacebookIE
+from .faz import FazIE
 from .fktv import (
     FKTVIE,
     FKTVPosteckeIE,
@@ -47,6 +57,7 @@ from .francetv import (
 )
 from .freesound import FreesoundIE
 from .funnyordie import FunnyOrDieIE
+from .gamekings import GamekingsIE
 from .gamespot import GameSpotIE
 from .gametrailers import GametrailersIE
 from .generic import GenericIE
@@ -60,10 +71,12 @@ from .ign import IGNIE, OneUPIE
 from .ina import InaIE
 from .infoq import InfoQIE
 from .instagram import InstagramIE
+from .internetvideoarchive import InternetVideoArchiveIE
 from .jeuxvideo import JeuxVideoIE
 from .jukebox import JukeboxIE
 from .justintv import JustinTVIE
 from .kankan import KankanIE
+from .keezmovies import KeezMoviesIE
 from .kickstarter import KickStarterIE
 from .keek import KeekIE
 from .liveleak import LiveLeakIE
@@ -72,44 +85,57 @@ from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mit import TechTVMITIE, MITIE
 from .mixcloud import MixcloudIE
+from .mofosex import MofosexIE
 from .mtv import MTVIE
 from .muzu import MuzuTVIE
+from .myspace import MySpaceIE
 from .myspass import MySpassIE
 from .myvideo import MyVideoIE
 from .naver import NaverIE
 from .nba import NBAIE
 from .nbc import NBCNewsIE
 from .newgrounds import NewgroundsIE
+from .nhl import NHLIE, NHLVideocenterIE
+from .nowvideo import NowVideoIE
 from .ooyala import OoyalaIE
 from .orf import ORFIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
+from .pornhub import PornHubIE
 from .pornotube import PornotubeIE
 from .rbmaradio import RBMARadioIE
 from .redtube import RedTubeIE
 from .ringtv import RingTVIE
 from .ro220 import Ro220IE
+from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtlnow import RTLnowIE
+from .rutube import RutubeIE
 from .sina import SinaIE
 from .slashdot import SlashdotIE
 from .slideshare import SlideshareIE
 from .sohu import SohuIE
 from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE
 from .southparkstudios import SouthParkStudiosIE
+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 .sztvhu import SztvHuIE
 from .teamcoco import TeamcocoIE
+from .techtalks import TechTalksIE
 from .ted import TEDIE
 from .tf1 import TF1IE
 from .thisav import ThisAVIE
 from .traileraddict import TrailerAddictIE
 from .trilulilu import TriluliluIE
+from .tube8 import Tube8IE
 from .tudou import TudouIE
 from .tumblr import TumblrIE
 from .tutv import TutvIE
+from .tvp import TvpIE
 from .unistra import UnistraIE
 from .ustream import UstreamIE, UstreamChannelIE
 from .vbox7 import Vbox7IE
@@ -117,16 +143,22 @@ from .veehd import VeeHDIE
 from .veoh import VeohIE
 from .vevo import VevoIE
 from .vice import ViceIE
+from .viddler import ViddlerIE
+from .videodetective import VideoDetectiveIE
 from .videofyme import VideofyMeIE
+from .videopremium import VideoPremiumIE
 from .vimeo import VimeoIE, VimeoChannelIE
 from .vine import VineIE
+from .vk import VKIE
 from .wat import WatIE
+from .websurg import WeBSurgIE
 from .weibo import WeiboIE
 from .wimp import WimpIE
 from .worldstarhiphop import WorldStarHipHopIE
 from .xhamster import XHamsterIE
 from .xnxx import XNXXIE
 from .xvideos import XVideosIE
+from .xtube import XTubeIE
 from .yahoo import YahooIE, YahooSearchIE
 from .youjizz import YouJizzIE
 from .youku import YoukuIE
@@ -135,11 +167,13 @@ from .youtube import (
     YoutubeIE,
     YoutubePlaylistIE,
     YoutubeSearchIE,
+    YoutubeSearchDateIE,
     YoutubeUserIE,
     YoutubeChannelIE,
     YoutubeShowIE,
     YoutubeSubscriptionsIE,
     YoutubeRecommendedIE,
+    YoutubeTruncatedURLIE,
     YoutubeWatchLaterIE,
     YoutubeFavouritesIE,
 )

youtube_dl/extractor/addanime.py

@@ -17,8 +17,8 @@ class AddAnimeIE(InfoExtractor):
     IE_NAME = u'AddAnime'
     _TEST = {
         u'url': u'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9',
-        u'file': u'24MR3YO5SAS9.flv',
-        u'md5': u'1036a0e0cd307b95bd8a8c3a5c8cfaf1',
+        u'file': u'24MR3YO5SAS9.mp4',
+        u'md5': u'72954ea10bc979ab5e2eb288b21425a0',
         u'info_dict': {
             u"description": u"One Piece 606",
             u"title": u"One Piece 606"
@@ -31,7 +31,8 @@ class AddAnimeIE(InfoExtractor):
             video_id = mobj.group('video_id')
             webpage = self._download_webpage(url, video_id)
         except ExtractorError as ee:
-            if not isinstance(ee.cause, compat_HTTPError):
+            if not isinstance(ee.cause, compat_HTTPError) or \
+               ee.cause.code != 503:
                 raise
 
             redir_webpage = ee.cause.read().decode('utf-8')
@@ -60,16 +61,26 @@ class AddAnimeIE(InfoExtractor):
                 note=u'Confirming after redirect')
             webpage = self._download_webpage(url, video_id)
 
-        video_url = self._search_regex(r"var normal_video_file = '(.*?)';",
-                                       webpage, u'video file URL')
+        formats = []
+        for format_id in ('normal', 'hq'):
+            rex = r"var %s_video_file = '(.*?)';" % re.escape(format_id)
+            video_url = self._search_regex(rex, webpage, u'video file URLx',
+                                           fatal=False)
+            if not video_url:
+                continue
+            formats.append({
+                'format_id': format_id,
+                'url': video_url,
+            })
+        if not formats:
+            raise ExtractorError(u'Cannot find any video format!')
         video_title = self._og_search_title(webpage)
         video_description = self._og_search_description(webpage)
 
         return {
             '_type': 'video',
             'id':  video_id,
-            'url': video_url,
-            'ext': 'flv',
+            'formats': formats,
             'title': video_title,
             'description': video_description
         }

youtube_dl/extractor/appletrailers.py

@@ -1,8 +1,10 @@
 import re
 import xml.etree.ElementTree
+import json
 
 from .common import InfoExtractor
 from ..utils import (
+    compat_urlparse,
     determine_ext,
 )
 
@@ -14,10 +16,9 @@ class AppleTrailersIE(InfoExtractor):
         u"playlist": [
             {
                 u"file": u"manofsteel-trailer4.mov",
-                u"md5": u"11874af099d480cc09e103b189805d5f",
+                u"md5": u"d97a8e575432dbcb81b7c3acb741f8a8",
                 u"info_dict": {
                     u"duration": 111,
-                    u"thumbnail": u"http://trailers.apple.com/trailers/wb/manofsteel/images/thumbnail_11624.jpg",
                     u"title": u"Trailer 4",
                     u"upload_date": u"20130523",
                     u"uploader_id": u"wb",
@@ -25,10 +26,9 @@ class AppleTrailersIE(InfoExtractor):
             },
             {
                 u"file": u"manofsteel-trailer3.mov",
-                u"md5": u"07a0a262aae5afe68120eed61137ab34",
+                u"md5": u"b8017b7131b721fb4e8d6f49e1df908c",
                 u"info_dict": {
                     u"duration": 182,
-                    u"thumbnail": u"http://trailers.apple.com/trailers/wb/manofsteel/images/thumbnail_10793.jpg",
                     u"title": u"Trailer 3",
                     u"upload_date": u"20130417",
                     u"uploader_id": u"wb",
@@ -36,10 +36,9 @@ class AppleTrailersIE(InfoExtractor):
             },
             {
                 u"file": u"manofsteel-trailer.mov",
-                u"md5": u"e401fde0813008e3307e54b6f384cff1",
+                u"md5": u"d0f1e1150989b9924679b441f3404d48",
                 u"info_dict": {
                     u"duration": 148,
-                    u"thumbnail": u"http://trailers.apple.com/trailers/wb/manofsteel/images/thumbnail_8703.jpg",
                     u"title": u"Trailer",
                     u"upload_date": u"20121212",
                     u"uploader_id": u"wb",
@@ -47,10 +46,9 @@ class AppleTrailersIE(InfoExtractor):
             },
             {
                 u"file": u"manofsteel-teaser.mov",
-                u"md5": u"76b392f2ae9e7c98b22913c10a639c97",
+                u"md5": u"5fe08795b943eb2e757fa95cb6def1cb",
                 u"info_dict": {
                     u"duration": 93,
-                    u"thumbnail": u"http://trailers.apple.com/trailers/wb/manofsteel/images/thumbnail_6899.jpg",
                     u"title": u"Teaser",
                     u"upload_date": u"20120721",
                     u"uploader_id": u"wb",
@@ -59,87 +57,61 @@ class AppleTrailersIE(InfoExtractor):
         ]
     }
 
+    _JSON_RE = r'iTunes.playURL\((.*?)\);'
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         movie = mobj.group('movie')
         uploader_id = mobj.group('company')
 
-        playlist_url = url.partition(u'?')[0] + u'/includes/playlists/web.inc'
+        playlist_url = compat_urlparse.urljoin(url, u'includes/playlists/itunes.inc')
         playlist_snippet = self._download_webpage(playlist_url, movie)
-        playlist_cleaned = re.sub(r'(?s)<script>.*?</script>', u'', playlist_snippet)
+        playlist_cleaned = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', playlist_snippet)
+        playlist_cleaned = re.sub(r'<img ([^<]*?)>', r'<img \1/>', playlist_cleaned)
+        # The ' in the onClick attributes are not escaped, it couldn't be parsed
+        # with xml.etree.ElementTree.fromstring
+        # like: http://trailers.apple.com/trailers/wb/gravity/
+        def _clean_json(m):
+            return u'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
+        playlist_cleaned = re.sub(self._JSON_RE, _clean_json, playlist_cleaned)
         playlist_html = u'<html>' + playlist_cleaned + u'</html>'
 
-        size_cache = {}
-
         doc = xml.etree.ElementTree.fromstring(playlist_html)
         playlist = []
         for li in doc.findall('./div/ul/li'):
-            title = li.find('.//h3').text
+            on_click = li.find('.//a').attrib['onClick']
+            trailer_info_json = self._search_regex(self._JSON_RE,
+                on_click, u'trailer info')
+            trailer_info = json.loads(trailer_info_json)
+            title = trailer_info['title']
             video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
             thumbnail = li.find('.//img').attrib['src']
+            upload_date = trailer_info['posted'].replace('-', '')
 
-            date_el = li.find('.//p')
-            upload_date = None
-            m = re.search(r':\s?(?P<month>[0-9]{2})/(?P<day>[0-9]{2})/(?P<year>[0-9]{2})', date_el.text)
-            if m:
-                upload_date = u'20' + m.group('year') + m.group('month') + m.group('day')
-            runtime_el = date_el.find('./br')
-            m = re.search(r':\s?(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime_el.tail)
+            runtime = trailer_info['runtime']
+            m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)
             duration = None
             if m:
                 duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
 
+            first_url = trailer_info['url']
+            trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
+            settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
+            settings_json = self._download_webpage(settings_json_url, trailer_id, u'Downloading settings json')
+            settings = json.loads(settings_json)
+
             formats = []
-            for formats_el in li.findall('.//a'):
-                if formats_el.attrib['class'] != 'OverlayPanel':
-                    continue
-                target = formats_el.attrib['target']
-
-                format_code = formats_el.text
-                if 'Automatic' in format_code:
-                    continue
-
-                size_q = formats_el.attrib['href']
-                size_id = size_q.rpartition('#videos-')[2]
-                if size_id not in size_cache:
-                    size_url = url + size_q
-                    sizepage_html = self._download_webpage(
-                        size_url, movie,
-                        note=u'Downloading size info %s' % size_id,
-                        errnote=u'Error while downloading size info %s' % size_id,
-                    )
-                    _doc = xml.etree.ElementTree.fromstring(sizepage_html)
-                    size_cache[size_id] = _doc
-
-                sizepage_doc = size_cache[size_id]
-                links = sizepage_doc.findall('.//{http://www.w3.org/1999/xhtml}ul/{http://www.w3.org/1999/xhtml}li/{http://www.w3.org/1999/xhtml}a')
-                for vid_a in links:
-                    href = vid_a.get('href')
-                    if not href.endswith(target):
-                        continue
-                    detail_q = href.partition('#')[0]
-                    detail_url = url + '/' + detail_q
-
-                    m = re.match(r'includes/(?P<detail_id>[^/]+)/', detail_q)
-                    detail_id = m.group('detail_id')
-
-                    detail_html = self._download_webpage(
-                        detail_url, movie,
-                        note=u'Downloading detail %s %s' % (detail_id, size_id),
-                        errnote=u'Error while downloading detail %s %s' % (detail_id, size_id)
-                    )
-                    detail_doc = xml.etree.ElementTree.fromstring(detail_html)
-                    movie_link_el = detail_doc.find('.//{http://www.w3.org/1999/xhtml}a')
-                    assert movie_link_el.get('class') == 'movieLink'
-                    movie_link = movie_link_el.get('href').partition('?')[0].replace('_', '_h')
-                    ext = determine_ext(movie_link)
-                    assert ext == 'mov'
-
+            for format in settings['metadata']['sizes']:
+                # The src is a file pointing to the real video file
+                format_url = re.sub(r'_(\d*p.mov)', r'_h\1', format['src'])
                 formats.append({
-                        'format': format_code,
-                        'ext': ext,
-                        'url': movie_link,
+                    'url': format_url,
+                    'ext': determine_ext(format_url),
+                    'format': format['type'],
+                    'width': format['width'],
+                    'height': int(format['height']),
                 })
+            formats = sorted(formats, key=lambda f: (f['height'], f['width']))
 
             info = {
                 '_type': 'video',

youtube_dl/extractor/arte.py

@@ -1,3 +1,4 @@
+# encoding: utf-8
 import re
 import json
 import xml.etree.ElementTree
@@ -7,15 +8,16 @@ from ..utils import (
     ExtractorError,
     find_xpath_attr,
     unified_strdate,
+    determine_ext,
+    get_element_by_id,
+    compat_str,
 )
 
+# There are different sources of video in arte.tv, the extraction process 
+# is different for each one. The videos usually expire in 7 days, so we can't
+# add tests.
+
 class ArteTvIE(InfoExtractor):
-    """
-    There are two sources of video in arte.tv: videos.arte.tv and
-    www.arte.tv/guide, the extraction process is different for each one.
-    The videos expire in 7 days, so we can't add tests.
-    """
-    _EMISSION_URL = r'(?:http://)?www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
     _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?P<lang>fr|de)/.*-(?P<id>.*?).html'
     _LIVEWEB_URL = r'(?:http://)?liveweb.arte.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)'
     _LIVE_URL = r'index-[0-9]+\.html$'
@@ -24,7 +26,7 @@ class ArteTvIE(InfoExtractor):
 
     @classmethod
     def suitable(cls, url):
-        return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL, cls._LIVEWEB_URL))
+        return any(re.match(regex, url) for regex in (cls._VIDEOS_URL, cls._LIVEWEB_URL))
 
     # TODO implement Live Stream
     # from ..utils import compat_urllib_parse
@@ -55,14 +57,6 @@ class ArteTvIE(InfoExtractor):
     #     video_url = u'%s/%s' % (info.get('url'), info.get('path'))
 
     def _real_extract(self, url):
-        mobj = re.match(self._EMISSION_URL, url)
-        if mobj is not None:
-            lang = mobj.group('lang')
-            # This is not a real id, it can be for example AJT for the news
-            # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
-            video_id = mobj.group('id')
-            return self._extract_emission(url, video_id, lang)
-
         mobj = re.match(self._VIDEOS_URL, url)
         if mobj is not None:
             id = mobj.group('id')
@@ -80,49 +74,6 @@ class ArteTvIE(InfoExtractor):
             # self.extractLiveStream(url)
             # return
 
-    def _extract_emission(self, url, video_id, lang):
-        """Extract from www.arte.tv/guide"""
-        webpage = self._download_webpage(url, video_id)
-        json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
-
-        json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
-        self.report_extraction(video_id)
-        info = json.loads(json_info)
-        player_info = info['videoJsonPlayer']
-
-        info_dict = {'id': player_info['VID'],
-                     'title': player_info['VTI'],
-                     'description': player_info.get('VDE'),
-                     'upload_date': unified_strdate(player_info['VDA'].split(' ')[0]),
-                     'thumbnail': player_info['programImage'],
-                     'ext': 'flv',
-                     }
-
-        formats = player_info['VSR'].values()
-        def _match_lang(f):
-            # Return true if that format is in the language of the url
-            if lang == 'fr':
-                l = 'F'
-            elif lang == 'de':
-                l = 'A'
-            regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
-            return any(re.match(r, f['versionCode']) for r in regexes)
-        # Some formats may not be in the same language as the url
-        formats = filter(_match_lang, formats)
-        # We order the formats by quality
-        formats = sorted(formats, key=lambda f: int(f['height']))
-        # Prefer videos without subtitles in the same language
-        formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f['versionCode']) is None)
-        # Pick the best quality
-        format_info = formats[-1]
-        if format_info['mediaType'] == u'rtmp':
-            info_dict['url'] = format_info['streamer']
-            info_dict['play_path'] = 'mp4:' + format_info['url']
-        else:
-            info_dict['url'] = format_info['url']
-
-        return info_dict
-
     def _extract_video(self, url, video_id, lang):
         """Extract from videos.arte.tv"""
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
@@ -172,3 +123,140 @@ class ArteTvIE(InfoExtractor):
                 'ext': 'flv',
                 'thumbnail': self._og_search_thumbnail(webpage),
                 }
+
+
+class ArteTVPlus7IE(InfoExtractor):
+    IE_NAME = u'arte.tv:+7'
+    _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
+
+    @classmethod
+    def _extract_url_info(cls, url):
+        mobj = re.match(cls._VALID_URL, url)
+        lang = mobj.group('lang')
+        # This is not a real id, it can be for example AJT for the news
+        # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
+        video_id = mobj.group('id')
+        return video_id, lang
+
+    def _real_extract(self, url):
+        video_id, lang = self._extract_url_info(url)
+        webpage = self._download_webpage(url, video_id)
+        return self._extract_from_webpage(webpage, video_id, lang)
+
+    def _extract_from_webpage(self, webpage, video_id, lang):
+        json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
+
+        json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
+        self.report_extraction(video_id)
+        info = json.loads(json_info)
+        player_info = info['videoJsonPlayer']
+
+        info_dict = {
+            'id': player_info['VID'],
+            'title': player_info['VTI'],
+            'description': player_info.get('VDE'),
+            'upload_date': unified_strdate(player_info.get('VDA', '').split(' ')[0]),
+            'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
+        }
+
+        all_formats = player_info['VSR'].values()
+        # Some formats use the m3u8 protocol
+        all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats))
+        def _match_lang(f):
+            if f.get('versionCode') is None:
+                return True
+            # Return true if that format is in the language of the url
+            if lang == 'fr':
+                l = 'F'
+            elif lang == 'de':
+                l = 'A'
+            regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
+            return any(re.match(r, f['versionCode']) for r in regexes)
+        # Some formats may not be in the same language as the url
+        formats = filter(_match_lang, all_formats)
+        formats = list(formats) # in python3 filter returns an iterator
+        if not formats:
+            # Some videos are only available in the 'Originalversion'
+            # they aren't tagged as being in French or German
+            if all(f['versionCode'] == 'VO' for f in all_formats):
+                formats = all_formats
+            else:
+                raise ExtractorError(u'The formats list is empty')
+
+        if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
+            def sort_key(f):
+                return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
+        else:
+            def sort_key(f):
+                return (
+                    # Sort first by quality
+                    int(f.get('height',-1)),
+                    int(f.get('bitrate',-1)),
+                    # The original version with subtitles has lower relevance
+                    re.match(r'VO-ST(F|A)', f.get('versionCode', '')) is None,
+                    # The version with sourds/mal subtitles has also lower relevance
+                    re.match(r'VO?(F|A)-STM\1', f.get('versionCode', '')) is None,
+                )
+        formats = sorted(formats, key=sort_key)
+        def _format(format_info):
+            quality = ''
+            height = format_info.get('height')
+            if height is not None:
+                quality = compat_str(height)
+            bitrate = format_info.get('bitrate')
+            if bitrate is not None:
+                quality += '-%d' % bitrate
+            if format_info.get('versionCode') is not None:
+                format_id = u'%s-%s' % (quality, format_info['versionCode'])
+            else:
+                format_id = quality
+            info = {
+                'format_id': format_id,
+                'format_note': format_info.get('versionLibelle'),
+                'width': format_info.get('width'),
+                'height': height,
+            }
+            if format_info['mediaType'] == u'rtmp':
+                info['url'] = format_info['streamer']
+                info['play_path'] = 'mp4:' + format_info['url']
+                info['ext'] = 'flv'
+            else:
+                info['url'] = format_info['url']
+                info['ext'] = determine_ext(info['url'])
+            return info
+        info_dict['formats'] = [_format(f) for f in formats]
+
+        return info_dict
+
+
+# It also uses the arte_vp_url url from the webpage to extract the information
+class ArteTVCreativeIE(ArteTVPlus7IE):
+    IE_NAME = u'arte.tv:creative'
+    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/magazine?/(?P<id>.+)'
+
+    _TEST = {
+        u'url': u'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
+        u'file': u'050489-002.mp4',
+        u'info_dict': {
+            u'title': u'Agentur Amateur / Agence Amateur #2 : Corporate Design',
+        },
+    }
+
+
+class ArteTVFutureIE(ArteTVPlus7IE):
+    IE_NAME = u'arte.tv:future'
+    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
+        u'file': u'050940-003.mp4',
+        u'info_dict': {
+            u'title': u'Les champignons au secours de la planète',
+        },
+    }
+
+    def _real_extract(self, url):
+        anchor_id, lang = self._extract_url_info(url)
+        webpage = self._download_webpage(url, anchor_id)
+        row = get_element_by_id(anchor_id, webpage)
+        return self._extract_from_webpage(row, anchor_id, lang)

