release 2015.11.13

For providing idea for vidto.me (#7167) and extending nowvideo support (#6760)

AUTHORS

@@ -144,3 +144,5 @@ Lee Jenkins
 Anssi Hannula
 Lukáš Lalinský
 Qijiang Fan
+Rémy Léone
+Marco Ferragina

CONTRIBUTING.md

@@ -1,6 +1,6 @@
 **Please include the full output of youtube-dl when run with `-v`**.
 
-The output (including the first lines) contain important debugging information. Issues without the full output are often not reproducible and therefore do not get solved in short order, if ever.
+The output (including the first lines) contains important debugging information. Issues without the full output are often not reproducible and therefore do not get solved in short order, if ever.
 
 Please re-read your issue once again to avoid a couple of common mistakes (you can and should use this as a checklist):
 

README.md

@@ -795,7 +795,7 @@ Bugs and suggestions should be reported at: <https://github.com/rg3/youtube-dl/i
 
 **Please include the full output of youtube-dl when run with `-v`**.
 
-The output (including the first lines) contain important debugging information. Issues without the full output are often not reproducible and therefore do not get solved in short order, if ever.
+The output (including the first lines) contains important debugging information. Issues without the full output are often not reproducible and therefore do not get solved in short order, if ever.
 
 Please re-read your issue once again to avoid a couple of common mistakes (you can and should use this as a checklist):
 

docs/supportedsites.md

@@ -93,6 +93,7 @@
  - **Clipsyndicate**
  - **Cloudy**
  - **Clubic**
+ - **Clyp**
  - **cmt.com**
  - **CNET**
  - **CNN**
@@ -122,6 +123,7 @@
  - **DctpTv**
  - **DeezerPlaylist**
  - **defense.gouv.fr**
+ - **democracynow**
  - **DHM**: Filmarchiv - Deutsches Historisches Museum
  - **Discovery**
  - **Dotsub**
@@ -194,10 +196,10 @@
  - **Giga**
  - **Glide**: Glide mobile video messages (glide.me)
  - **Globo**
+ - **GloboArticle**
  - **GodTube**
  - **GoldenMoustache**
  - **Golem**
- - **GorillaVid**: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net and filehoot.com
  - **Goshgay**
  - **Groupon**
  - **Hark**
@@ -281,7 +283,7 @@
  - **macgamestore**: MacGameStore trailers
  - **mailru**: Видео@Mail.Ru
  - **Malemotion**
- - **MDR**
+ - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
  - **metacafe**
  - **Metacritic**
@@ -616,7 +618,6 @@
  - **video.mit.edu**
  - **VideoDetective**
  - **videofy.me**
- - **videolectures.net**
  - **VideoMega**
  - **VideoPremium**
  - **VideoTt**: video.tt - Your True Tube
@@ -626,6 +627,7 @@
  - **vier**
  - **vier:videos**
  - **Viewster**
+ - **Viidea**
  - **viki**
  - **viki:channel**
  - **vimeo**
@@ -668,6 +670,7 @@
  - **WSJ**: Wall Street Journal
  - **XBef**
  - **XboxClips**
+ - **XFileShare**: XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me
  - **XHamster**
  - **XHamsterEmbed**
  - **XMinus**

setup.py

@@ -28,7 +28,7 @@ py2exe_options = {
     "compressed": 1,
     "optimize": 2,
     "dist_dir": '.',
-    "dll_excludes": ['w9xpopen.exe'],
+    "dll_excludes": ['w9xpopen.exe', 'crypt32.dll'],
 }
 
 py2exe_console = [{

test/test_compat.py

@@ -13,8 +13,10 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from youtube_dl.utils import get_filesystem_encoding
 from youtube_dl.compat import (
     compat_getenv,
+    compat_etree_fromstring,
     compat_expanduser,
     compat_shlex_split,
+    compat_str,
     compat_urllib_parse_unquote,
     compat_urllib_parse_unquote_plus,
 )
@@ -71,5 +73,20 @@ class TestCompat(unittest.TestCase):
     def test_compat_shlex_split(self):
         self.assertEqual(compat_shlex_split('-option "one two"'), ['-option', 'one two'])
 
+    def test_compat_etree_fromstring(self):
+        xml = '''
+            <root foo="bar" spam="中文">
+                <normal>foo</normal>
+                <chinese>中文</chinese>
+                <foo><bar>spam</bar></foo>
+            </root>
+        '''
+        doc = compat_etree_fromstring(xml.encode('utf-8'))
+        self.assertTrue(isinstance(doc.attrib['foo'], compat_str))
+        self.assertTrue(isinstance(doc.attrib['spam'], compat_str))
+        self.assertTrue(isinstance(doc.find('normal').text, compat_str))
+        self.assertTrue(isinstance(doc.find('chinese').text, compat_str))
+        self.assertTrue(isinstance(doc.find('foo/bar').text, compat_str))
+
 if __name__ == '__main__':
     unittest.main()

test/test_download.py

@@ -102,7 +102,7 @@ def generator(test_case):
 
         params = get_params(test_case.get('params', {}))
         if is_playlist and 'playlist' not in test_case:
-            params.setdefault('extract_flat', True)
+            params.setdefault('extract_flat', 'in_playlist')
             params.setdefault('skip_download', True)
 
         ydl = YoutubeDL(params, auto_init=False)

test/test_jsinterp.py

@@ -19,6 +19,9 @@ class TestJSInterpreter(unittest.TestCase):
         jsi = JSInterpreter('function x3(){return 42;}')
         self.assertEqual(jsi.call_function('x3'), 42)
 
+        jsi = JSInterpreter('var x5 = function(){return 42;}')
+        self.assertEqual(jsi.call_function('x5'), 42)
+
     def test_calc(self):
         jsi = JSInterpreter('function x4(a){return 2*a+1;}')
         self.assertEqual(jsi.call_function('x4', 3), 7)

test/test_subtitles.py

@@ -28,6 +28,7 @@ from youtube_dl.extractor import (
     ThePlatformFeedIE,
     RTVEALaCartaIE,
     FunnyOrDieIE,
+    DemocracynowIE,
 )
 
 
@@ -346,5 +347,25 @@ class TestFunnyOrDieSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), 'c5593c193eacd353596c11c2d4f9ecc4')
 
 
+class TestDemocracynowSubtitles(BaseTestSubtitles):
+    url = 'http://www.democracynow.org/shows/2015/7/3'
+    IE = DemocracynowIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), 'acaca989e24a9e45a6719c9b3d60815c')
+
+    def test_subtitles_in_page(self):
+        self.url = 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), 'acaca989e24a9e45a6719c9b3d60815c')
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_utils.py

@@ -68,6 +68,9 @@ from youtube_dl.utils import (
     cli_valueless_option,
     cli_bool_option,
 )
+from youtube_dl.compat import (
+    compat_etree_fromstring,
+)
 
 
 class TestUtil(unittest.TestCase):
@@ -233,6 +236,7 @@ class TestUtil(unittest.TestCase):
             unified_strdate('2/2/2015 6:47:40 PM', day_first=False),
             '20150202')
         self.assertEqual(unified_strdate('25-09-2014'), '20140925')
+        self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None)
 
     def test_find_xpath_attr(self):
         testxml = '''<root>
@@ -242,7 +246,7 @@ class TestUtil(unittest.TestCase):
             <node x="b" y="d" />
             <node x="" />
         </root>'''
-        doc = xml.etree.ElementTree.fromstring(testxml)
+        doc = compat_etree_fromstring(testxml)
 
         self.assertEqual(find_xpath_attr(doc, './/fourohfour', 'n'), None)
         self.assertEqual(find_xpath_attr(doc, './/fourohfour', 'n', 'v'), None)
@@ -263,7 +267,7 @@ class TestUtil(unittest.TestCase):
                 <url>http://server.com/download.mp3</url>
             </media:song>
         </root>'''
-        doc = xml.etree.ElementTree.fromstring(testxml)
+        doc = compat_etree_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, 'The Author')
@@ -275,9 +279,16 @@ class TestUtil(unittest.TestCase):
         p = xml.etree.ElementTree.SubElement(div, 'p')
         p.text = 'Foo'
         self.assertEqual(xpath_element(doc, 'div/p'), p)
+        self.assertEqual(xpath_element(doc, ['div/p']), p)
+        self.assertEqual(xpath_element(doc, ['div/bar', 'div/p']), p)
         self.assertEqual(xpath_element(doc, 'div/bar', default='default'), 'default')
+        self.assertEqual(xpath_element(doc, ['div/bar'], default='default'), 'default')
         self.assertTrue(xpath_element(doc, 'div/bar') is None)
+        self.assertTrue(xpath_element(doc, ['div/bar']) is None)
+        self.assertTrue(xpath_element(doc, ['div/bar'], 'div/baz') is None)
         self.assertRaises(ExtractorError, xpath_element, doc, 'div/bar', fatal=True)
+        self.assertRaises(ExtractorError, xpath_element, doc, ['div/bar'], fatal=True)
+        self.assertRaises(ExtractorError, xpath_element, doc, ['div/bar', 'div/baz'], fatal=True)
 