youtube_dl/extractor/bambuser.py

@@ -0,0 +1,80 @@
+import re
+import json
+import itertools
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+)
+
+
+class BambuserIE(InfoExtractor):
+    IE_NAME = u'bambuser'
+    _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
+    _API_KEY = '005f64509e19a868399060af746a00aa'
+
+    _TEST = {
+        u'url': u'http://bambuser.com/v/4050584',
+        u'md5': u'fba8f7693e48fd4e8641b3fd5539a641',
+        u'info_dict': {
+            u'id': u'4050584',
+            u'ext': u'flv',
+            u'title': u'Education engineering days - lightning talks',
+            u'duration': 3741,
+            u'uploader': u'pixelversity',
+            u'uploader_id': u'344706',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        info_url = ('http://player-c.api.bambuser.com/getVideo.json?'
+            '&api_key=%s&vid=%s' % (self._API_KEY, video_id))
+        info_json = self._download_webpage(info_url, video_id)
+        info = json.loads(info_json)['result']
+
+        return {
+            'id': video_id,
+            'title': info['title'],
+            'url': info['url'],
+            'thumbnail': info.get('preview'),
+            'duration': int(info['length']),
+            'view_count': int(info['views_total']),
+            'uploader': info['username'],
+            'uploader_id': info['uid'],
+        }
+
+
+class BambuserChannelIE(InfoExtractor):
+    IE_NAME = u'bambuser:channel'
+    _VALID_URL = r'http://bambuser.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
+    # The maximum number we can get with each request
+    _STEP = 50
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        user = mobj.group('user')
+        urls = []
+        last_id = ''
+        for i in itertools.count(1):
+            req_url = ('http://bambuser.com/xhr-api/index.php?username={user}'
+                '&sort=created&access_mode=0%2C1%2C2&limit={count}'
+                '&method=broadcast&format=json&vid_older_than={last}'
+                ).format(user=user, count=self._STEP, last=last_id)
+            req = compat_urllib_request.Request(req_url)
+            # Without setting this header, we wouldn't get any result
+            req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
+            info_json = self._download_webpage(req, user,
+                u'Downloading page %d' % i)
+            results = json.loads(info_json)['result']
+            if len(results) == 0:
+                break
+            last_id = results[-1]['vid']
+            urls.extend(self.url_result(v['page'], 'Bambuser') for v in results)
+
+        return {
+            '_type': 'playlist',
+            'title': user,
+            'entries': urls,
+        }

youtube_dl/extractor/bliptv.py

@@ -115,7 +115,7 @@ class BlipTVIE(InfoExtractor):
                 ext = umobj.group(1)
 
                 info = {
-                    'id': data['item_id'],
+                    'id': compat_str(data['item_id']),
                     'url': video_url,
                     'uploader': data['display_name'],
                     'upload_date': upload_date,

youtube_dl/extractor/brightcove.py

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 import re
 import json
 import xml.etree.ElementTree
@@ -7,15 +9,53 @@ from ..utils import (
     compat_urllib_parse,
     find_xpath_attr,
     compat_urlparse,
+    compat_str,
+    compat_urllib_request,
+
+    ExtractorError,
 )
 
+
 class BrightcoveIE(InfoExtractor):
     _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
     _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'
 
-    # There is a test for Brigtcove in GenericIE, that way we test both the download
-    # and the detection of videos, and we don't have to find an URL that is always valid
+    _TESTS = [
+        {
+            # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
+            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
+            u'file': u'2371591881001.mp4',
+            u'md5': u'8eccab865181d29ec2958f32a6a754f5',
+            u'note': u'Test Brightcove downloads and detection in GenericIE',
+            u'info_dict': {
+                u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
+                u'uploader': u'8TV',
+                u'description': u'md5:a950cc4285c43e44d763d036710cd9cd',
+            }
+        },
+        {
+            # From http://medianetwork.oracle.com/video/player/1785452137001
+            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
+            u'file': u'1785452137001.flv',
+            u'info_dict': {
+                u'title': u'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
+                u'description': u'John Rose speaks at the JVM Language Summit, August 1, 2012.',
+                u'uploader': u'Oracle',
+            },
+        },
+        {
+            # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/
+            u'url': u'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
+            u'info_dict': {
+                u'id': u'2750934548001',
+                u'ext': u'mp4',
+                u'title': u'This Bracelet Acts as a Personal Thermostat',
+                u'description': u'md5:547b78c64f4112766ccf4e151c20b6a0',
+                u'uploader': u'Mashable',
+            },
+        },
+    ]
 
     @classmethod
     def _build_brighcove_url(cls, object_str):
@@ -23,6 +63,13 @@ class BrightcoveIE(InfoExtractor):
         Build a Brightcove url from a xml string containing
         <object class="BrightcoveExperience">{params}</object>
         """
+
+        # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553
+        object_str = re.sub(r'(<param name="[^"]+" value="[^"]+")>',
+                            lambda m: m.group(1) + '/>', object_str)
+        # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
+        object_str = object_str.replace(u'<--', u'<!--')
+
         object_doc = xml.etree.ElementTree.fromstring(object_str)
         assert u'BrightcoveExperience' in object_doc.attrib['class']
         params = {'flashID': object_doc.attrib['id'],
@@ -35,24 +82,48 @@ class BrightcoveIE(InfoExtractor):
         videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer')
         if videoPlayer is not None:
             params['@videoPlayer'] = videoPlayer.attrib['value']
+        linkBase = find_xpath_attr(object_doc, './param', 'name', 'linkBaseURL')
+        if linkBase is not None:
+            params['linkBaseURL'] = linkBase.attrib['value']
         data = compat_urllib_parse.urlencode(params)
         return cls._FEDERATED_URL_TEMPLATE % data
 
+    @classmethod
+    def _extract_brightcove_url(cls, webpage):
+        """Try to extract the brightcove url from the wepbage, returns None
+        if it can't be found
+        """
+        m_brightcove = re.search(
+            r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>',
+            webpage, re.DOTALL)
+        if m_brightcove is not None:
+            return cls._build_brighcove_url(m_brightcove.group())
+        else:
+            return None
+
     def _real_extract(self, url):
+        # Change the 'videoId' and others field to '@videoPlayer'
+        url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url)
+        # Change bckey (used by bcove.me urls) to playerKey
+        url = re.sub(r'(?<=[?&])bckey', 'playerKey', url)
         mobj = re.match(self._VALID_URL, url)
         query_str = mobj.group('query')
         query = compat_urlparse.parse_qs(query_str)
 
         videoPlayer = query.get('@videoPlayer')
         if videoPlayer:
-            return self._get_video_info(videoPlayer[0], query_str)
+            return self._get_video_info(videoPlayer[0], query_str, query)
         else:
             player_key = query['playerKey']
             return self._get_playlist_info(player_key[0])
 
-    def _get_video_info(self, video_id, query):
-        request_url = self._FEDERATED_URL_TEMPLATE % query
-        webpage = self._download_webpage(request_url, video_id)
+    def _get_video_info(self, video_id, query_str, query):
+        request_url = self._FEDERATED_URL_TEMPLATE % query_str
+        req = compat_urllib_request.Request(request_url)
+        linkBase = query.get('linkBaseURL')
+        if linkBase is not None:
+            req.add_header('Referer', linkBase[0])
+        webpage = self._download_webpage(req, video_id)
 
         self.report_extraction(video_id)
         info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
@@ -65,22 +136,36 @@ class BrightcoveIE(InfoExtractor):
         playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
                                                player_key, u'Downloading playlist information')
 
-        playlist_info = json.loads(playlist_info)['videoList']
+        json_data = json.loads(playlist_info)
+        if 'videoList' not in json_data:
+            raise ExtractorError(u'Empty playlist')
+        playlist_info = json_data['videoList']
         videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
 
         return self.playlist_result(videos, playlist_id=playlist_info['id'],
                                     playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
 
     def _extract_video_info(self, video_info):
-        renditions = video_info['renditions']
-        renditions = sorted(renditions, key=lambda r: r['size'])
-        best_format = renditions[-1]
-
-        return {'id': video_info['id'],
+        info = {
+            'id': compat_str(video_info['id']),
             'title': video_info['displayName'],
-                'url': best_format['defaultURL'], 
-                'ext': 'mp4',
             'description': video_info.get('shortDescription'),
             'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
             'uploader': video_info.get('publisherName'),
         }
+
+        renditions = video_info.get('renditions')
+        if renditions:
+            renditions = sorted(renditions, key=lambda r: r['size'])
+            info['formats'] = [{
+                'url': rend['defaultURL'],
+                'height': rend.get('frameHeight'),
+                'width': rend.get('frameWidth'),
+            } for rend in renditions]
+        elif video_info.get('FLVFullLengthURL') is not None:
+            info.update({
+                'url': video_info['FLVFullLengthURL'],
+            })
+        else:
+            raise ExtractorError(u'Unable to extract video url for %s' % info['id'])
+        return info

youtube_dl/extractor/canalc2.py

@@ -6,7 +6,7 @@ from .common import InfoExtractor
 
 class Canalc2IE(InfoExtractor):
     IE_NAME = 'canalc2.tv'
-    _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?idVideo=(\d+)&voir=oui'
+    _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?.*?idVideo=(?P<id>\d+)'
 
     _TEST = {
         u'url': u'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui',
@@ -18,7 +18,9 @@ class Canalc2IE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        video_id = re.match(self._VALID_URL, url).group(1)
+        video_id = re.match(self._VALID_URL, url).group('id')
+        # We need to set the voir field for getting the file name
+        url = 'http://www.canalc2.tv/video.asp?idVideo=%s&voir=oui' % video_id
         webpage = self._download_webpage(url, video_id)
         file_name = self._search_regex(
             r"so\.addVariable\('file','(.*?)'\);",

youtube_dl/extractor/cinemassacre.py

@@ -0,0 +1,90 @@
+# encoding: utf-8
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+)
+
+
+class CinemassacreIE(InfoExtractor):
+    _VALID_URL = r'(?:http://)?(?:www\.)?(?P<url>cinemassacre\.com/(?P<date_Y>[0-9]{4})/(?P<date_m>[0-9]{2})/(?P<date_d>[0-9]{2})/.+?)(?:[/?].*)?'
+    _TESTS = [{
+        u'url': u'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
+        u'file': u'19911.flv',
+        u'info_dict': {
+            u'upload_date': u'20121110',
+            u'title': u'“Angry Video Game Nerd: The Movie” – Trailer',
+            u'description': u'md5:fb87405fcb42a331742a0dce2708560b',
+        },
+        u'params': {
+            # rtmp download
+            u'skip_download': True,
+        },
+    },
+    {
+        u'url': u'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
+        u'file': u'521be8ef82b16.flv',
+        u'info_dict': {
+            u'upload_date': u'20131002',
+            u'title': u'The Mummy’s Hand (1940)',
+        },
+        u'params': {
+            # rtmp download
+            u'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        webpage_url = u'http://' + mobj.group('url')
+        webpage = self._download_webpage(webpage_url, None) # Don't know video id yet
+        video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d')
+        mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?id=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage)
+        if not mobj:
+            raise ExtractorError(u'Can\'t extract embed url and video id')
+        playerdata_url = mobj.group(u'embed_url')
+        video_id = mobj.group(u'video_id')
+
+        video_title = self._html_search_regex(r'<title>(?P<title>.+?)\|',
+            webpage, u'title')
+        video_description = self._html_search_regex(r'<div class="entry-content">(?P<description>.+?)</div>',
+            webpage, u'description', flags=re.DOTALL, fatal=False)
+        if len(video_description) == 0:
+            video_description = None
+
+        playerdata = self._download_webpage(playerdata_url, video_id)
+        url = self._html_search_regex(r'\'streamer\': \'(?P<url>[^\']+)\'', playerdata, u'url')
+
+        sd_file = self._html_search_regex(r'\'file\': \'(?P<sd_file>[^\']+)\'', playerdata, u'sd_file')
+        hd_file = self._html_search_regex(r'\'?file\'?: "(?P<hd_file>[^"]+)"', playerdata, u'hd_file')
+        video_thumbnail = self._html_search_regex(r'\'image\': \'(?P<thumbnail>[^\']+)\'', playerdata, u'thumbnail', fatal=False)
+
+        formats = [
+            {
+                'url': url,
+                'play_path': 'mp4:' + sd_file,
+                'rtmp_live': True, # workaround
+                'ext': 'flv',
+                'format': 'sd',
+                'format_id': 'sd',
+            },
+            {
+                'url': url,
+                'play_path': 'mp4:' + hd_file,
+                'rtmp_live': True, # workaround
+                'ext': 'flv',
+                'format': 'hd',
+                'format_id': 'hd',
+            },
+        ]
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'formats': formats,
+            'description': video_description,
+            'upload_date': video_date,
+            'thumbnail': video_thumbnail,
+        }

youtube_dl/extractor/cnn.py

@@ -6,7 +6,7 @@ from ..utils import determine_ext
 
 
 class CNNIE(InfoExtractor):
-    _VALID_URL = r'''(?x)https?://(edition\.)?cnn\.com/video/(data/.+?|\?)/
+    _VALID_URL = r'''(?x)https?://((edition|www)\.)?cnn\.com/video/(data/.+?|\?)/
         (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn|(?=&)))'''
 
     _TESTS = [{

youtube_dl/extractor/comedycentral.py

@@ -51,12 +51,12 @@ class ComedyCentralIE(InfoExtractor):
         '400': 'mp4',
     }
     _video_dimensions = {
-        '3500': '1280x720',
-        '2200': '960x540',
-        '1700': '768x432',
-        '1200': '640x360',
-        '750': '512x288',
-        '400': '384x216',
+        '3500': (1280, 720),
+        '2200': (960, 540),
+        '1700': (768, 432),
+        '1200': (640, 360),
+        '750': (512, 288),
+        '400': (384, 216),
     }
 
     @classmethod
@@ -64,11 +64,13 @@ class ComedyCentralIE(InfoExtractor):
         """Receives a URL and returns True if suitable for this IE."""
         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
 
-    def _print_formats(self, formats):
-        print('Available formats:')
-        for x in formats:
-            print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'mp4'), self._video_dimensions.get(x, '???')))
-
+    @staticmethod
+    def _transform_rtmp_url(rtmp_video_url):
+        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
+        if not m:
+            raise ExtractorError(u'Cannot transform RTMP url')
+        base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
+        return base + m.group('finalid')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
@@ -155,40 +157,31 @@ class ComedyCentralIE(InfoExtractor):
                 self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
                 continue
 
-            if self._downloader.params.get('listformats', None):
-                self._print_formats([i[0] for i in turls])
-                return
-
-            # For now, just pick the highest bitrate
-            format,rtmp_video_url = turls[-1]
-
-            # Get the format arg from the arg stream
-            req_format = self._downloader.params.get('format', None)
-
-            # Select format if we can find one
-            for f,v in turls:
-                if f == req_format:
-                    format, rtmp_video_url = f, v
-                    break
-
-            m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
-            if not m:
-                raise ExtractorError(u'Cannot transform RTMP url')
-            base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
-            video_url = base + m.group('finalid')
+            formats = []
+            for format, rtmp_video_url in turls:
+                w, h = self._video_dimensions.get(format, (None, None))
+                formats.append({
+                    'url': self._transform_rtmp_url(rtmp_video_url),
+                    'ext': self._video_extensions.get(format, 'mp4'),
+                    'format_id': format,
+                    'height': h,
+                    'width': w,
+                })
 
             effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
             info = {
                 'id': shortMediaId,
-                'url': video_url,
+                'formats': formats,
                 'uploader': showId,
                 'upload_date': officialDate,
                 'title': effTitle,
-                'ext': 'mp4',
-                'format': format,
                 'thumbnail': None,
                 'description': compat_str(officialTitle),
             }
+
+            # TODO: Remove when #980 has been merged
+            info.update(info['formats'][-1])
+
             results.append(info)
 
         return results

youtube_dl/extractor/common.py

@@ -14,6 +14,8 @@ from ..utils import (
     clean_html,
     compiled_regex_type,
     ExtractorError,
+    RegexNotFoundError,
+    sanitize_filename,
     unescapeHTML,
 )
 
@@ -35,6 +37,8 @@ class InfoExtractor(object):
     title:          Video title, unescaped.
     ext:            Video filename extension.
 
+    Instead of url and ext, formats can also specified.
+
     The following fields are optional:
 
     format:         The video format, defaults to ext (used for --get-format)
@@ -52,8 +56,26 @@ class InfoExtractor(object):
     view_count:     How many users have watched the video on the platform.
     urlhandle:      [internal] The urlHandle to be used to download the file,
                     like returned by urllib.request.urlopen
+    age_limit:      Age restriction for the video, as an integer (years)
+    formats:        A list of dictionaries for each format available, it must
+                    be ordered from worst to best quality. Potential fields:
+                    * url       Mandatory. The URL of the video file
+                    * ext       Will be calculated from url if missing
+                    * format    A human-readable description of the format
+                                ("mp4 container with h264/opus").
+                                Calculated from the format_id, width, height.
+                                and format_note fields if missing.
+                    * format_id A short description of the format
+                                ("mp4_h264_opus" or "19")
+                    * format_note Additional info about the format
+                                ("3D" or "DASH video")
+                    * width     Width of the video, if known
+                    * height    Height of the video, if known
+    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)
 
-    The fields should all be Unicode strings.
+    Unless mentioned otherwise, the fields should be Unicode strings.
 
     Subclasses of this one should re-define the _real_initialize() and
     _real_extract() methods and define a _VALID_URL regexp.
@@ -164,6 +186,17 @@ class InfoExtractor(object):
             self.to_screen(u'Dumping request to ' + url)
             dump = base64.b64encode(webpage_bytes).decode('ascii')
             self._downloader.to_screen(dump)
+        if self._downloader.params.get('write_pages', False):
+            try:
+                url = url_or_request.get_full_url()
+            except AttributeError:
+                url = url_or_request
+            raw_filename = ('%s_%s.dump' % (video_id, url))
+            filename = sanitize_filename(raw_filename, restricted=True)
+            self.to_screen(u'Saving request to ' + filename)
+            with open(filename, 'wb') as outf:
+                outf.write(webpage_bytes)
+
         content = webpage_bytes.decode(encoding, 'replace')
         return (content, urlh)
 
@@ -214,7 +247,7 @@ class InfoExtractor(object):
         Perform a regex search on the given string, using a single or a list of
         patterns returning the first matching group.
         In case of failure return a default value or raise a WARNING or a
-        ExtractorError, depending on fatal, specifying the field name.
+        RegexNotFoundError, depending on fatal, specifying the field name.
         """
         if isinstance(pattern, (str, compat_str, compiled_regex_type)):
             mobj = re.search(pattern, string, flags)
@@ -234,7 +267,7 @@ class InfoExtractor(object):
         elif default is not None:
             return default
         elif fatal:
-            raise ExtractorError(u'Unable to extract %s' % _name)
+            raise RegexNotFoundError(u'Unable to extract %s' % _name)
         else:
             self._downloader.report_warning(u'unable to extract %s; '
                 u'please report this issue on http://yt-dl.org/bug' % _name)
@@ -289,6 +322,8 @@ class InfoExtractor(object):
         if name is None:
             name = 'OpenGraph %s' % prop
         escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs)
+        if escaped is None:
+            return None
         return unescapeHTML(escaped)
 
     def _og_search_thumbnail(self, html, **kargs):
@@ -300,10 +335,19 @@ class InfoExtractor(object):
     def _og_search_title(self, html, **kargs):
         return self._og_search_property('title', html, **kargs)
 
-    def _og_search_video_url(self, html, name='video url', **kargs):
-        return self._html_search_regex([self._og_regex('video:secure_url'),
-                                        self._og_regex('video')],
-                                       html, name, **kargs)
+    def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
+        regexes = [self._og_regex('video')]
+        if secure: regexes.insert(0, self._og_regex('video:secure_url'))
+        return self._html_search_regex(regexes, html, name, **kargs)
+
+    def _rta_search(self, html):
+        # See http://www.rtalabel.org/index.php?content=howtofaq#single
+        if re.search(r'(?ix)<meta\s+name="rating"\s+'
+                     r'     content="RTA-5042-1996-1400-1577-RTA"',
+                     html):
+            return 18
+        return 0
+
 
 class SearchInfoExtractor(InfoExtractor):
     """
@@ -342,7 +386,7 @@ class SearchInfoExtractor(InfoExtractor):
 
     def _get_n_results(self, query, n):
         """Get a specified number of results for a query"""
-        raise NotImplementedError("This method must be implemented by sublclasses")
+        raise NotImplementedError("This method must be implemented by subclasses")
 
     @property
     def SEARCH_KEY(self):

youtube_dl/extractor/dailymotion.py

@@ -10,17 +10,36 @@ from ..utils import (
     compat_str,
     get_element_by_attribute,
     get_element_by_id,
+    orderedSet,
 
     ExtractorError,
 )
 
+class DailymotionBaseInfoExtractor(InfoExtractor):
+    @staticmethod
+    def _build_request(url):
+        """Build a request with the family filter disabled"""
+        request = compat_urllib_request.Request(url)
+        request.add_header('Cookie', 'family_filter=off')
+        request.add_header('Cookie', 'ff=off')
+        return request
 
-class DailymotionIE(SubtitlesInfoExtractor):
+class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
     """Information Extractor for Dailymotion"""
 
     _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
     IE_NAME = u'dailymotion'
-    _TEST = {
+
+    _FORMATS = [
+        (u'stream_h264_ld_url', u'ld'),
+        (u'stream_h264_url', u'standard'),
+        (u'stream_h264_hq_url', u'hq'),
+        (u'stream_h264_hd_url', u'hd'),
+        (u'stream_h264_hd1080_url', u'hd180'),
+    ]
+
+    _TESTS = [
+        {
             u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech',
             u'file': u'x33vw9.mp4',
             u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
@@ -28,29 +47,64 @@ class DailymotionIE(SubtitlesInfoExtractor):
                 u"uploader": u"Amphora Alex and Van .", 
                 u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
             }
+        },
+        # Vevo video
+        {
+            u'url': u'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
+            u'file': u'USUV71301934.mp4',
+            u'info_dict': {
+                u'title': u'Roar (Official)',
+                u'uploader': u'Katy Perry',
+                u'upload_date': u'20130905',
+            },
+            u'params': {
+                u'skip_download': True,
+            },
+            u'skip': u'VEVO is only available in some countries',
+        },
+        # age-restricted video
+        {
+            u'url': u'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
+            u'file': u'xyh2zz.mp4',
+            u'md5': u'0d667a7b9cebecc3c89ee93099c4159d',
+            u'info_dict': {
+                u'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
+                u'uploader': 'HotWaves1012',
+                u'age_limit': 18,
             }
 
+        }
+    ]
+
     def _real_extract(self, url):
         # Extract id and simplified title from URL
         mobj = re.match(self._VALID_URL, url)
 
         video_id = mobj.group(1).split('_')[0].split('?')[0]
 
-        video_extension = 'mp4'
         url = 'http://www.dailymotion.com/video/%s' % video_id
 
         # Retrieve video webpage to extract further information
-        request = compat_urllib_request.Request(url)
-        request.add_header('Cookie', 'family_filter=off')
+        request = self._build_request(url)
         webpage = self._download_webpage(request, video_id)
 
         # Extract URL, uploader and title from webpage
         self.report_extraction(video_id)
 
+        # It may just embed a vevo video:
+        m_vevo = re.search(
+            r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
+            webpage)
+        if m_vevo is not None:
+            vevo_id = m_vevo.group('id')
+            self.to_screen(u'Vevo video detected: %s' % vevo_id)
+            return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
+
         video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
                                              # Looking for official user
                                              r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
-                                            webpage, 'video uploader')
+                                            webpage, 'video uploader', fatal=False)
+        age_limit = self._rta_search(webpage)
 
         video_upload_date = None
         mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
@@ -67,37 +121,43 @@ class DailymotionIE(SubtitlesInfoExtractor):
             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
             raise ExtractorError(msg, expected=True)
 
-        # TODO: support choosing qualities
-
-        for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
-                    'stream_h264_hq_url','stream_h264_url',
-                    'stream_h264_ld_url']:
-            if info.get(key):#key in info and info[key]:
-                max_quality = key
-                self.to_screen(u'Using %s' % key)
-                break
+        formats = []
+        for (key, format_id) in self._FORMATS:
+            video_url = info.get(key)
+            if video_url is not None:
+                m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
+                if m_size is not None:
+                    width, height = m_size.group(1), m_size.group(2)
                 else:
+                    width, height = None, None
+                formats.append({
+                    'url': video_url,
+                    'ext': 'mp4',
+                    'format_id': format_id,
+                    'width': width,
+                    'height': height,
+                })
+        if not formats:
             raise ExtractorError(u'Unable to extract video URL')
-        video_url = info[max_quality]
 
         # subtitles
-        video_subtitles = self.extract_subtitles(video_id)
+        video_subtitles = self.extract_subtitles(video_id, webpage)
         if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id)
+            self._list_available_subtitles(video_id, webpage)
             return
 
-        return [{
+        return {
             'id':       video_id,
-            'url':      video_url,
+            'formats': formats,
             'uploader': video_uploader,
             'upload_date':  video_upload_date,
             'title':    self._og_search_title(webpage),
-            'ext':      video_extension,
             'subtitles':    video_subtitles,
-            'thumbnail': info['thumbnail_url']
-        }]
+            'thumbnail': info['thumbnail_url'],
+            'age_limit': age_limit,
+        }
 
-    def _get_available_subtitles(self, video_id):
+    def _get_available_subtitles(self, video_id, webpage):
         try:
             sub_list = self._download_webpage(
                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
@@ -113,7 +173,7 @@ class DailymotionIE(SubtitlesInfoExtractor):
         return {}
 
 
-class DailymotionPlaylistIE(InfoExtractor):
+class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
     IE_NAME = u'dailymotion:playlist'
     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
     _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
@@ -122,16 +182,17 @@ class DailymotionPlaylistIE(InfoExtractor):
     def _extract_entries(self, id):
         video_ids = []
         for pagenum in itertools.count(1):
-            webpage = self._download_webpage(self._PAGE_TEMPLATE % (id, pagenum),
+            request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
+            webpage = self._download_webpage(request,
                                              id, u'Downloading page %s' % pagenum)
 
             playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
-            video_ids.extend(re.findall(r'data-id="(.+?)" data-ext-id', playlist_el))
+            video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el))
 
             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
                 break
         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
-                   for video_id in video_ids]
+                   for video_id in orderedSet(video_ids)]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/eighttracks.py

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

youtube_dl/extractor/eitb.py

@@ -0,0 +1,37 @@
+# encoding: utf-8
+import re
+
+from .common import InfoExtractor
+from .brightcove import BrightcoveIE
+from ..utils import ExtractorError
+
+
+class EitbIE(InfoExtractor):
+    IE_NAME = u'eitb.tv'
+    _VALID_URL = r'https?://www\.eitb\.tv/(eu/bideoa|es/video)/[^/]+/(?P<playlist_id>\d+)/(?P<chapter_id>\d+)'
+
+    _TEST = {
+        u'add_ie': ['Brightcove'],
+        u'url': u'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/2677100210001/2743577154001/lasa-y-zabala-30-anos/',
+        u'md5': u'edf4436247185adee3ea18ce64c47998',
+        u'info_dict': {
+            u'id': u'2743577154001',
+            u'ext': u'mp4',
+            u'title': u'60 minutos (Lasa y Zabala, 30 años)',
+            # All videos from eitb has this description in the brightcove info
+            u'description': u'.',
+            u'uploader': u'Euskal Telebista',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        chapter_id = mobj.group('chapter_id')
+        webpage = self._download_webpage(url, chapter_id)
+        bc_url = BrightcoveIE._extract_brightcove_url(webpage)
+        if bc_url is None:
+            raise ExtractorError(u'Could not extract the Brightcove url')
+        # The BrightcoveExperience object doesn't contain the video id, we set
+        # it manually
+        bc_url += '&%40videoPlayer={0}'.format(chapter_id)
+        return self.url_result(bc_url, BrightcoveIE.ie_key())

youtube_dl/extractor/exfm.py

@@ -11,16 +11,17 @@ class ExfmIE(InfoExtractor):
     _SOUNDCLOUD_URL = r'(?:http://)?(?:www\.)?api\.soundcloud.com/tracks/([^/]+)/stream'
     _TESTS = [
         {
-            u'url': u'http://ex.fm/song/1bgtzg',
-            u'file': u'95223130.mp3',
-            u'md5': u'8a7967a3fef10e59a1d6f86240fd41cf',
+            u'url': u'http://ex.fm/song/eh359',
+            u'file': u'44216187.mp3',
+            u'md5': u'e45513df5631e6d760970b14cc0c11e7',
             u'info_dict': {
-                u"title": u"We Can't Stop - Miley Cyrus",
-                u"uploader": u"Miley Cyrus",
-                u'upload_date': u'20130603',
-                u'description': u'Download "We Can\'t Stop" \r\niTunes: http://smarturl.it/WeCantStop?IQid=SC\r\nAmazon: http://smarturl.it/WeCantStopAMZ?IQid=SC',
+                u"title": u"Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive",
+                u"uploader": u"deadjournalist",
+                u'upload_date': u'20120424',
+                u'description': u'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive',
             },
             u'note': u'Soundcloud song',
+            u'skip': u'The site is down too often',
         },
         {
             u'url': u'http://ex.fm/song/wddt8',
@@ -30,6 +31,7 @@ class ExfmIE(InfoExtractor):
                 u'title': u'Safe and Sound',
                 u'uploader': u'Capital Cities',
             },
+            u'skip': u'The site is down too often',
         },
     ]
 

youtube_dl/extractor/extremetube.py

@@ -0,0 +1,50 @@
+import os
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
+    compat_urllib_parse,
+)
+
+class ExtremeTubeIE(InfoExtractor):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>extremetube\.com/video/.+?(?P<videoid>[0-9]+))(?:[/?&]|$)'
+    _TEST = {
+        u'url': u'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431',
+        u'file': u'652431.mp4',
+        u'md5': u'1fb9228f5e3332ec8c057d6ac36f33e0',
+        u'info_dict': {
+            u"title": u"Music Video 14 british euro brit european cumshots swallow",
+            u"uploader": u"unknown",
+            u"age_limit": 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+        url = 'http://www.' + mobj.group('url')
+
+        req = compat_urllib_request.Request(url)
+        req.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(req, video_id)
+
+        video_title = self._html_search_regex(r'<h1 [^>]*?title="([^"]+)"[^>]*>\1<', webpage, u'title')
+        uploader = self._html_search_regex(r'>Posted by:(?=<)(?:\s|<[^>]*>)*(.+?)\|', webpage, u'uploader', fatal=False)
+        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&amp;', webpage, u'video_url'))
+        path = compat_urllib_parse_urlparse(video_url).path
+        extension = os.path.splitext(path)[1][1:]
+        format = path.split('/')[5].split('_')[:2]
+        format = "-".join(format)
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'uploader': uploader,
+            'url': video_url,
+            'ext': extension,
+            'format': format,
+            'format_id': format,
+            'age_limit': 18,
+        }

youtube_dl/extractor/facebook.py

@@ -19,7 +19,8 @@ class FacebookIE(InfoExtractor):
     """Information Extractor for Facebook"""
 
     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
-    _LOGIN_URL = 'https://login.facebook.com/login.php?m&next=http%3A%2F%2Fm.facebook.com%2Fhome.php&'
+    _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
+    _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
     _NETRC_MACHINE = 'facebook'
     IE_NAME = u'facebook'
     _TEST = {
@@ -36,50 +37,56 @@ class FacebookIE(InfoExtractor):
         """Report attempt to log in."""
         self.to_screen(u'Logging in')
 
-    def _real_initialize(self):
-        if self._downloader is None:
-            return
-
-        useremail = None
-        password = None
-        downloader_params = self._downloader.params
-
-        # Attempt to use provided username and password or .netrc data
-        if downloader_params.get('username', None) is not None:
-            useremail = downloader_params['username']
-            password = downloader_params['password']
-        elif downloader_params.get('usenetrc', False):
-            try:
-                info = netrc.netrc().authenticators(self._NETRC_MACHINE)
-                if info is not None:
-                    useremail = info[0]
-                    password = info[2]
-                else:
-                    raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
-            except (IOError, netrc.NetrcParseError) as err:
-                self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
-                return
-
+    def _login(self):
+        (useremail, password) = self._get_login_info()
         if useremail is None:
             return
 
-        # Log in
+        login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
+        login_page_req.add_header('Cookie', 'locale=en_US')
+        self.report_login()
+        login_page = self._download_webpage(login_page_req, None, note=False,
+            errnote=u'Unable to download login page')
+        lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd')
+        lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd')
+
         login_form = {
             'email': useremail,
             'pass': password,
-            'login': 'Log+In'
+            'lsd': lsd,
+            'lgnrnd': lgnrnd,
+            'next': 'http://facebook.com/home.php',
+            'default_persistent': '0',
+            'legacy_return': '1',
+            'timezone': '-60',
+            'trynum': '1',
             }
         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:
-            self.report_login()
             login_results = compat_urllib_request.urlopen(request).read()
             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
                 self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
                 return
+
+            check_form = {
+                'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'),
+                'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'),
+                'name_action_selected': 'dont_save',
+                'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'),
+            }
+            check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
+            check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+            check_response = compat_urllib_request.urlopen(check_req).read()
+            if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
+                self._downloader.report_warning(u'Unable to confirm login, you have to login in your brower and authorize the login.')
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
             self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
             return
 
+    def _real_initialize(self):
+        self._login()
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
@@ -93,6 +100,12 @@ class FacebookIE(InfoExtractor):
         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
         if not m:
+            m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
+            if m_msg is not None:
+                raise ExtractorError(
+                    u'The video is not available, Facebook said: "%s"' % m_msg.group(1),
+                    expected=True)
+            else:
                 raise ExtractorError(u'Cannot parse data')
         data = dict(json.loads(m.group(1)))
         params_raw = compat_urllib_parse.unquote(data['params'])

youtube_dl/extractor/faz.py

@@ -0,0 +1,58 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+)
+
+
+class FazIE(InfoExtractor):
+    IE_NAME = u'faz.net'
+    _VALID_URL = r'https?://www\.faz\.net/multimedia/videos/.*?-(?P<id>\d+).html'
+
+    _TEST = {
+        u'url': u'http://www.faz.net/multimedia/videos/stockholm-chemie-nobelpreis-fuer-drei-amerikanische-forscher-12610585.html',
+        u'file': u'12610585.mp4',
+        u'info_dict': {
+            u'title': u'Stockholm: Chemie-Nobelpreis für drei amerikanische Forscher',
+            u'description': u'md5:1453fbf9a0d041d985a47306192ea253',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        self.to_screen(video_id)
+        webpage = self._download_webpage(url, video_id)
+        config_xml_url = self._search_regex(r'writeFLV\(\'(.+?)\',', webpage,
+            u'config xml url')
+        config_xml = self._download_webpage(config_xml_url, video_id,
+            u'Downloading config xml')
+        config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
+
+        encodings = config.find('ENCODINGS')
+        formats = []
+        for code in ['LOW', 'HIGH', 'HQ']:
+            encoding = encodings.find(code)
+            if encoding is None:
+                continue
+            encoding_url = encoding.find('FILENAME').text
+            formats.append({
+                'url': encoding_url,
+                'ext': determine_ext(encoding_url),
+                'format_id': code.lower(),
+            })
+
+        descr = self._html_search_regex(r'<p class="Content Copy">(.*?)</p>', webpage, u'description')
+        info = {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'description': descr,
+            'thumbnail': config.find('STILL/STILL_BIG').text,
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/flickr.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 class FlickrIE(InfoExtractor):
     """Information Extractor for Flickr videos"""
-    _VALID_URL = r'(?:https?://)?(?:www\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
+    _VALID_URL = r'(?:https?://)?(?:www\.|secure\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
     _TEST = {
         u'url': u'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
         u'file': u'5645318632.mp4',

youtube_dl/extractor/francetv.py

@@ -70,7 +70,11 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
 
 class France2IE(FranceTVBaseInfoExtractor):
     IE_NAME = u'france2.fr'
-    _VALID_URL = r'https?://www\.france2\.fr/emissions/.*?/videos/(?P<id>\d+)'
+    _VALID_URL = r'''(?x)https?://www\.france2\.fr/
+        (?:
+            emissions/.*?/videos/(?P<id>\d+)
+        |   emission/(?P<key>[^/?]+)
+        )'''
 