     def test_xpath_text(self):
         testxml = '''<root>
@@ -285,7 +296,7 @@ class TestUtil(unittest.TestCase):
                 <p>Foo</p>
             </div>
         </root>'''
-        doc = xml.etree.ElementTree.fromstring(testxml)
+        doc = compat_etree_fromstring(testxml)
         self.assertEqual(xpath_text(doc, 'div/p'), 'Foo')
         self.assertEqual(xpath_text(doc, 'div/bar', default='default'), 'default')
         self.assertTrue(xpath_text(doc, 'div/bar') is None)
@@ -297,7 +308,7 @@ class TestUtil(unittest.TestCase):
                 <p x="a">Foo</p>
             </div>
         </root>'''
-        doc = xml.etree.ElementTree.fromstring(testxml)
+        doc = compat_etree_fromstring(testxml)
         self.assertEqual(xpath_attr(doc, 'div/p', 'x'), 'a')
         self.assertEqual(xpath_attr(doc, 'div/bar', 'x'), None)
         self.assertEqual(xpath_attr(doc, 'div/p', 'y'), None)
@@ -425,6 +436,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26+0000'), 1395612266)
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26Z'), 1395612266)
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26.1234Z'), 1395612266)
+        self.assertEqual(parse_iso8601('2015-09-29T08:27:31.727'), 1443515251)
+        self.assertEqual(parse_iso8601('2015-09-29T08-27-31.727'), None)
 
     def test_strip_jsonp(self):
         stripped = strip_jsonp('cb ([ {"id":"532cb",\n\n\n"x":\n3}\n]\n);')

youtube_dl/YoutubeDL.py

@@ -572,7 +572,7 @@ class YoutubeDL(object):
                                  if v is not None)
             template_dict = collections.defaultdict(lambda: 'NA', template_dict)
 
-            outtmpl = sanitize_path(self.params.get('outtmpl', DEFAULT_OUTTMPL))
+            outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
             tmpl = compat_expanduser(outtmpl)
             filename = tmpl % template_dict
             # Temporary fix for #4787
@@ -580,7 +580,7 @@ class YoutubeDL(object):
             # to workaround encoding issues with subprocess on python2 @ Windows
             if sys.version_info < (3, 0) and sys.platform == 'win32':
                 filename = encodeFilename(filename, True).decode(preferredencoding())
-            return filename
+            return sanitize_path(filename)
         except ValueError as err:
             self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
             return None

youtube_dl/compat.py

@@ -14,6 +14,7 @@ import socket
 import subprocess
 import sys
 import itertools
+import xml.etree.ElementTree
 
 
 try:
@@ -212,6 +213,43 @@ try:
 except ImportError:  # Python 2.6
     from xml.parsers.expat import ExpatError as compat_xml_parse_error
 
+if sys.version_info[0] >= 3:
+    compat_etree_fromstring = xml.etree.ElementTree.fromstring
+else:
+    # python 2.x tries to encode unicode strings with ascii (see the
+    # XMLParser._fixtext method)
+    etree = xml.etree.ElementTree
+
+    try:
+        _etree_iter = etree.Element.iter
+    except AttributeError:  # Python <=2.6
+        def _etree_iter(root):
+            for el in root.findall('*'):
+                yield el
+                for sub in _etree_iter(el):
+                    yield sub
+
+    # on 2.6 XML doesn't have a parser argument, function copied from CPython
+    # 2.7 source
+    def _XML(text, parser=None):
+        if not parser:
+            parser = etree.XMLParser(target=etree.TreeBuilder())
+        parser.feed(text)
+        return parser.close()
+
+    def _element_factory(*args, **kwargs):
+        el = etree.Element(*args, **kwargs)
+        for k, v in el.items():
+            if isinstance(v, bytes):
+                el.set(k, v.decode('utf-8'))
+        return el
+
+    def compat_etree_fromstring(text):
+        doc = _XML(text, parser=etree.XMLParser(target=etree.TreeBuilder(element_factory=_element_factory)))
+        for el in _etree_iter(doc):
+            if el.text is not None and isinstance(el.text, bytes):
+                el.text = el.text.decode('utf-8')
+        return doc
 