     _TEST = {
         u'url': u'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
@@ -86,12 +90,20 @@ class France2IE(FranceTVBaseInfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
+        if mobj.group('key'):
+            webpage = self._download_webpage(url, mobj.group('key'))
+            video_id = self._html_search_regex(
+                r'''(?x)<div\s+class="video-player">\s*
+                    <a\s+href="http://videos.francetv.fr/video/([0-9]+)"\s+
+                    class="francetv-video-player">''',
+                webpage, u'video ID')
+        else:
             video_id = mobj.group('id')
         return self._extract_video(video_id)
 
 
 class GenerationQuoiIE(InfoExtractor):
-    IE_NAME = u'http://generation-quoi.france2.fr'
+    IE_NAME = u'france2.fr:generation-quoi'
     _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<name>.*)(\?|$)'
 
     _TEST = {

youtube_dl/extractor/gamekings.py

@@ -0,0 +1,40 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+)
+
+
+class GamekingsIE(InfoExtractor):
+    _VALID_URL = r'http?://www\.gamekings\.tv/videos/(?P<name>[0-9a-z\-]+)'
+    _TEST = {
+        u"url": u"http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/",
+        u'file': u'20130811.mp4',
+        u'md5': u'17f6088f7d0149ff2b46f2714bdb1954',
+        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.",
+        }
+    }
+
+    def _real_extract(self, url):
+
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        webpage = self._download_webpage(url, name)
+        video_url = self._og_search_video_url(webpage)
+
+        video = re.search(r'[0-9]+', video_url)
+        video_id = video.group(0)
+
+        # Todo: add medium format
+        video_url = video_url.replace(video_id, 'large/' + video_id)
+
+        return {
+            'id': video_id,
+            'ext': 'mp4',
+            'url': video_url,
+            'title': self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
+        }

youtube_dl/extractor/gamespot.py

@@ -1,55 +1,59 @@
 import re
-import xml.etree.ElementTree
+import json
 
 from .common import InfoExtractor
 from ..utils import (
-    unified_strdate,
     compat_urllib_parse,
+    compat_urlparse,
+    unescapeHTML,
+    get_meta_content,
 )
 
+
 class GameSpotIE(InfoExtractor):
     _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
     _TEST = {
         u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
-        u"file": u"6410818.mp4",
+        u"file": u"gs-2300-6410818.mp4",
         u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
         u"info_dict": {
             u"title": u"Arma 3 - Community Guide: SITREP I",
-            u"upload_date": u"20130627", 
+            u'description': u'Check out this video where some of the basics of Arma 3 is explained.',
         }
     }
 
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        page_id = mobj.group('page_id')
+        page_id = video_id = mobj.group('page_id')
         webpage = self._download_webpage(url, page_id)
-        video_id = self._html_search_regex([r'"og:video" content=".*?\?id=(\d+)"',
-                                            r'http://www\.gamespot\.com/videoembed/(\d+)'],
-                                           webpage, 'video id')
-        data = compat_urllib_parse.urlencode({'id': video_id, 'newplayer': '1'})
-        info_url = 'http://www.gamespot.com/pages/video_player/xml.php?' + data
-        info_xml = self._download_webpage(info_url, video_id)
-        doc = xml.etree.ElementTree.fromstring(info_xml)
-        clip_el = doc.find('./playList/clip')
+        data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video')
+        data_video = json.loads(unescapeHTML(data_video_json))
 
-        http_urls = [{'url': node.find('filePath').text,
-                      'rate': int(node.find('rate').text)}
-            for node in clip_el.find('./httpURI')]
-        best_quality = sorted(http_urls, key=lambda f: f['rate'])[-1]
-        video_url = best_quality['url']
-        title = clip_el.find('./title').text
-        ext = video_url.rpartition('.')[2]
-        thumbnail_url = clip_el.find('./screenGrabURI').text
-        view_count = int(clip_el.find('./views').text)
-        upload_date = unified_strdate(clip_el.find('./postDate').text)
+        # Transform the manifest url to a link to the mp4 files
+        # they are used in mobile devices.
+        f4m_url = data_video['videoStreams']['f4m_stream']
+        f4m_path = compat_urlparse.urlparse(f4m_url).path
+        QUALITIES_RE = r'((,\d+)+,?)'
+        qualities = self._search_regex(QUALITIES_RE, f4m_path, u'qualities').strip(',').split(',')
+        http_path = f4m_path[1:].split('/', 1)[1]
+        http_template = re.sub(QUALITIES_RE, r'%s', http_path)
+        http_template = http_template.replace('.csmil/manifest.f4m', '')
+        http_template = compat_urlparse.urljoin('http://video.gamespotcdn.com/', http_template)
+        formats = []
+        for q in qualities:
+            formats.append({
+                'url': http_template % q,
+                'ext': 'mp4',
+                'format_id': q,
+            })
 
-        return [{
-            'id'          : video_id,
-            'url'         : video_url,
-            'ext'         : ext,
-            'title'       : title,
-            'thumbnail'   : thumbnail_url,
-            'upload_date' : upload_date,
-            'view_count'  : view_count,
-        }]
+        info = {
+            'id': data_video['guid'],
+            'title': compat_urllib_parse.unquote(data_video['title']),
+            'formats': formats,
+            'description': get_meta_content('description', webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }
+        # TODO: Remove when #980 has been merged
+        info.update(formats[-1])
+        return info

youtube_dl/extractor/generic.py

@@ -11,6 +11,8 @@ from ..utils import (
     compat_urlparse,
 
     ExtractorError,
+    smuggle_url,
+    unescapeHTML,
 )
 from .brightcove import BrightcoveIE
 
@@ -23,23 +25,52 @@ class GenericIE(InfoExtractor):
         {
             u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
             u'file': u'13601338388002.mp4',
-            u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89',
+            u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd',
             u'info_dict': {
                 u"uploader": u"www.hodiho.fr",
                 u"title": u"R\u00e9gis plante sa Jeep"
             }
         },
+        # embedded vimeo video
         {
-            u'url': u'http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/',
-            u'file': u'2371591881001.mp4',
-            u'md5': u'9e80619e0a94663f0bdc849b4566af19',
-            u'note': u'Test Brightcove downloads and detection in GenericIE',
+            u'add_ie': ['Vimeo'],
+            u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references',
+            u'file': u'22444065.mp4',
+            u'md5': u'2903896e23df39722c33f015af0666e2',
             u'info_dict': {
-                u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
-                u'uploader': u'8TV',
-                u'description': u'md5:a950cc4285c43e44d763d036710cd9cd',
+                u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011',
+                u"uploader_id": u"skillsmatter",
+                u"uploader": u"Skills Matter",
             }
         },
+        # bandcamp page with custom domain
+        {
+            u'add_ie': ['Bandcamp'],
+            u'url': u'http://bronyrock.com/track/the-pony-mash',
+            u'file': u'3235767654.mp3',
+            u'info_dict': {
+                u'title': u'The Pony Mash',
+                u'uploader': u'M_Pallante',
+            },
+            u'skip': u'There is a limit of 200 free downloads / month for the test song',
+        },
+        # embedded brightcove video
+        # it also tests brightcove videos that need to set the 'Referer' in the
+        # http requests
+        {
+            u'add_ie': ['Brightcove'],
+            u'url': u'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
+            u'info_dict': {
+                u'id': u'2765128793001',
+                u'ext': u'mp4',
+                u'title': u'Le cours de bourse : l’analyse technique',
+                u'description': u'md5:7e9ad046e968cb2d1114004aba466fd9',
+                u'uploader': u'BFM BUSINESS',
+            },
+            u'params': {
+                u'skip_download': True,
+            },
+        },
     ]
 
     def report_download_webpage(self, video_id):
@@ -128,16 +159,36 @@ class GenericIE(InfoExtractor):
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Failed to download URL: %s' % url)
 
         self.report_extraction(video_id)
         # Look for BrightCove:
-        m_brightcove = re.search(r'<object.+?class=([\'"]).*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL)
-        if m_brightcove is not None:
+        bc_url = BrightcoveIE._extract_brightcove_url(webpage)
+        if bc_url is not None:
             self.to_screen(u'Brightcove video detected.')
-            bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group())
             return self.url_result(bc_url, 'Brightcove')
 
+        # Look for embedded Vimeo player
+        mobj = re.search(
+            r'<iframe[^>]+?src="(https?://player.vimeo.com/video/.+?)"', webpage)
+        if mobj:
+            player_url = unescapeHTML(mobj.group(1))
+            surl = smuggle_url(player_url, {'Referer': url})
+            return self.url_result(surl, 'Vimeo')
+
+        # Look for embedded YouTube player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?youtube.com/embed/.+?)\1', webpage)
+        if mobj:
+            surl = unescapeHTML(mobj.group(u'url'))
+            return self.url_result(surl, 'Youtube')
+
+        # Look for Bandcamp pages with custom domain
+        mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
+        if mobj is not None:
+            burl = unescapeHTML(mobj.group(1))
+            return self.url_result(burl, 'Bandcamp')
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:
@@ -160,12 +211,12 @@ class GenericIE(InfoExtractor):
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Unsupported URL: %s' % url)
 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Did not find a valid video URL at %s' % url)
 
         video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)

youtube_dl/extractor/googleplus.py

@@ -41,8 +41,9 @@ class GooglePlusIE(InfoExtractor):
 
         # Extract update date
         upload_date = self._html_search_regex(
-            ['title="Timestamp">(.*?)</a>', r'<a.+?class="g-M.+?>(.+?)</a>'],
-            webpage, u'upload date', fatal=False)
+            r'''(?x)<a.+?class="o-U-s\s[^"]+"\s+style="display:\s*none"\s*>
+                    ([0-9]{4}-[0-9]{2}-[0-9]{2})</a>''',
+            webpage, u'upload date', fatal=False, flags=re.VERBOSE)
         if upload_date:
             # Convert timestring to a format suitable for filename
             upload_date = datetime.datetime.strptime(upload_date, "%Y-%m-%d")

youtube_dl/extractor/ign.py

@@ -13,7 +13,7 @@ class IGNIE(InfoExtractor):
     Some videos of it.ign.com are also supported
     """
 
-    _VALID_URL = r'https?://.+?\.ign\.com/(?P<type>videos|show_videos|articles)(/.+)?/(?P<name_or_id>.+)'
+    _VALID_URL = r'https?://.+?\.ign\.com/(?P<type>videos|show_videos|articles|(?:[^/]*/feature))(/.+)?/(?P<name_or_id>.+)'
     IE_NAME = u'ign.com'
 
     _CONFIG_URL_TEMPLATE = 'http://www.ign.com/videos/configs/id/%s.config'
@@ -21,7 +21,8 @@ class IGNIE(InfoExtractor):
                        r'id="my_show_video">.*?<p>(.*?)</p>',
                        ]
 
-    _TEST = {
+    _TESTS = [
+        {
             u'url': u'http://www.ign.com/videos/2013/06/05/the-last-of-us-review',
             u'file': u'8f862beef863986b2785559b9e1aa599.mp4',
             u'md5': u'eac8bdc1890980122c3b66f14bdd02e9',
@@ -29,7 +30,30 @@ class IGNIE(InfoExtractor):
                 u'title': u'The Last of Us Review',
                 u'description': u'md5:c8946d4260a4d43a00d5ae8ed998870c',
             }
-    }
+        },
+        {
+            u'url': u'http://me.ign.com/en/feature/15775/100-little-things-in-gta-5-that-will-blow-your-mind',
+            u'playlist': [
+                {
+                    u'file': u'5ebbd138523268b93c9141af17bec937.mp4',
+                    u'info_dict': {
+                        u'title': u'GTA 5 Video Review',
+                        u'description': u'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.',
+                    },
+                },
+                {
+                    u'file': u'638672ee848ae4ff108df2a296418ee2.mp4',
+                    u'info_dict': {
+                        u'title': u'GTA 5\'s Twisted Beauty in Super Slow Motion',
+                        u'description': u'The twisted beauty of GTA 5 in stunning slow motion.',
+                    },
+                },
+            ],
+            u'params': {
+                u'skip_download': True,
+            },
+        },
+    ]
 
     def _find_video_id(self, webpage):
         res_id = [r'data-video-id="(.+?)"',
@@ -46,6 +70,13 @@ class IGNIE(InfoExtractor):
         if page_type == 'articles':
             video_url = self._search_regex(r'var videoUrl = "(.+?)"', webpage, u'video url')
             return self.url_result(video_url, ie='IGN')
+        elif page_type != 'video':
+            multiple_urls = re.findall(
+                '<param name="flashvars" value="[^"]*?url=(https?://www\.ign\.com/videos/.*?)["&]',
+                webpage)
+            if multiple_urls:
+                return [self.url_result(u, ie='IGN') for u in multiple_urls]
+
         video_id = self._find_video_id(webpage)
         result = self._get_video_info(video_id)
         description = self._html_search_regex(self._DESCRIPTION_RE,
@@ -87,6 +118,9 @@ class OneUPIE(IGNIE):
         }
     }
 
+    # Override IGN tests
+    _TESTS = []
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         id = mobj.group('name_or_id')

youtube_dl/extractor/instagram.py

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

youtube_dl/extractor/internetvideoarchive.py

@@ -0,0 +1,84 @@
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    compat_urllib_parse,
+    xpath_with_ns,
+    determine_ext,
+)
+
+
+class InternetVideoArchiveIE(InfoExtractor):
+    _VALID_URL = r'https?://video\.internetvideoarchive\.net/flash/players/.*?\?.*?publishedid.*?'
+
+    _TEST = {
+        u'url': u'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?customerid=69249&publishedid=452693&playerid=247',
+        u'file': u'452693.mp4',
+        u'info_dict': {
+            u'title': u'SKYFALL',
+            u'description': u'In SKYFALL, Bond\'s loyalty to M is tested as her past comes back to haunt her. As MI6 comes under attack, 007 must track down and destroy the threat, no matter how personal the cost.',
+            u'duration': 153,
+        },
+    }
+
+    @staticmethod
+    def _build_url(query):
+        return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
+
+    @staticmethod
+    def _clean_query(query):
+        NEEDED_ARGS = ['publishedid', 'customerid']
+        query_dic = compat_urlparse.parse_qs(query)
+        cleaned_dic = dict((k,v[0]) for (k,v) in query_dic.items() if k in NEEDED_ARGS)
+        # Other player ids return m3u8 urls
+        cleaned_dic['playerid'] = '247'
+        cleaned_dic['videokbrate'] = '100000'
+        return compat_urllib_parse.urlencode(cleaned_dic)
+
+    def _real_extract(self, url):
+        query = compat_urlparse.urlparse(url).query
+        query_dic = compat_urlparse.parse_qs(query)
+        video_id = query_dic['publishedid'][0]
+        url = self._build_url(query)
+
+        flashconfiguration_xml = self._download_webpage(url, video_id,
+            u'Downloading flash configuration')
+        flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8'))
+        file_url = flashconfiguration.find('file').text
+        file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
+        # Replace some of the parameters in the query to get the best quality
+        # and http links (no m3u8 manifests)
+        file_url = re.sub(r'(?<=\?)(.+)$',
+            lambda m: self._clean_query(m.group()),
+            file_url)
+        info_xml = self._download_webpage(file_url, video_id,
+            u'Downloading video info')
+        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+        item = info.find('channel/item')
+
+        def _bp(p):
+            return xpath_with_ns(p,
+                {'media': 'http://search.yahoo.com/mrss/',
+                'jwplayer': 'http://developer.longtailvideo.com/trac/wiki/FlashFormats'})
+        formats = []
+        for content in item.findall(_bp('media:group/media:content')):
+            attr = content.attrib
+            f_url = attr['url']
+            formats.append({
+                'url': f_url,
+                'ext': determine_ext(f_url),
+                'width': int(attr['width']),
+                'bitrate': int(attr['bitrate']),
+            })
+        formats = sorted(formats, key=lambda f: f['bitrate'])
+
+        return {
+            'id': video_id,
+            'title': item.find('title').text,
+            'formats': formats,
+            'thumbnail': item.find(_bp('media:thumbnail')).attrib['url'],
+            'description': item.find('description').text,
+            'duration': int(attr['duration']),
+        }

youtube_dl/extractor/jeuxvideo.py

@@ -6,13 +6,14 @@ import xml.etree.ElementTree
 
 from .common import InfoExtractor
 
+
 class JeuxVideoIE(InfoExtractor):
     _VALID_URL = r'http://.*?\.jeuxvideo\.com/.*/(.*?)-\d+\.htm'
 
     _TEST = {
         u'url': u'http://www.jeuxvideo.com/reportages-videos-jeux/0004/00046170/tearaway-playstation-vita-gc-2013-tearaway-nous-presente-ses-papiers-d-identite-00115182.htm',
         u'file': u'5182.mp4',
-        u'md5': u'e0fdb0cd3ce98713ef9c1e1e025779d0',
+        u'md5': u'046e491afb32a8aaac1f44dd4ddd54ee',
         u'info_dict': {
             u'title': u'GC 2013 : Tearaway nous présente ses papiers d\'identité',
             u'description': u'Lorsque les développeurs de LittleBigPlanet proposent un nouveau titre, on ne peut que s\'attendre à un résultat original et fort attrayant.\n',
@@ -23,22 +24,26 @@ class JeuxVideoIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         title = re.match(self._VALID_URL, url).group(1)
         webpage = self._download_webpage(url, title)
-        m_download = re.search(r'<param name="flashvars" value="config=(.*?)" />', webpage)
+        xml_link = self._html_search_regex(
+            r'<param name="flashvars" value="config=(.*?)" />',
+            webpage, u'config URL')
         
-        xml_link = m_download.group(1)
+        video_id = self._search_regex(
+            r'http://www\.jeuxvideo\.com/config/\w+/\d+/(.*?)/\d+_player\.xml',
+            xml_link, u'video ID')
 
-        id = re.search(r'http://www.jeuxvideo.com/config/\w+/0011/(.*?)/\d+_player\.xml', xml_link).group(1)
-
-        xml_config = self._download_webpage(xml_link, title,
-                                                  'Downloading XML config')
+        xml_config = self._download_webpage(
+            xml_link, title, u'Downloading XML config')
         config = xml.etree.ElementTree.fromstring(xml_config.encode('utf-8'))
-        info = re.search(r'<format\.json>(.*?)</format\.json>',
-                         xml_config, re.MULTILINE|re.DOTALL).group(1)
-        info = json.loads(info)['versions'][0]
+        info_json = self._search_regex(
+            r'(?sm)<format\.json>(.*?)</format\.json>',
+            xml_config, u'JSON information')
+        info = json.loads(info_json)['versions'][0]
         
         video_url = 'http://video720.jeuxvideo.com/' + info['file']
 
-        return {'id': id,
+        return {
+            'id': video_id,
             'title': config.find('titre_video').text,
             'ext': 'mp4',
             'url': video_url,

youtube_dl/extractor/kankan.py

@@ -1,8 +1,10 @@
 import re
+import hashlib
 
 from .common import InfoExtractor
 from ..utils import determine_ext
 
+_md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
 
 class KankanIE(InfoExtractor):
     _VALID_URL = r'https?://(?:.*?\.)?kankan\.com/.+?/(?P<id>\d+)\.shtml'
@@ -30,7 +32,10 @@ class KankanIE(InfoExtractor):
                                                  video_id, u'Downloading video url info')
         ip = self._search_regex(r'ip:"(.+?)"', video_info_page, u'video url ip')
         path = self._search_regex(r'path:"(.+?)"', video_info_page, u'video url path')
-        video_url = 'http://%s%s' % (ip, path)
+        param1 = self._search_regex(r'param1:(\d+)', video_info_page, u'param1')
+        param2 = self._search_regex(r'param2:(\d+)', video_info_page, u'param2')
+        key = _md5('xl_mp43651' + param1 + param2)
+        video_url = 'http://%s%s?key=%s&key1=%s' % (ip, path, key, param2)
 
         return {'id': video_id,
                 'title': title,

youtube_dl/extractor/keezmovies.py

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

youtube_dl/extractor/livestream.py

@@ -40,13 +40,9 @@ class LivestreamIE(InfoExtractor):
 
         if video_id is None:
             # This is an event page:
-            player = get_meta_content('twitter:player', webpage)
-            if player is None:
-                raise ExtractorError('Couldn\'t extract event api url')
-            api_url = player.replace('/player', '')
-            api_url = re.sub(r'^(https?://)(new\.)', r'\1api.\2', api_url)
-            info = json.loads(self._download_webpage(api_url, event_name,
-                                                     u'Downloading event info'))
+            config_json = self._search_regex(r'window.config = ({.*?});',
+                webpage, u'window config')
+            info = json.loads(config_json)['event']
             videos = [self._extract_video_info(video_data['data'])
                 for video_data in info['feed']['data'] if video_data['type'] == u'video']
             return self.playlist_result(videos, info['id'], info['full_name'])

youtube_dl/extractor/metacafe.py

@@ -20,10 +20,12 @@ class MetacafeIE(InfoExtractor):
     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
     IE_NAME = u'metacafe'
-    _TESTS = [{
+    _TESTS = [
+    # Youtube video
+    {
         u"add_ie": ["Youtube"],
         u"url":  u"http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/",
-        u"file":  u"_aUehQsCQtM.flv",
+        u"file":  u"_aUehQsCQtM.mp4",
         u"info_dict": {
             u"upload_date": u"20090102",
             u"title": u"The Electric Company | \"Short I\" | PBS KIDS GO!",
@@ -32,15 +34,42 @@ class MetacafeIE(InfoExtractor):
             u"uploader_id": u"PBS"
         }
     },
+    # Normal metacafe video
+    {
+        u'url': u'http://www.metacafe.com/watch/11121940/news_stuff_you_wont_do_with_your_playstation_4/',
+        u'md5': u'6e0bca200eaad2552e6915ed6fd4d9ad',
+        u'info_dict': {
+            u'id': u'11121940',
+            u'ext': u'mp4',
+            u'title': u'News: Stuff You Won\'t Do with Your PlayStation 4',
+            u'uploader': u'ign',
+            u'description': u'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.',
+        },
+    },
+    # AnyClip video
     {
         u"url": u"http://www.metacafe.com/watch/an-dVVXnuY7Jh77J/the_andromeda_strain_1971_stop_the_bomb_part_3/",
         u"file": u"an-dVVXnuY7Jh77J.mp4",
         u"info_dict": {
             u"title": u"The Andromeda Strain (1971): Stop the Bomb Part 3",
             u"uploader": u"anyclip",
-            u"description": u"md5:38c711dd98f5bb87acf973d573442e67"
-        }
-    }]
+            u"description": u"md5:38c711dd98f5bb87acf973d573442e67",
+        },
+    },
+    # age-restricted video
+    {
+        u'url': u'http://www.metacafe.com/watch/5186653/bbc_internal_christmas_tape_79_uncensored_outtakes_etc/',
+        u'md5': u'98dde7c1a35d02178e8ab7560fe8bd09',
+        u'info_dict': {
+            u'id': u'5186653',
+            u'ext': u'mp4',
+            u'title': u'BBC INTERNAL Christmas Tape \'79 - UNCENSORED Outtakes, Etc.',
+            u'uploader': u'Dwayne Pipe',
+            u'description': u'md5:950bf4c581e2c059911fa3ffbe377e4b',
+            u'age_limit': 18,
+        },
+    },
+    ]
 
 
     def report_disclaimer(self):
@@ -62,6 +91,7 @@ class MetacafeIE(InfoExtractor):
             'submit': "Continue - I'm over 18",
             }
         request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:
             self.report_age_confirmation()
             compat_urllib_request.urlopen(request).read()
@@ -83,6 +113,11 @@ class MetacafeIE(InfoExtractor):
 
         # Retrieve video webpage to extract further information
         req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)
+
+        # AnyClip videos require the flashversion cookie so that we get the link
+        # to the mp4 file
+        mobj_an = re.match(r'^an-(.*?)$', video_id)
+        if mobj_an:
             req.headers['Cookie'] = 'flashVersion=0;'
         webpage = self._download_webpage(req, video_id)
 
@@ -125,6 +160,11 @@ class MetacafeIE(InfoExtractor):
                 r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
                 webpage, u'uploader nickname', fatal=False)
 
+        if re.search(r'"contentRating":"restricted"', webpage) is not None:
+            age_limit = 18
+        else:
+            age_limit = 0
+
         return {
             '_type':    'video',
             'id':       video_id,
@@ -134,4 +174,5 @@ class MetacafeIE(InfoExtractor):
             'upload_date':  None,
             'title':    video_title,
             'ext':      video_ext,
+            'age_limit': age_limit,
         }

youtube_dl/extractor/mofosex.py

@@ -0,0 +1,49 @@
+import os
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
+    compat_urllib_parse,
+)
+
+class MofosexIE(InfoExtractor):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>mofosex\.com/videos/(?P<videoid>[0-9]+)/.*?\.html)'
+    _TEST = {
+        u'url': u'http://www.mofosex.com/videos/5018/japanese-teen-music-video.html',
+        u'file': u'5018.mp4',
+        u'md5': u'1b2eb47ac33cc75d4a80e3026b613c5a',
+        u'info_dict': {
+            u"title": u"Japanese Teen Music Video",
+            u"age_limit": 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+        url = 'http://www.' + mobj.group('url')
+
+        req = compat_urllib_request.Request(url)
+        req.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(req, video_id)
+
+        video_title = self._html_search_regex(r'<h1>(.+?)<', webpage, u'title')
+        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'flashvars.video_url = \'([^\']+)', webpage, u'video_url'))
+        path = compat_urllib_parse_urlparse(video_url).path
+        extension = os.path.splitext(path)[1][1:]
+        format = path.split('/')[5].split('_')[:2]
+        format = "-".join(format)
+
+        age_limit = self._rta_search(webpage)
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'url': video_url,
+            'ext': extension,
+            'format': format,
+            'format_id': format,
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/mtv.py

@@ -26,6 +26,7 @@ class MTVIE(InfoExtractor):
             },
         },
         {
+            u'add_ie': ['Vevo'],
             u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
             u'file': u'USCJY1331283.mp4',
             u'md5': u'73b4e7fcadd88929292fe52c3ced8caf',
@@ -54,46 +55,57 @@ class MTVIE(InfoExtractor):
     def _get_thumbnail_url(self, uri, itemdoc):
         return 'http://mtv.mtvnimages.com/uri/' + uri
 
-    def _extract_video_url(self, metadataXml):
+    def _extract_video_formats(self, metadataXml):
         if '/error_country_block.swf' in metadataXml:
             raise ExtractorError(u'This video is not available from your country.', expected=True)
         mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8'))
         renditions = mdoc.findall('.//rendition')
 
-        # For now, always pick the highest quality.
-        rendition = renditions[-1]
-
+        formats = []
+        for rendition in mdoc.findall('.//rendition'):
             try:
                 _, _, ext = rendition.attrib['type'].partition('/')
-            format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate']
                 rtmp_video_url = rendition.find('./src').text
-        except KeyError:
+                formats.append({'ext': ext,
+                                'url': self._transform_rtmp_url(rtmp_video_url),
+                                'format_id': rendition.get('bitrate'),
+                                'width': int(rendition.get('width')),
+                                'height': int(rendition.get('height')),
+                                })
+            except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')
-        video_url = self._transform_rtmp_url(rtmp_video_url)
-        return {'ext': ext, 'url': video_url, 'format': format}
+        return formats
 
     def _get_video_info(self, itemdoc):
         uri = itemdoc.find('guid').text
         video_id = self._id_from_uri(uri)
         self.report_extraction(video_id)
         mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
+        # Remove the templates, like &device={device}
+        mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', u'', mediagen_url)
         if 'acceptMethods' not in mediagen_url:
             mediagen_url += '&acceptMethods=fms'
         mediagen_page = self._download_webpage(mediagen_url, video_id,
                                                u'Downloading video urls')
-        video_info = self._extract_video_url(mediagen_page)
 
         description_node = itemdoc.find('description')
         if description_node is not None:
-            description = description_node.text
+            description = description_node.text.strip()
         else:
             description = None
-        video_info.update({'title': itemdoc.find('title').text,
+
+        info = {
+            'title': itemdoc.find('title').text,
+            'formats': self._extract_video_formats(mediagen_page),
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,
-                           })
-        return video_info
+        }
+
+        # TODO: Remove when #980 has been merged
+        info.update(info['formats'][-1])
+
+        return info
 
     def _get_videos_info(self, uri):
         video_id = self._id_from_uri(uri)

youtube_dl/extractor/myspace.py

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

youtube_dl/extractor/nhl.py

@@ -0,0 +1,120 @@
+import re
+import json
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    compat_urllib_parse,
+    determine_ext,
+    unified_strdate,
+)
+
+
+class NHLBaseInfoExtractor(InfoExtractor):
+    @staticmethod
+    def _fix_json(json_string):
+        return json_string.replace('\\\'', '\'')
+
+    def _extract_video(self, info):
+        video_id = info['id']
+        self.report_extraction(video_id)
+
+        initial_video_url = info['publishPoint']
+        data = compat_urllib_parse.urlencode({
+            'type': 'fvod',
+            'path': initial_video_url.replace('.mp4', '_sd.mp4'),
+        })
+        path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
+        path_response = self._download_webpage(path_url, video_id,
+            u'Downloading final video url')
+        path_doc = xml.etree.ElementTree.fromstring(path_response)
+        video_url = path_doc.find('path').text
+
+        join = compat_urlparse.urljoin
+        return {
+            'id': video_id,
+            'title': info['name'],
+            'url': video_url,
+            'ext': determine_ext(video_url),
+            'description': info['description'],
+            'duration': int(info['duration']),
+            'thumbnail': join(join(video_url, '/u/'), info['bigImage']),
+            'upload_date': unified_strdate(info['releaseDate'].split('.')[0]),
+        }
+
+
+class NHLIE(NHLBaseInfoExtractor):
+    IE_NAME = u'nhl.com'
+    _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console\?.*?(?<=[?&])id=(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
+        u'file': u'453614.mp4',
+        u'info_dict': {
+            u'title': u'Quick clip: Weise 4-3 goal vs Flames',
+            u'description': u'Dale Weise scores his first of the season to put the Canucks up 4-3.',
+            u'duration': 18,
+            u'upload_date': u'20131006',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
+        info_json = self._download_webpage(json_url, video_id,
+            u'Downloading info json')
+        info_json = self._fix_json(info_json)
+        info = json.loads(info_json)[0]
+        return self._extract_video(info)
+
+
+class NHLVideocenterIE(NHLBaseInfoExtractor):
+    IE_NAME = u'nhl.com:videocenter'
+    IE_DESC = u'Download the first 12 videos from a videocenter category'
+    _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?.*?catid=(?P<catid>[^&]+))?'
+
+    @classmethod
+    def suitable(cls, url):
+        if NHLIE.suitable(url):
+            return False
+        return super(NHLVideocenterIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        team = mobj.group('team')
+        webpage = self._download_webpage(url, team)
+        cat_id = self._search_regex(
+            [r'var defaultCatId = "(.+?)";',
+             r'{statusIndex:0,index:0,.*?id:(.*?),'],
+            webpage, u'category id')
+        playlist_title = self._html_search_regex(
+            r'tab0"[^>]*?>(.*?)</td>',
+            webpage, u'playlist title', flags=re.DOTALL).lower().capitalize()
+
+        data = compat_urllib_parse.urlencode({
+            'cid': cat_id,
+            # This is the default value
+            'count': 12,
+            'ptrs': 3,
+            'format': 'json',
+        })
+        path = '/videocenter/servlets/browse?' + data
+        request_url = compat_urlparse.urljoin(url, path)
+        response = self._download_webpage(request_url, playlist_title)
+        response = self._fix_json(response)
+        if not response.strip():
+            self._downloader.report_warning(u'Got an empty reponse, trying '
+                                            u'adding the "newvideos" parameter')
+            response = self._download_webpage(request_url + '&newvideos=true',
+                playlist_title)
+            response = self._fix_json(response)
+        videos = json.loads(response)
+
+        return {
+            '_type': 'playlist',
+            'title': playlist_title,
+            'id': cat_id,
+            'entries': [self._extract_video(i) for i in videos],
+        }

youtube_dl/extractor/nowvideo.py

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

youtube_dl/extractor/pornhub.py

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

youtube_dl/extractor/pornotube.py

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

youtube_dl/extractor/redtube.py

@@ -10,7 +10,8 @@ class RedTubeIE(InfoExtractor):
         u'file': u'66418.mp4',
         u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d',
         u'info_dict': {
-            u"title": u"Sucked on a toilet"
+            u"title": u"Sucked on a toilet",
+            u"age_limit": 18,
         }
     }
 
@@ -23,15 +24,21 @@ class RedTubeIE(InfoExtractor):
 
         self.report_extraction(video_id)
 
-        video_url = self._html_search_regex(r'<source src="(.+?)" type="video/mp4">',
-            webpage, u'video URL')
+        video_url = self._html_search_regex(
+            r'<source src="(.+?)" type="video/mp4">', webpage, u'video URL')
 
-        video_title = self._html_search_regex('<h1 class="videoTitle slidePanelMovable">(.+?)</h1>',
+        video_title = self._html_search_regex(
+            r'<h1 class="videoTitle slidePanelMovable">(.+?)</h1>',
             webpage, u'title')
 
-        return [{
+        # No self-labeling, but they describe themselves as
+        # "Home of Videos Porno"
+        age_limit = 18
+
+        return {
             'id':        video_id,
             'url':       video_url,
             'ext':       video_extension,
             'title':     video_title,
-        }]
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/rottentomatoes.py

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

youtube_dl/extractor/rtlnow.py

@@ -8,8 +8,8 @@ from ..utils import (
 )
 
 class RTLnowIE(InfoExtractor):
-    """Information Extractor for RTL NOW, RTL2 NOW, SUPER RTL NOW and VOX NOW"""
-    _VALID_URL = r'(?:http://)?(?P<url>(?P<base_url>rtl-now\.rtl\.de/|rtl2now\.rtl2\.de/|(?:www\.)?voxnow\.de/|(?:www\.)?superrtlnow\.de/)[a-zA-Z0-9-]+/[a-zA-Z0-9-]+\.php\?(?:container_id|film_id)=(?P<video_id>[0-9]+)&player=1(?:&season=[0-9]+)?(?:&.*)?)'
+    """Information Extractor for RTL NOW, RTL2 NOW, RTL NITRO, SUPER RTL NOW, VOX NOW and n-tv NOW"""
+    _VALID_URL = r'(?:http://)?(?P<url>(?P<base_url>rtl-now\.rtl\.de/|rtl2now\.rtl2\.de/|(?:www\.)?voxnow\.de/|(?:www\.)?rtlnitronow\.de/|(?:www\.)?superrtlnow\.de/|(?:www\.)?n-tvnow\.de/)[a-zA-Z0-9-]+/[a-zA-Z0-9-]+\.php\?(?:container_id|film_id)=(?P<video_id>[0-9]+)&player=1(?:&season=[0-9]+)?(?:&.*)?)'
     _TESTS = [{
         u'url': u'http://rtl-now.rtl.de/ahornallee/folge-1.php?film_id=90419&player=1&season=1',
         u'file': u'90419.flv',
@@ -61,8 +61,34 @@ class RTLnowIE(InfoExtractor):
         u'params': {
             u'skip_download': True,
         },
+    },
+    {
+        u'url': u'http://www.rtlnitronow.de/recht-ordnung/stadtpolizei-frankfurt-gerichtsvollzieher-leipzig.php?film_id=129679&player=1&season=1',
+        u'file': u'129679.flv',
+        u'info_dict': {
+            u'upload_date': u'20131016', 
+            u'title': u'Recht & Ordnung - Stadtpolizei Frankfurt/ Gerichtsvollzieher...',
+            u'description': u'Stadtpolizei Frankfurt/ Gerichtsvollzieher Leipzig',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    },
+    {
+        u'url': u'http://www.n-tvnow.de/top-gear/episode-1-2013-01-01-00-00-00.php?film_id=124903&player=1&season=10',
+        u'file': u'124903.flv',
+        u'info_dict': {
+            u'upload_date': u'20130101', 
+            u'title': u'Top Gear vom 01.01.2013',
+            u'description': u'Episode 1',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+        u'skip': u'Only works from Germany',
     }]
 
+
     def _real_extract(self,url):
         mobj = re.match(self._VALID_URL, url)
 
@@ -79,19 +105,22 @@ class RTLnowIE(InfoExtractor):
             msg = clean_html(note_m.group(1))
             raise ExtractorError(msg)
 
-        video_title = self._html_search_regex(r'<title>(?P<title>[^<]+)</title>',
+        video_title = self._html_search_regex(r'<title>(?P<title>[^<]+?)( \| [^<]*)?</title>',
             webpage, u'title')
         playerdata_url = self._html_search_regex(r'\'playerdata\': \'(?P<playerdata_url>[^\']+)\'',
             webpage, u'playerdata_url')
 
         playerdata = self._download_webpage(playerdata_url, video_id)
-        mobj = re.search(r'<title><!\[CDATA\[(?P<description>.+?)\s+- (?:Sendung )?vom (?P<upload_date_d>[0-9]{2})\.(?P<upload_date_m>[0-9]{2})\.(?:(?P<upload_date_Y>[0-9]{4})|(?P<upload_date_y>[0-9]{2})) [0-9]{2}:[0-9]{2} Uhr\]\]></title>', playerdata)
+        mobj = re.search(r'<title><!\[CDATA\[(?P<description>.+?)(?:\s+- (?:Sendung )?vom (?P<upload_date_d>[0-9]{2})\.(?P<upload_date_m>[0-9]{2})\.(?:(?P<upload_date_Y>[0-9]{4})|(?P<upload_date_y>[0-9]{2})) [0-9]{2}:[0-9]{2} Uhr)?\]\]></title>', playerdata)
         if mobj:
             video_description = mobj.group(u'description')
             if mobj.group('upload_date_Y'):
                 video_upload_date = mobj.group('upload_date_Y')
-            else:
+            elif mobj.group('upload_date_y'):
                 video_upload_date = u'20' + mobj.group('upload_date_y')
+            else:
+                video_upload_date = None
+            if video_upload_date:
                 video_upload_date += mobj.group('upload_date_m')+mobj.group('upload_date_d')
         else:
             video_description = None

youtube_dl/extractor/rutube.py

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

youtube_dl/extractor/slashdot.py

@@ -7,6 +7,7 @@ class SlashdotIE(InfoExtractor):
     _VALID_URL = r'https?://tv.slashdot.org/video/\?embed=(?P<id>.*?)(&|$)'
 
     _TEST = {
+        u'add_ie': ['Ooyala'],
         u'url': u'http://tv.slashdot.org/video/?embed=JscHMzZDplD0p-yNLOzTfzC3Q3xzJaUz',
         u'file': u'JscHMzZDplD0p-yNLOzTfzC3Q3xzJaUz.mp4',
         u'md5': u'd2222e7a4a4c1541b3e0cf732fb26735',

youtube_dl/extractor/soundcloud.py

@@ -29,7 +29,8 @@ class SoundcloudIE(InfoExtractor):
                     )
                     '''
     IE_NAME = u'soundcloud'
-    _TEST = {
+    _TESTS = [
+        {
             u'url': u'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
             u'file': u'62986583.mp3',
             u'md5': u'ebef0a451b909710ed1d7787dddbf0d7',
@@ -39,7 +40,23 @@ class SoundcloudIE(InfoExtractor):
                 u"uploader": u"E.T. ExTerrestrial Music", 
                 u"title": u"Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1"
             }
-    }
+        },
+        # not streamable song
+        {
+            u'url': u'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
+            u'info_dict': {
+                u'id': u'47127627',
+                u'ext': u'mp3',
+                u'title': u'Goldrushed',
+                u'uploader': u'The Royal Concept',
+                u'upload_date': u'20120521',
+            },
+            u'params': {
+                # rtmp
+                u'skip_download': True,
+            },
+        },
+    ]
 
     _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28'
 
@@ -56,16 +73,16 @@ class SoundcloudIE(InfoExtractor):
         return 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID
 
     def _extract_info_dict(self, info, full_title=None, quiet=False):
-        video_id = info['id']
-        name = full_title or video_id
+        track_id = compat_str(info['id'])
+        name = full_title or track_id
         if quiet == False:
             self.report_extraction(name)
 
         thumbnail = info['artwork_url']
         if thumbnail is not None:
             thumbnail = thumbnail.replace('-large', '-t500x500')
-        return {
-            'id':       info['id'],
+        result = {
+            'id':       track_id,
             'url':      info['stream_url'] + '?client_id=' + self._CLIENT_ID,
             'uploader': info['user']['username'],
             'upload_date': unified_strdate(info['created_at']),
@@ -74,6 +91,21 @@ class SoundcloudIE(InfoExtractor):
             '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)
+        if not info.get('streamable', False):
+            # We have to get the rtmp url
+            stream_json = self._download_webpage(
+                'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._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
+            url, path = rtmp_url.split('mp3:', 1)
+            result.update({
+                'url': url,
+                'play_path': 'mp3:' + path,
+            })
+        return result
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
@@ -106,70 +138,8 @@ class SoundcloudIE(InfoExtractor):
 class SoundcloudSetIE(SoundcloudIE):
     _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
     IE_NAME = u'soundcloud:set'
-    _TEST = {
-        u"url":"https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep",
-        u"playlist": [
-            {
-                u"file":"30510138.mp3",
-                u"md5":"f9136bf103901728f29e419d2c70f55d",
-                u"info_dict": {
-                    u"upload_date": u"20111213",
-                    u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
-                    u"uploader": u"The Royal Concept",
-                    u"title": u"D-D-Dance"
-                }
-            },
-            {
-                u"file":"47127625.mp3",
-                u"md5":"09b6758a018470570f8fd423c9453dd8",
-                u"info_dict": {
-                    u"upload_date": u"20120521",
-                    u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
-                    u"uploader": u"The Royal Concept",
-                    u"title": u"The Royal Concept - Gimme Twice"
-                }
-            },
-            {
-                u"file":"47127627.mp3",
-                u"md5":"154abd4e418cea19c3b901f1e1306d9c",
-                u"info_dict": {
-                    u"upload_date": u"20120521",
-                    u"uploader": u"The Royal Concept",
-                    u"title": u"Goldrushed"
-                }
-            },
-            {
-                u"file":"47127629.mp3",
-                u"md5":"2f5471edc79ad3f33a683153e96a79c1",
-                u"info_dict": {
-                    u"upload_date": u"20120521",
-                    u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
-                    u"uploader": u"The Royal Concept",
-                    u"title": u"In the End"
-                }
-            },
-            {
-                u"file":"47127631.mp3",
-                u"md5":"f9ba87aa940af7213f98949254f1c6e2",
-                u"info_dict": {
-                    u"upload_date": u"20120521",
-                    u"description": u"The Royal Concept from Stockholm\r\nFilip / David / Povel / Magnus\r\nwww.theroyalconceptband.com",
-                    u"uploader": u"The Royal Concept",
-                    u"title": u"Knocked Up"
-                }
-            },
-            {
-                u"file":"75206121.mp3",
-                u"md5":"f9d1fe9406717e302980c30de4af9353",
-                u"info_dict": {
-                    u"upload_date": u"20130116",
-                    u"description": u"The unreleased track World on Fire premiered on the CW's hit show Arrow (8pm/7pm central).  \r\nAs a gift to our fans we would like to offer you a free download of the track!  ",
-                    u"uploader": u"The Royal Concept",
-                    u"title": u"World On Fire"
-                }
-            }
-        ]
-    }
+    # it's in tests/test_playlists.py
+    _TESTS = []
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -208,7 +178,7 @@ class SoundcloudUserIE(SoundcloudIE):
     IE_NAME = u'soundcloud:user'
 
     # it's in tests/test_playlists.py
-    _TEST = None
+    _TESTS = []
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/space.py

@@ -0,0 +1,35 @@
+import re
+
+from .common import InfoExtractor
+from .brightcove import BrightcoveIE
+from ..utils import RegexNotFoundError, ExtractorError
+
+
+class SpaceIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.space\.com/\d+-(?P<title>[^/\.\?]*?)-video.html'
+    _TEST = {
+        u'add_ie': ['Brightcove'],
+        u'url': u'http://www.space.com/23373-huge-martian-landforms-detail-revealed-by-european-probe-video.html',
+        u'info_dict': {
+            u'id': u'2780937028001',
+            u'ext': u'mp4',
+            u'title': u'Huge Martian Landforms\' Detail Revealed By European Probe | Video',
+            u'description': u'md5:db81cf7f3122f95ed234b631a6ea1e61',
+            u'uploader': u'TechMedia Networks',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        title = mobj.group('title')
+        webpage = self._download_webpage(url, title)
+        try:
+            # Some videos require the playerKey field, which isn't define in
+            # the BrightcoveExperience object
+            brightcove_url = self._og_search_video_url(webpage)
+        except RegexNotFoundError:
+            # Other videos works fine with the info from the object
+            brightcove_url = BrightcoveIE._extract_brightcove_url(webpage)
+        if brightcove_url is None:
+            raise ExtractorError(u'The webpage does not contain a video', expected=True)
+        return self.url_result(brightcove_url, BrightcoveIE.ie_key())

youtube_dl/extractor/spankwire.py

@@ -0,0 +1,74 @@
+import os
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
+    compat_urllib_parse,
+    unescapeHTML,
+)
+from ..aes import (
+    aes_decrypt_text
+)
+
+class SpankwireIE(InfoExtractor):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>spankwire\.com/[^/]*/video(?P<videoid>[0-9]+)/?)'
+    _TEST = {
+        u'url': u'http://www.spankwire.com/Buckcherry-s-X-Rated-Music-Video-Crazy-Bitch/video103545/',
+        u'file': u'103545.mp4',
+        u'md5': u'1b3f55e345500552dbc252a3e9c1af43',
+        u'info_dict': {
+            u"uploader": u"oreusz", 
+            u"title": u"Buckcherry`s X Rated Music Video Crazy Bitch",
+            u"description": u"Crazy Bitch X rated music video.",
+            u"age_limit": 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+        url = 'http://www.' + mobj.group('url')
+
+        req = compat_urllib_request.Request(url)
+        req.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(req, video_id)
+
+        video_title = self._html_search_regex(r'<h1>([^<]+)', webpage, u'title')
+        video_uploader = self._html_search_regex(r'by:\s*<a [^>]*>(.+?)</a>', webpage, u'uploader', fatal=False)
+        thumbnail = self._html_search_regex(r'flashvars\.image_url = "([^"]+)', webpage, u'thumbnail', fatal=False)
+        description = self._html_search_regex(r'>\s*Description:</div>\s*<[^>]*>([^<]+)', webpage, u'description', fatal=False)
+        if len(description) == 0:
+            description = None
+
+        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:
+            password = self._html_search_regex(r'flashvars\.video_title = "([^"]+)', webpage, u'password').replace('+', ' ')
+            video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
+
+        formats = []
+        for video_url in video_urls:
+            path = compat_urllib_parse_urlparse(video_url).path
+            extension = os.path.splitext(path)[1][1:]
+            format = path.split('/')[4].split('_')[:2]
+            format = "-".join(format)
+            formats.append({
+                'url': video_url,
+                'ext': extension,
+                'format': format,
+                'format_id': format,
+            })
+        formats.sort(key=lambda format: list(map(lambda s: s.zfill(6), format['format'].split('-'))))
+
+        age_limit = self._rta_search(webpage)
+
+        return {
+            'id': video_id,
+            'uploader': video_uploader,
+            'title': video_title,
+            'thumbnail': thumbnail,
+            'description': description,
+            'formats': formats,
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/subtitles.py

@@ -12,9 +12,9 @@ class SubtitlesInfoExtractor(InfoExtractor):
         return any([self._downloader.params.get('writesubtitles', False),
                     self._downloader.params.get('writeautomaticsub')])
 
-    def _list_available_subtitles(self, video_id, webpage=None):
+    def _list_available_subtitles(self, video_id, webpage):
         """ outputs the available subtitles for the video """
-        sub_lang_list = self._get_available_subtitles(video_id)
+        sub_lang_list = self._get_available_subtitles(video_id, webpage)
         auto_captions_list = self._get_available_automatic_caption(video_id, webpage)
         sub_lang = ",".join(list(sub_lang_list.keys()))
         self.to_screen(u'%s: Available subtitles for video: %s' %
@@ -23,7 +23,7 @@ class SubtitlesInfoExtractor(InfoExtractor):
         self.to_screen(u'%s: Available automatic captions for video: %s' %
                        (video_id, auto_lang))
 
-    def extract_subtitles(self, video_id, video_webpage=None):
+    def extract_subtitles(self, video_id, webpage):
         """
         returns {sub_lang: sub} ,{} if subtitles not found or None if the
         subtitles aren't requested.
@@ -32,9 +32,9 @@ class SubtitlesInfoExtractor(InfoExtractor):
             return None
         available_subs_list = {}
         if self._downloader.params.get('writeautomaticsub', False):
-            available_subs_list.update(self._get_available_automatic_caption(video_id, video_webpage))
+            available_subs_list.update(self._get_available_automatic_caption(video_id, webpage))
         if self._downloader.params.get('writesubtitles', False):
-            available_subs_list.update(self._get_available_subtitles(video_id))
+            available_subs_list.update(self._get_available_subtitles(video_id, webpage))
 
         if not available_subs_list:  # error, it didn't get the available subtitles
             return {}
@@ -74,7 +74,7 @@ class SubtitlesInfoExtractor(InfoExtractor):
             return
         return sub
 
-    def _get_available_subtitles(self, video_id):
+    def _get_available_subtitles(self, video_id, webpage):
         """
         returns {sub_lang: url} or {} if not available
         Must be redefined by the subclasses

youtube_dl/extractor/sztvhu.py

@@ -0,0 +1,44 @@
+# -*- coding: utf-8 -*-
+
+import re
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+
+class SztvHuIE(InfoExtractor):
+    _VALID_URL = r'(?:http://)?(?:(?:www\.)?sztv\.hu|www\.tvszombathely\.hu)/(?:[^/]+)/.+-(?P<id>[0-9]+)'
+    _TEST = {
+        u'url': u'http://sztv.hu/hirek/cserkeszek-nepszerusitettek-a-kornyezettudatos-eletmodot-a-savaria-teren-20130909',
+        u'file': u'20130909.mp4',
+        u'md5': u'a6df607b11fb07d0e9f2ad94613375cb',
+        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...',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        video_file = self._search_regex(
+            r'file: "...:(.*?)",', webpage, 'video file')
+        title = self._html_search_regex(
+            r'<meta name="title" content="([^"]*?) - [^-]*? - [^-]*?"',
+            webpage, 'video title')
+        description = self._html_search_regex(
+            r'<meta name="description" content="([^"]*)"/>',
+            webpage, 'video description', fatal=False)
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        video_url = 'http://media.sztv.hu/vod/' + video_file
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'ext': determine_ext(video_url),
+            'description': description,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/teamcoco.py

@@ -1,4 +1,5 @@
 import re
+import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -11,7 +12,7 @@ class TeamcocoIE(InfoExtractor):
     _TEST = {
         u'url': u'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
         u'file': u'19705.mp4',
-        u'md5': u'27b6f7527da5acf534b15f21b032656e',
+        u'md5': u'cde9ba0fa3506f5f017ce11ead928f9a',
         u'info_dict': {
             u"description": u"Louis C.K. got starstruck by George W. Bush, so what? Part one.", 
             u"title": u"Louis C.K. Interview Pt. 1 11/3/11"
@@ -31,16 +32,40 @@ class TeamcocoIE(InfoExtractor):
         self.report_extraction(video_id)
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
-        data = self._download_webpage(data_url, video_id, 'Downloading data webpage')
+        data_xml = self._download_webpage(data_url, video_id, 'Downloading data webpage')
+        data = xml.etree.ElementTree.fromstring(data_xml.encode('utf-8'))
 
-        video_url = self._html_search_regex(r'<file [^>]*type="high".*?>(.*?)</file>',
-            data, u'video URL')
 
-        return [{
-            'id':          video_id,
-            'url':         video_url,
+        qualities = ['500k', '480p', '1000k', '720p', '1080p']
+        formats = []
+        for file in data.findall('files/file'):
+            if file.attrib.get('playmode') == 'all':
+                # it just duplicates one of the entries
+                break
+            file_url = file.text
+            m_format = re.search(r'(\d+(k|p))\.mp4', file_url)
+            if m_format is not None:
+                format_id = m_format.group(1)
+            else:
+                format_id = file.attrib['bitrate']
+            formats.append({
+                'url': file_url,
                 'ext': 'mp4',
+                'format_id': format_id,
+            })
+        def sort_key(f):
+            try:
+                return qualities.index(f['format_id'])
+            except ValueError:
+                return -1
+        formats.sort(key=sort_key)
+        if not formats:
+            raise RegexNotFoundError(u'Unable to extract video URL')
+
+        return {
+            'id':          video_id,
+            'formats': formats,
             'title':       self._og_search_title(webpage),
             'thumbnail':   self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),
-        }]
+        }

youtube_dl/extractor/techtalks.py

@@ -0,0 +1,65 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    get_element_by_attribute,
+    clean_html,
+)
+
+
+class TechTalksIE(InfoExtractor):
+    _VALID_URL = r'https?://techtalks\.tv/talks/[^/]*/(?P<id>\d+)/'
+
+    _TEST = {
+        u'url': u'http://techtalks.tv/talks/learning-topic-models-going-beyond-svd/57758/',
+        u'playlist': [
+            {
+                u'file': u'57758.flv',
+                u'info_dict': {
+                    u'title': u'Learning Topic Models --- Going beyond SVD',
+                },
+            },
+            {
+                u'file': u'57758-slides.flv',
+                u'info_dict': {
+                    u'title': u'Learning Topic Models --- Going beyond SVD',
+                },
+            },
+        ],
+        u'params': {
+            # rtmp download
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        talk_id = mobj.group('id')
+        webpage = self._download_webpage(url, talk_id)
+        rtmp_url = self._search_regex(r'netConnectionUrl: \'(.*?)\'', webpage,
+            u'rtmp url')
+        play_path = self._search_regex(r'href=\'(.*?)\' [^>]*id="flowplayer_presenter"',
+            webpage, u'presenter play path')
+        title = clean_html(get_element_by_attribute('class', 'title', webpage))
+        video_info = {
+                'id': talk_id,
+                'title': title,
+                'url': rtmp_url,
+                'play_path': play_path,
+                'ext': 'flv',
+            }
+        m_slides = re.search(r'<a class="slides" href=\'(.*?)\'', webpage)
+        if m_slides is None:
+            return video_info
+        else:
+            return [
+                video_info,
+                # The slides video
+                {
+                    'id': talk_id + '-slides',
+                    'title': title,
+                    'url': rtmp_url,
+                    'play_path': m_slides.group(1),
+                    'ext': 'flv',
+                },
+            ]

youtube_dl/extractor/ted.py

@@ -1,10 +1,14 @@
 import json
 import re
 
-from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
 
+from ..utils import (
+    compat_str,
+    RegexNotFoundError,
+)
 
-class TEDIE(InfoExtractor):
+class TEDIE(SubtitlesInfoExtractor):
     _VALID_URL=r'''http://www\.ted\.com/
                    (
                         ((?P<type_playlist>playlists)/(?P<playlist_id>\d+)) # We have a playlist
@@ -32,7 +36,7 @@ class TEDIE(InfoExtractor):
     def _real_extract(self, url):
         m=re.match(self._VALID_URL, url, re.VERBOSE)
         if m.group('type_talk'):
-            return [self._talk_info(url)]
+            return self._talk_info(url)
         else :
             playlist_id=m.group('playlist_id')
             name=m.group('name')
@@ -77,12 +81,44 @@ class TEDIE(InfoExtractor):
         
         thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"',
                                        webpage, 'thumbnail')
-        info = {
-                'id': info['id'],
-                'url': info['htmlStreams'][-1]['file'],
+        formats = [{
             'ext': 'mp4',
+            'url': stream['file'],
+            'format': stream['id']
+            } for stream in info['htmlStreams']]
+
+        video_id = info['id']
+
+        # subtitles
+        video_subtitles = self.extract_subtitles(video_id, webpage)
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id, webpage)
+            return
+
+        info = {
+            'id': video_id,
             'title': title,
             'thumbnail': thumbnail,
             'description': desc,
+            'subtitles': video_subtitles,
+            '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)
+            languages = re.findall(r'(?:<option value=")(\S+)"', options)
+            if languages:
+                sub_lang_list = {}
+                for l in languages:
+                    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:
+            self._downloader.report_warning(u'video doesn\'t have subtitles')
+        return {}

youtube_dl/extractor/tube8.py

@@ -0,0 +1,65 @@
+import os
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
+    compat_urllib_parse,
+    unescapeHTML,
+)
+from ..aes import (
+    aes_decrypt_text
+)
+
+class Tube8IE(InfoExtractor):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>tube8\.com/[^/]+/[^/]+/(?P<videoid>[0-9]+)/?)'
+    _TEST = {
+        u'url': u'http://www.tube8.com/teen/kasia-music-video/229795/',
+        u'file': u'229795.mp4',
+        u'md5': u'e9e0b0c86734e5e3766e653509475db0',
+        u'info_dict': {
+            u"description": u"hot teen Kasia grinding", 
+            u"uploader": u"unknown", 
+            u"title": u"Kasia music video",
+            u"age_limit": 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+        url = 'http://www.' + mobj.group('url')
+
+        req = compat_urllib_request.Request(url)
+        req.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(req, video_id)
+
+        video_title = self._html_search_regex(r'videotitle	="([^"]+)', webpage, u'title')
+        video_description = self._html_search_regex(r'>Description:</strong>(.+?)<', webpage, u'description', fatal=False)
+        video_uploader = self._html_search_regex(r'>Submitted by:</strong>(?:\s|<[^>]*>)*(.+?)<', webpage, u'uploader', fatal=False)
+        thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, u'thumbnail', fatal=False)
+        if thumbnail:
+            thumbnail = thumbnail.replace('\\/', '/')
+
+        video_url = self._html_search_regex(r'"video_url":"([^"]+)', webpage, u'video_url')
+        if webpage.find('"encrypted":true')!=-1:
+            password = self._html_search_regex(r'"video_title":"([^"]+)', webpage, u'password')
+            video_url = aes_decrypt_text(video_url, password, 32).decode('utf-8')
+        path = compat_urllib_parse_urlparse(video_url).path
+        extension = os.path.splitext(path)[1][1:]
+        format = path.split('/')[4].split('_')[:2]
+        format = "-".join(format)
+
+        return {
+            'id': video_id,
+            'uploader': video_uploader,
+            'title': video_title,
+            'thumbnail': thumbnail,
+            'description': video_description,
+            'url': video_url,
+            'ext': extension,
+            'format': format,
+            'format_id': format,
+            'age_limit': 18,
+        }

youtube_dl/extractor/tudou.py

@@ -7,15 +7,25 @@ from .common import InfoExtractor
 
 
 class TudouIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
-    _TEST = {
+    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs|albumplay)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
+    _TESTS = [{
         u'url': u'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
         u'file': u'159448201.f4v',
         u'md5': u'140a49ed444bd22f93330985d8475fcb',
         u'info_dict': {
             u"title": u"卡马乔国足开大脚长传冲吊集锦"
         }
-    }
+    },
+    {
+        u'url': u'http://www.tudou.com/albumplay/TenTw_JgiPM/PzsAs5usU9A.html',
+        u'file': u'todo.mp4',
+        u'md5': u'todo.mp4',
+        u'info_dict': {
+            u'title': u'todo.mp4',
+        },
+        u'add_ie': [u'Youku'],
+        u'skip': u'Only works from China'
+    }]
 
     def _url_for_id(self, id, quality = None):
         info_url = "http://v2.tudou.com/f?id="+str(id)
@@ -29,14 +39,19 @@ class TudouIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(2)
         webpage = self._download_webpage(url, video_id)
-        title = re.search(",kw:\"(.+)\"",webpage)
-        if title is None:
-            title = re.search(",kw: \'(.+)\'",webpage)
-        title = title.group(1)
-        thumbnail_url = re.search(",pic: \'(.+?)\'",webpage)
-        if thumbnail_url is None:
-            thumbnail_url = re.search(",pic:\"(.+?)\"",webpage)
-        thumbnail_url = thumbnail_url.group(1)
+
+        m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage)
+        if m and m.group(1):
+            return {
+                '_type': 'url',
+                'url': u'youku:' + m.group(1),
+                'ie_key': 'Youku'
+            }
+
+        title = self._search_regex(
+            r",kw:\s*['\"](.+?)[\"']", webpage, u'title')
+        thumbnail_url = self._search_regex(
+            r",pic:\s*[\"'](.+?)[\"']", webpage, u'thumbnail URL', fatal=False)
 
         segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments')
         segments = json.loads(segs_json)

youtube_dl/extractor/tvp.py

@@ -0,0 +1,41 @@
+import json
+import re
+
+from .common import InfoExtractor
+
+
+class TvpIE(InfoExtractor):
+    IE_NAME = u'tvp.pl'
+    _VALID_URL = r'https?://www\.tvp\.pl/.*?wideo/(?P<date>\d+)/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.tvp.pl/warszawa/magazyny/campusnews/wideo/31102013/12878238',
+        u'md5': u'148408967a6a468953c0a75cbdaf0d7a',
+        u'file': u'12878238.wmv',
+        u'info_dict': {
+            u'title': u'31.10.2013',
+            u'description': u'31.10.2013',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        json_url = 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id
+        json_params = self._download_webpage(
+            json_url, video_id, u"Downloading video metadata")
+
+        params = json.loads(json_params)
+        self.report_extraction(video_id)
+        video_url = params['video_url']
+
+        title = self._og_search_title(webpage, fatal=True)
+        return {
+            'id': video_id,
+            'title': title,
+            'ext': 'wmv',
+            'url': video_url,
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }

youtube_dl/extractor/vevo.py

@@ -1,53 +1,127 @@
 import re
 import json
+import xml.etree.ElementTree
+import datetime
 
 from .common import InfoExtractor
 from ..utils import (
+    compat_HTTPError,
     ExtractorError,
 )
 
+
 class VevoIE(InfoExtractor):
     """
     Accepts urls from vevo.com or in the format 'vevo:{id}'
     (currently used by MTVIE)
     """
     _VALID_URL = r'((http://www.vevo.com/watch/.*?/.*?/)|(vevo:))(?P<id>.*?)(\?|$)'
-    _TEST = {
+    _TESTS = [{
         u'url': u'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
         u'file': u'GB1101300280.mp4',
-        u'md5': u'06bea460acb744eab74a9d7dcb4bfd61',
+        u"md5": u"06bea460acb744eab74a9d7dcb4bfd61",
         u'info_dict': {
             u"upload_date": u"20130624",
             u"uploader": u"Hurts",
-            u"title": u"Somebody to Die For"
-        }
+            u"title": u"Somebody to Die For",
+            u"duration": 230,
+            u"width": 1920,
+            u"height": 1080,
         }
+    }]
+    _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/'
+
+    def _formats_from_json(self, video_info):
+        last_version = {'version': -1}
+        for version in video_info['videoVersions']:
+            # These are the HTTP downloads, other types are for different manifests
+            if version['sourceType'] == 2:
+                if version['version'] > last_version['version']:
+                    last_version = version
+        if last_version['version'] == -1:
+            raise ExtractorError(u'Unable to extract last version of the video')
+
+        renditions = xml.etree.ElementTree.fromstring(last_version['data'])
+        formats = []
+        # Already sorted from worst to best quality
+        for rend in renditions.findall('rendition'):
+            attr = rend.attrib
+            format_note = '%(videoCodec)s@%(videoBitrate)4sk, %(audioCodec)s@%(audioBitrate)3sk' % attr
+            formats.append({
+                'url': attr['url'],
+                'format_id': attr['name'],
+                'format_note': format_note,
+                'height': int(attr['frameheight']),
+                'width': int(attr['frameWidth']),
+            })
+        return formats
+
+    def _formats_from_smil(self, smil_xml):
+        formats = []
+        smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8'))
+        els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
+        for el in els:
+            src = el.attrib['src']
+            m = re.match(r'''(?xi)
+                (?P<ext>[a-z0-9]+):
+                (?P<path>
+                    [/a-z0-9]+     # The directory and main part of the URL
+                    _(?P<cbr>[0-9]+)k
+                    _(?P<width>[0-9]+)x(?P<height>[0-9]+)
+                    _(?P<vcodec>[a-z0-9]+)
+                    _(?P<vbr>[0-9]+)
+                    _(?P<acodec>[a-z0-9]+)
+                    _(?P<abr>[0-9]+)
+                    \.[a-z0-9]+  # File extension
+                )''', src)
+            if not m:
+                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,
+                'ext': m.group('ext'),
+                'width': int(m.group('width')),
+                'height': int(m.group('height')),
+            })
+        return formats
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
 
-        json_url = 'http://www.vevo.com/data/video/%s' % video_id
-        base_url = 'http://smil.lvl3.vevo.com'
-        videos_url = '%s/Video/V2/VFILE/%s/%sr.smil' % (base_url, video_id, video_id.lower())
+        json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
         info_json = self._download_webpage(json_url, video_id, u'Downloading json info')
-        links_webpage = self._download_webpage(videos_url, video_id, u'Downloading videos urls')
+        video_info = json.loads(info_json)['video']
 
-        self.report_extraction(video_id)
-        video_info = json.loads(info_json)
-        m_urls = list(re.finditer(r'<video src="(?P<ext>.*?):/?(?P<url>.*?)"', links_webpage))
-        if m_urls is None or len(m_urls) == 0:
-            raise ExtractorError(u'Unable to extract video url')
-        # They are sorted from worst to best quality
-        m_url = m_urls[-1]
-        video_url = base_url + '/' + m_url.group('url')
-        ext = m_url.group('ext')
+        formats = self._formats_from_json(video_info)
+        try:
+            smil_url = '%s/Video/V2/VFILE/%s/%sr.smil' % (
+                self._SMIL_BASE_URL, video_id, video_id.lower())
+            smil_xml = self._download_webpage(smil_url, video_id,
+                                              u'Downloading SMIL info')
+            formats.extend(self._formats_from_smil(smil_xml))
+        except ExtractorError as ee:
+            if not isinstance(ee.cause, compat_HTTPError):
+                raise
+            self._downloader.report_warning(
+                u'Cannot download SMIL information, falling back to JSON ..')
 
-        return {'url': video_url,
-                'ext': ext,
+        timestamp_ms = int(self._search_regex(
+            r'/Date\((\d+)\)/', video_info['launchDate'], u'launch date'))
+        upload_date = datetime.datetime.fromtimestamp(timestamp_ms // 1000)
+        info = {
             'id': video_id,
             'title': video_info['title'],
-                'thumbnail': video_info['img'],
-                'upload_date': video_info['launchDate'].replace('/',''),
-                'uploader': video_info['Artists'][0]['title'],
+            'formats': formats,
+            'thumbnail': video_info['imageUrl'],
+            'upload_date': upload_date.strftime('%Y%m%d'),
+            'uploader': video_info['mainArtists'][0]['artistName'],
+            'duration': video_info['duration'],
         }
+
+        return info

youtube_dl/extractor/viddler.py

@@ -0,0 +1,64 @@
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+)
+
+
+class ViddlerIE(InfoExtractor):
+    _VALID_URL = r'(?P<domain>https?://(?:www\.)?viddler.com)/(?:v|embed|player)/(?P<id>[a-z0-9]+)'
+    _TEST = {
+        u"url": u"http://www.viddler.com/v/43903784",
+        u'file': u'43903784.mp4',
+        u'md5': u'fbbaedf7813e514eb7ca30410f439ac9',
+        u'info_dict': {
+            u"title": u"Video Made Easy",
+            u"uploader": u"viddler",
+            u"duration": 100.89,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        embed_url = mobj.group('domain') + u'/embed/' + video_id
+        webpage = self._download_webpage(embed_url, video_id)
+
+        video_sources_code = self._search_regex(
+            r"(?ms)sources\s*:\s*(\{.*?\})", webpage, u'video URLs')
+        video_sources = json.loads(video_sources_code.replace("'", '"'))
+
+        formats = [{
+            'url': video_url,
+            'format': format_id,
+        } for video_url, format_id in video_sources.items()]
+
+        title = self._html_search_regex(
+            r"title\s*:\s*'([^']*)'", webpage, u'title')
+        uploader = self._html_search_regex(
+            r"authorName\s*:\s*'([^']*)'", webpage, u'uploader', fatal=False)
+        duration_s = self._html_search_regex(
+            r"duration\s*:\s*([0-9.]*)", webpage, u'duration', fatal=False)
+        duration = float(duration_s) if duration_s else None
+        thumbnail = self._html_search_regex(
+            r"thumbnail\s*:\s*'([^']*)'",
+            webpage, u'thumbnail', fatal=False)
+
+        info = {
+            '_type': 'video',
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'duration': duration,
+            'formats': formats,
+        }
+
+        # TODO: Remove when #980 has been merged
+        info['formats'][-1]['ext'] = determine_ext(info['formats'][-1]['url'])
+        info.update(info['formats'][-1])
+
+        return info

youtube_dl/extractor/videodetective.py

@@ -0,0 +1,30 @@
+import re
+
+from .common import InfoExtractor
+from .internetvideoarchive import InternetVideoArchiveIE
+from ..utils import (
+    compat_urlparse,
+)
+
+
+class VideoDetectiveIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.videodetective\.com/[^/]+/[^/]+/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://www.videodetective.com/movies/kick-ass-2/194487',
+        u'file': u'194487.mp4',
+        u'info_dict': {
+            u'title': u'KICK-ASS 2',
+            u'description': u'md5:65ba37ad619165afac7d432eaded6013',
+            u'duration': 135,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        og_video = self._og_search_video_url(webpage)
+        query = compat_urlparse.urlparse(og_video).query
+        return self.url_result(InternetVideoArchiveIE._build_url(query),
+            ie=InternetVideoArchiveIE.ie_key())

youtube_dl/extractor/videopremium.py

@@ -0,0 +1,40 @@
+import re
+import random
+
+from .common import InfoExtractor
+
+
+class VideoPremiumIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:www\.)?videopremium\.tv/(?P<id>\w+)(?:/.*)?'
+    _TEST = {
+        u'url': u'http://videopremium.tv/4w7oadjsf156',
+        u'file': u'4w7oadjsf156.f4v',
+        u'info_dict': {
+            u"title": u"youtube-dl_test_video____a_________-BaW_jenozKc.mp4.mp4"
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        webpage_url = 'http://videopremium.tv/' + video_id
+        webpage = self._download_webpage(webpage_url, video_id)
+
+        self.report_extraction(video_id)
+
+        video_title = self._html_search_regex(r'<h2(?:.*?)>\s*(.+?)\s*<',
+            webpage, u'video title')
+
+        return [{
+            'id':          video_id,
+            'url':         "rtmp://e%d.md.iplay.md/play" % random.randint(1, 16),
+            'play_path':   "mp4:%s.f4v" % video_id,
+            'page_url':    "http://videopremium.tv/" + video_id,
+            'player_url':  "http://videopremium.tv/uplayer/uppod.swf",
+            'ext':         'f4v',
+            'title':       video_title,
+        }]

youtube_dl/extractor/vimeo.py

@@ -1,3 +1,4 @@
+# encoding: utf-8
 import json
 import re
 import itertools
@@ -10,19 +11,21 @@ from ..utils import (
     clean_html,
     get_element_by_attribute,
     ExtractorError,
+    RegexNotFoundError,
     std_headers,
+    unsmuggle_url,
 )
 
 class VimeoIE(InfoExtractor):
     """Information extractor for vimeo.com."""
 
     # _VALID_URL matches Vimeo URLs
-    _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)(?:[?].*)?$'
+    _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|(?P<player>player))\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)/?(?:[?].*)?(?:#.*)?$'
     _NETRC_MACHINE = 'vimeo'
     IE_NAME = u'vimeo'
     _TESTS = [
         {
-            u'url': u'http://vimeo.com/56015672',
+            u'url': u'http://vimeo.com/56015672#at=0',
             u'file': u'56015672.mp4',
             u'md5': u'8879b6cc097e987f02484baf890129e5',
             u'info_dict': {
@@ -54,6 +57,21 @@ class VimeoIE(InfoExtractor):
                 u'uploader': u'The BLN & Business of Software',
             },
         },
+        {
+            u'url': u'http://vimeo.com/68375962',
+            u'file': u'68375962.mp4',
+            u'md5': u'aaf896bdb7ddd6476df50007a0ac0ae7',
+            u'note': u'Video protected with password',
+            u'info_dict': {
+                u'title': u'youtube-dl password protected test video',
+                u'upload_date': u'20130614',
+                u'uploader_id': u'user18948128',
+                u'uploader': u'Jaime Marquínez Ferrándiz',
+            },
+            u'params': {
+                u'videopassword': u'youtube-dl',
+            },
+        },
     ]
 
     def _login(self):
@@ -98,21 +116,25 @@ class VimeoIE(InfoExtractor):
         self._login()
 
     def _real_extract(self, url, new_video=True):
+        url, data = unsmuggle_url(url)
+        headers = std_headers
+        if data is not None:
+            headers = headers.copy()
+            headers.update(data)
+
         # Extract ID from URL
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
 
         video_id = mobj.group('id')
-        if not mobj.group('proto'):
-            url = 'https://' + url
-        elif mobj.group('pro'):
+        if mobj.group('pro') or mobj.group('player'):
             url = 'http://player.vimeo.com/video/' + video_id
-        elif mobj.group('direct_link'):
+        else:
             url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information
-        request = compat_urllib_request.Request(url, None, std_headers)
+        request = compat_urllib_request.Request(url, None, headers)
         webpage = self._download_webpage(request, video_id)
 
         # Now we begin extracting as much information as we can from what we
@@ -122,18 +144,26 @@ class VimeoIE(InfoExtractor):
 
         # Extract the config JSON
         try:
+            try:
+                config_url = self._html_search_regex(
+                    r' data-config-url="(.+?)"', webpage, u'config URL')
+                config_json = self._download_webpage(config_url, video_id)
+                config = json.loads(config_json)
+            except RegexNotFoundError:
+                # For pro videos or player.vimeo.com urls
                 config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'],
                     webpage, u'info section', flags=re.DOTALL)
                 config = json.loads(config)
-        except:
+        except Exception as e:
             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
                 raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option')
 
-            if re.search('If so please provide the correct password.', webpage):
+            if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
                 self._verify_video_password(url, video_id, webpage)
                 return self._real_extract(url)
             else:
-                raise ExtractorError(u'Unable to extract info section')
+                raise ExtractorError(u'Unable to extract info section',
+                                     cause=e)
 
         # Extract title
         video_title = config["video"]["title"]
@@ -172,47 +202,47 @@ class VimeoIE(InfoExtractor):
 
         # Vimeo specific: extract video codec and quality information
         # First consider quality, then codecs, then take everything
-        # TODO bind to format param
-        codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
+        codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
         files = {'hd': [], 'sd': [], 'other': []}
         config_files = config["video"].get("files") or config["request"].get("files")
         for codec_name, codec_extension in codecs:
-            if codec_name in config_files:
-                if 'hd' in config_files[codec_name]:
-                    files['hd'].append((codec_name, codec_extension, 'hd'))
-                elif 'sd' in config_files[codec_name]:
-                    files['sd'].append((codec_name, codec_extension, 'sd'))
-                else:
-                    files['other'].append((codec_name, codec_extension, config_files[codec_name][0]))
-
-        for quality in ('hd', 'sd', 'other'):
-            if len(files[quality]) > 0:
-                video_quality = files[quality][0][2]
-                video_codec = files[quality][0][0]
-                video_extension = files[quality][0][1]
-                self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality))
-                break
-        else:
-            raise ExtractorError(u'No known codec found')
-
+            for quality in config_files.get(codec_name, []):
+                format_id = '-'.join((codec_name, quality)).lower()
+                key = quality if quality in files else 'other'
                 video_url = None
-        if isinstance(config_files[video_codec], dict):
-            video_url = config_files[video_codec][video_quality].get("url")
+                if isinstance(config_files[codec_name], dict):
+                    file_info = config_files[codec_name][quality]
+                    video_url = file_info.get('url')
+                else:
+                    file_info = {}
                 if video_url is None:
                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
-                        %(video_id, sig, timestamp, video_quality, video_codec.upper())
+                        %(video_id, sig, timestamp, quality, codec_name.upper())
 
-        return [{
-            'id':       video_id,
+                files[key].append({
+                    'ext': codec_extension,
                     'url': video_url,
+                    'format_id': format_id,
+                    'width': file_info.get('width'),
+                    'height': file_info.get('height'),
+                })
+        formats = []
+        for key in ('other', 'sd', 'hd'):
+            formats += files[key]
+        if len(formats) == 0:
+            raise ExtractorError(u'No known codec found')
+
+        return {
+            'id':       video_id,
             'uploader': video_uploader,
             'uploader_id': video_uploader_id,
             'upload_date':  video_upload_date,
             'title':    video_title,
-            'ext':      video_extension,
             'thumbnail':    video_thumbnail,
             'description':  video_description,
-        }]
+            'formats': formats,
+            'webpage_url': url,
+        }
 
 
 class VimeoChannelIE(InfoExtractor):

youtube_dl/extractor/vine.py

@@ -27,7 +27,7 @@ class VineIE(InfoExtractor):
         video_url = self._html_search_regex(r'<meta property="twitter:player:stream" content="(.+?)"',
             webpage, u'video URL')
 
-        uploader = self._html_search_regex(r'<div class="user">.*?<h2>(.+?)</h2>',
+        uploader = self._html_search_regex(r'<p class="username">(.*?)</p>',
             webpage, u'uploader', fatal=False, flags=re.DOTALL)
 
         return [{

youtube_dl/extractor/vk.py

@@ -0,0 +1,45 @@
+# encoding: utf-8
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_str,
+    unescapeHTML,
+)
+
+
+class VKIE(InfoExtractor):
+    IE_NAME = u'vk.com'
+    _VALID_URL = r'https?://vk\.com/(?:videos.*?\?.*?z=)?video(?P<id>.*?)(?:\?|%2F|$)'
+
+    _TEST = {
+        u'url': u'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
+        u'md5': u'0deae91935c54e00003c2a00646315f0',
+        u'info_dict': {
+            u'id': u'162222515',
+            u'ext': u'flv',
+            u'title': u'ProtivoGunz - Хуёвая песня',
+            u'uploader': u'Noize MC',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        info_url = 'http://vk.com/al_video.php?act=show&al=1&video=%s' % video_id
+        info_page = self._download_webpage(info_url, video_id)
+        m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page)
+        if m_yt is not None:
+            self.to_screen(u'Youtube video detected')
+            return self.url_result(m_yt.group(1), 'Youtube')
+        vars_json = self._search_regex(r'var vars = ({.*?});', info_page, u'vars')
+        vars = json.loads(vars_json)
+
+        return {
+            'id': compat_str(vars['vid']),
+            'url': vars['url240'],
+            'title': unescapeHTML(vars['md_title']),
+            'thumbnail': vars['jpg'],
+            'uploader': vars['md_author'],
+        }

youtube_dl/extractor/websurg.py

@@ -0,0 +1,59 @@
+# coding: utf-8
+
+import re
+
+from ..utils import (
+    compat_urllib_request,
+    compat_urllib_parse
+)
+
+from .common import InfoExtractor
+
+class WeBSurgIE(InfoExtractor):
+    IE_NAME = u'websurg.com'
+    _VALID_URL = r'http://.*?\.websurg\.com/MEDIA/\?noheader=1&doi=(.*)'
+
+    _TEST = {
+        u'url': u'http://www.websurg.com/MEDIA/?noheader=1&doi=vd01en4012',
+        u'file': u'vd01en4012.mp4',
+        u'params': {
+            u'skip_download': True,
+        },
+        u'skip': u'Requires login information',
+    }
+    
+    _LOGIN_URL = 'http://www.websurg.com/inc/login/login_div.ajax.php?login=1'
+
+    def _real_initialize(self):
+
+        login_form = {
+            'username': self._downloader.params['username'],
+            'password': self._downloader.params['password'],
+            'Submit': 1
+        }
+        
+        request = compat_urllib_request.Request(
+            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
+        request.add_header(
+            'Content-Type', 'application/x-www-form-urlencoded;charset=utf-8')
+        compat_urllib_request.urlopen(request).info()
+        webpage = self._download_webpage(self._LOGIN_URL, '', 'Logging in')
+        
+        if webpage != 'OK':
+            self._downloader.report_error(
+                u'Unable to log in: bad username/password')
+        
+    def _real_extract(self, url):
+        video_id = re.match(self._VALID_URL, url).group(1)
+        
+        webpage = self._download_webpage(url, video_id)
+        
+        url_info = re.search(r'streamer="(.*?)" src="(.*?)"', webpage)
+        
+        return {'id': video_id,
+                'title': self._og_search_title(webpage),
+                'description': self._og_search_description(webpage),
+                'ext' : 'mp4',
+                'url' : url_info.group(1) + '/' + url_info.group(2),
+                'thumbnail': self._og_search_thumbnail(webpage)
+                }

youtube_dl/extractor/weibo.py

@@ -13,6 +13,7 @@ class WeiboIE(InfoExtractor):
     _VALID_URL = r'https?://video\.weibo\.com/v/weishipin/t_(?P<id>.+?)\.htm'
 
     _TEST = {
+        u'add_ie': ['Sina'],
         u'url': u'http://video.weibo.com/v/weishipin/t_zjUw2kZ.htm',
         u'file': u'98322879.flv',
         u'info_dict': {

youtube_dl/extractor/xhamster.py

@@ -19,7 +19,8 @@ class XHamsterIE(InfoExtractor):
         u'info_dict': {
             u"upload_date": u"20121014", 
             u"uploader_id": u"Ruseful2011", 
-            u"title": u"FemaleAgent Shy beauty takes the bait"
+            u"title": u"FemaleAgent Shy beauty takes the bait",
+            u"age_limit": 18,
         }
     },
     {
@@ -29,25 +30,30 @@ class XHamsterIE(InfoExtractor):
         u'info_dict': {
             u"upload_date": u"20130914",
             u"uploader_id": u"jojo747400",
-            u"title": u"Britney Spears  Sexy Booty"
+            u"title": u"Britney Spears  Sexy Booty",
+            u"age_limit": 18,
         }
     }]
 
     def _real_extract(self,url):
-        mobj = re.match(self._VALID_URL, url)
-
-        video_id = mobj.group('id')
-        seo = mobj.group('seo')
-        mrss_url = 'http://xhamster.com/movies/%s/%s.html?hd' % (video_id, seo)
-        webpage = self._download_webpage(mrss_url, video_id)
-
+        def extract_video_url(webpage):
             mobj = re.search(r'\'srv\': \'(?P<server>[^\']*)\',\s*\'file\': \'(?P<file>[^\']+)\',', webpage)
             if mobj is None:
                 raise ExtractorError(u'Unable to extract media URL')
             if len(mobj.group('server')) == 0:
-            video_url = compat_urllib_parse.unquote(mobj.group('file'))
+                return compat_urllib_parse.unquote(mobj.group('file'))
             else:
-            video_url = mobj.group('server')+'/key='+mobj.group('file')
+                return mobj.group('server')+'/key='+mobj.group('file')
+
+        def is_hd(webpage):
+            return webpage.find('<div class=\'icon iconHD\'>') != -1
+
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        seo = mobj.group('seo')
+        mrss_url = 'http://xhamster.com/movies/%s/%s.html' % (video_id, seo)
+        webpage = self._download_webpage(mrss_url, video_id)
 
         video_title = self._html_search_regex(r'<title>(?P<title>.+?) - xHamster\.com</title>',
             webpage, u'title')
@@ -72,13 +78,34 @@ class XHamsterIE(InfoExtractor):
         video_thumbnail = self._search_regex(r'\'image\':\'(?P<thumbnail>[^\']+)\'',
             webpage, u'thumbnail', fatal=False)
 
-        return [{
-            'id':       video_id,
+        age_limit = self._rta_search(webpage)
+
+        video_url = extract_video_url(webpage)
+        hd = is_hd(webpage)
+        formats = [{
             'url': video_url,
             'ext': determine_ext(video_url),
+            'format': 'hd' if hd else 'sd',
+            'format_id': 'hd' if hd else 'sd',
+        }]
+        if not hd:
+            webpage = self._download_webpage(mrss_url+'?hd', video_id)
+            if is_hd(webpage):
+                video_url = extract_video_url(webpage)
+                formats.append({
+                    'url': video_url,
+                    'ext': determine_ext(video_url),
+                    'format': 'hd',
+                    'format_id': 'hd',
+                })
+
+        return {
+            'id': video_id,
             'title': video_title,
+            'formats': formats,
             'description': video_description,
             'upload_date': video_upload_date,
             'uploader_id': video_uploader_id,
-            'thumbnail': video_thumbnail
-        }]
+            'thumbnail': video_thumbnail,
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/xnxx.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class XNXXIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?video\.xnxx\.com/video([0-9]+)/(.*)'
+    _VALID_URL = r'^(?:https?://)?(?:video|www)\.xnxx\.com/video([0-9]+)/(.*)'
     VIDEO_URL_RE = r'flv_url=(.*?)&'
     VIDEO_TITLE_RE = r'<title>(.*?)\s+-\s+XNXX.COM'
     VIDEO_THUMB_RE = r'url_bigthumb=(.*?)&amp;'
@@ -18,7 +18,8 @@ class XNXXIE(InfoExtractor):
         u'file': u'1135332.flv',
         u'md5': u'0831677e2b4761795f68d417e0b7b445',
         u'info_dict': {
-            u"title": u"lida \u00bb Naked Funny Actress  (5)"
+            u"title": u"lida \u00bb Naked Funny Actress  (5)",
+            u"age_limit": 18,
         }
     }
 
@@ -50,4 +51,5 @@ class XNXXIE(InfoExtractor):
             'ext': 'flv',
             'thumbnail': video_thumbnail,
             'description': None,
+            'age_limit': 18,
         }]

youtube_dl/extractor/xtube.py

@@ -0,0 +1,55 @@
+import os
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
+    compat_urllib_parse,
+)
+
+class XTubeIE(InfoExtractor):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<videoid>[^/?&]+))'
+    _TEST = {
+        u'url': u'http://www.xtube.com/watch.php?v=kVTUy_G222_',
+        u'file': u'kVTUy_G222_.mp4',
+        u'md5': u'092fbdd3cbe292c920ef6fc6a8a9cdab',
+        u'info_dict': {
+            u"title": u"strange erotica",
+            u"description": u"surreal gay themed erotica...almost an ET kind of thing",
+            u"uploader": u"greenshowers",
+            u"age_limit": 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+        url = 'http://www.' + mobj.group('url')
+
+        req = compat_urllib_request.Request(url)
+        req.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(req, video_id)
+
+        video_title = self._html_search_regex(r'<div class="p_5px[^>]*>([^<]+)', webpage, u'title')
+        video_uploader = self._html_search_regex(r'so_s\.addVariable\("owner_u", "([^"]+)', webpage, u'uploader', fatal=False)
+        video_description = self._html_search_regex(r'<p class="video_description">([^<]+)', webpage, u'description', default=None)
+        video_url= self._html_search_regex(r'var videoMp4 = "([^"]+)', webpage, u'video_url').replace('\\/', '/')
+        path = compat_urllib_parse_urlparse(video_url).path
+        extension = os.path.splitext(path)[1][1:]
+        format = path.split('/')[5].split('_')[:2]
+        format[0] += 'p'
+        format[1] += 'k'
+        format = "-".join(format)
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'uploader': video_uploader,
+            'description': video_description,
+            'url': video_url,
+            'ext': extension,
+            'format': format,
+            'format_id': format,
+            'age_limit': 18,
+        }

youtube_dl/extractor/xvideos.py

@@ -13,7 +13,8 @@ class XVideosIE(InfoExtractor):
         u'file': u'939581.flv',
         u'md5': u'1d0c835822f0a71a7bf011855db929d0',
         u'info_dict': {
-            u"title": u"Funny Porns By >>>>S<<<<<< -1"
+            u"title": u"Funny Porns By >>>>S<<<<<< -1",
+            u"age_limit": 18,
         }
     }
 
@@ -46,6 +47,7 @@ class XVideosIE(InfoExtractor):
             'ext': 'flv',
             'thumbnail': video_thumbnail,
             'description': None,
+            'age_limit': 18,
         }
 
         return [info]

youtube_dl/extractor/yahoo.py

@@ -1,4 +1,3 @@
-import datetime
 import itertools
 import json
 import re
@@ -6,86 +5,104 @@ import re
 from .common import InfoExtractor, SearchInfoExtractor
 from ..utils import (
     compat_urllib_parse,
-
-    ExtractorError,
+    compat_urlparse,
+    determine_ext,
+    clean_html,
 )
 
+
 class YahooIE(InfoExtractor):
     IE_DESC = u'Yahoo screen'
     _VALID_URL = r'http://screen\.yahoo\.com/.*?-(?P<id>\d*?)\.html'
-    _TEST = {
+    _TESTS = [
+        {
             u'url': u'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
             u'file': u'214727115.flv',
-        u'md5': u'2e717f169c1be93d84d3794a00d4a325',
             u'info_dict': {
-            u"title": u"Julian Smith & Travis Legg Watch Julian Smith"
+                u'title': u'Julian Smith & Travis Legg Watch Julian Smith',
+                u'description': u'Julian and Travis watch Julian Smith',
             },
-        u'skip': u'Requires rtmpdump'
-    }
+            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'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,
+            },
+        },
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
-        m_id = re.search(r'YUI\.namespace\("Media"\)\.CONTENT_ID = "(?P<new_id>.+?)";', webpage)
 
-        if m_id is None: 
-            # TODO: Check which url parameters are required
-            info_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;lmsoverride=1;outputformat=mrss;cb=974419660;id=%s;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
-            webpage = self._download_webpage(info_url, video_id, u'Downloading info webpage')
-            info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
-                        <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
-                        <media:pubStart><!\[CDATA\[(?P<date>.*?)\ .*\]\]></media:pubStart>.*
-                        <media:content\ medium="image"\ url="(?P<thumb>.*?)"\ name="LARGETHUMB"
-                        '''
-            self.report_extraction(video_id)
-            m_info = re.search(info_re, webpage, re.VERBOSE|re.DOTALL)
-            if m_info is None:
-                raise ExtractorError(u'Unable to extract video info')
-            video_title = m_info.group('title')
-            video_description = m_info.group('description')
-            video_thumb = m_info.group('thumb')
-            video_date = m_info.group('date')
-            video_date = datetime.datetime.strptime(video_date, '%m/%d/%Y').strftime('%Y%m%d')
+        items_json = self._search_regex(r'YVIDEO_INIT_ITEMS = ({.*?});$',
+            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']
+        query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"'
+                 ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2"' % long_id)
+        data = compat_urllib_parse.urlencode({
+            'q': query,
+            'env': 'prod',
+            'format': 'json',
+        })
+        query_result_json = self._download_webpage(
+            'http://video.query.yahoo.com/v1/public/yql?' + data,
+            video_id, u'Downloading video info')
+        query_result = json.loads(query_result_json)
+        info = query_result['query']['results']['mediaObj'][0]
+        meta = info['meta']
 
-            # TODO: Find a way to get mp4 videos
-            rest_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;element=stream;outputformat=mrss;id=%s;lmsoverride=1;bw=375;dynamicstream=1;cb=83521105;tech=flv,mp4;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
-            webpage = self._download_webpage(rest_url, video_id, u'Downloading video url webpage')
-            m_rest = re.search(r'<media:content url="(?P<url>.*?)" path="(?P<path>.*?)"', webpage)
-            video_url = m_rest.group('url')
-            video_path = m_rest.group('path')
-            if m_rest is None:
-                raise ExtractorError(u'Unable to extract video url')
-
-        else: # We have to use a different method if another id is defined
-            long_id = m_id.group('new_id')
-            info_url = 'http://video.query.yahoo.com/v1/public/yql?q=SELECT%20*%20FROM%20yahoo.media.video.streams%20WHERE%20id%3D%22' + long_id + '%22%20AND%20format%3D%22mp4%2Cflv%22%20AND%20protocol%3D%22rtmp%2Chttp%22%20AND%20plrs%3D%2286Gj0vCaSzV_Iuf6hNylf2%22%20AND%20acctid%3D%22389%22%20AND%20plidl%3D%22%22%20AND%20pspid%3D%22792700001%22%20AND%20offnetwork%3D%22false%22%20AND%20site%3D%22ivy%22%20AND%20lang%3D%22en-US%22%20AND%20region%3D%22US%22%20AND%20override%3D%22none%22%3B&env=prod&format=json&callback=YUI.Env.JSONP.yui_3_8_1_1_1368368376830_335'
-            webpage = self._download_webpage(info_url, video_id, u'Downloading info json')
-            json_str = re.search(r'YUI.Env.JSONP.yui.*?\((.*?)\);', webpage).group(1)
-            info = json.loads(json_str)
-            res = info[u'query'][u'results'][u'mediaObj'][0]
-            stream = res[u'streams'][0]
-            video_path = stream[u'path']
-            video_url = stream[u'host']
-            meta = res[u'meta']
-            video_title = meta[u'title']
-            video_description = meta[u'description']
-            video_thumb = meta[u'thumbnail']
-            video_date = None # I can't find it
-
-        info_dict = {
-                     'id': video_id,
-                     'url': video_url,
-                     'play_path': video_path,
-                     'title':video_title,
-                     'description': video_description,
-                     'thumbnail': video_thumb,
-                     'upload_date': video_date,
-                     'ext': 'flv',
+        formats = []
+        for s in info['streams']:
+            format_info = {
+                'width': s.get('width'),
+                'height': s.get('height'),
+                'bitrate': s.get('bitrate'),
             }
-        return info_dict
+
+            host = s['host']
+            path = s['path']
+            if host.startswith('rtmp'):
+                format_info.update({
+                    'url': host,
+                    'play_path': path,
+                    'ext': 'flv',
+                })
+            else:
+                format_url = compat_urlparse.urljoin(host, path)
+                format_info['url'] = format_url
+                format_info['ext'] = determine_ext(format_url)
+                
+            formats.append(format_info)
+        formats = sorted(formats, key=lambda f:(f['height'], f['width']))
+
+        info = {
+            '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 YahooSearchIE(SearchInfoExtractor):
     IE_DESC = u'Yahoo screen search'

youtube_dl/extractor/youjizz.py

@@ -13,7 +13,8 @@ class YouJizzIE(InfoExtractor):
         u'file': u'2189178.flv',
         u'md5': u'07e15fa469ba384c7693fd246905547c',
         u'info_dict': {
-            u"title": u"Zeichentrick 1"
+            u"title": u"Zeichentrick 1",
+            u"age_limit": 18,
         }
     }
 
@@ -25,6 +26,8 @@ class YouJizzIE(InfoExtractor):
         # Get webpage content
         webpage = self._download_webpage(url, video_id)
 
+        age_limit = self._rta_search(webpage)
+
         # Get the video title
         video_title = self._html_search_regex(r'<title>(?P<title>.*)</title>',
             webpage, u'title').strip()
@@ -60,6 +63,7 @@ class YouJizzIE(InfoExtractor):
                 'title': video_title,
                 'ext': 'flv',
                 'format': 'flv',
-                'player_url': embed_page_url}
+                'player_url': embed_page_url,
+                'age_limit': age_limit}
 
         return [info]