 try:
     from urllib.parse import parse_qs as compat_parse_qs
@@ -507,6 +545,7 @@ __all__ = [
     'compat_chr',
     'compat_cookiejar',
     'compat_cookies',
+    'compat_etree_fromstring',
     'compat_expanduser',
     'compat_get_terminal_size',
     'compat_getenv',

youtube_dl/downloader/f4m.py

@@ -5,12 +5,13 @@ import io
 import itertools
 import os
 import time
-import xml.etree.ElementTree as etree
 
 from .fragment import FragmentFD
 from ..compat import (
+    compat_etree_fromstring,
     compat_urlparse,
     compat_urllib_error,
+    compat_urllib_parse_urlparse,
 )
 from ..utils import (
     encodeFilename,
@@ -285,9 +286,11 @@ class F4mFD(FragmentFD):
         man_url = info_dict['url']
         requested_bitrate = info_dict.get('tbr')
         self.to_screen('[%s] Downloading f4m manifest' % self.FD_NAME)
-        manifest = self.ydl.urlopen(man_url).read()
+        urlh = self.ydl.urlopen(man_url)
+        man_url = urlh.geturl()
+        manifest = urlh.read()
 
-        doc = etree.fromstring(manifest)
+        doc = compat_etree_fromstring(manifest)
         formats = [(int(f.attrib.get('bitrate', -1)), f)
                    for f in self._get_unencrypted_media(doc)]
         if requested_bitrate is None:
@@ -329,20 +332,25 @@ class F4mFD(FragmentFD):
         if not live:
             write_metadata_tag(dest_stream, metadata)
 
+        base_url_parsed = compat_urllib_parse_urlparse(base_url)
+
         self._start_frag_download(ctx)
 
         frags_filenames = []
         while fragments_list:
             seg_i, frag_i = fragments_list.pop(0)
             name = 'Seg%d-Frag%d' % (seg_i, frag_i)
-            url = base_url + name
+            query = []
+            if base_url_parsed.query:
+                query.append(base_url_parsed.query)
             if akamai_pv:
-                url += '?' + akamai_pv.strip(';')
+                query.append(akamai_pv.strip(';'))
             if info_dict.get('extra_param_to_segment_url'):
-                url += info_dict.get('extra_param_to_segment_url')
+                query.append(info_dict['extra_param_to_segment_url'])
+            url_parsed = base_url_parsed._replace(path=base_url_parsed.path + name, query='&'.join(query))
             frag_filename = '%s-%s' % (ctx['tmpfilename'], name)
             try:
-                success = ctx['dl'].download(frag_filename, {'url': url})
+                success = ctx['dl'].download(frag_filename, {'url': url_parsed.geturl()})
                 if not success:
                     return False
                 (down, frag_sanitized) = sanitize_open(frag_filename, 'rb')

youtube_dl/extractor/__init__.py

@@ -90,6 +90,7 @@ from .cliphunter import CliphunterIE
 from .clipsyndicate import ClipsyndicateIE
 from .cloudy import CloudyIE
 from .clubic import ClubicIE
+from .clyp import ClypIE
 from .cmt import CMTIE
 from .cnet import CNETIE
 from .cnn import (
@@ -123,6 +124,7 @@ from .dbtv import DBTVIE
 from .dcn import DCNIE
 from .dctp import DctpTvIE
 from .deezer import DeezerPlaylistIE
+from .democracynow import DemocracynowIE
 from .dfb import DFBIE
 from .dhm import DHMIE
 from .dotsub import DotsubIE
@@ -210,13 +212,15 @@ from .gfycat import GfycatIE
 from .giantbomb import GiantBombIE
 from .giga import GigaIE
 from .glide import GlideIE
-from .globo import GloboIE
+from .globo import (
+    GloboIE,
+    GloboArticleIE,
+)
 from .godtube import GodTubeIE
 from .goldenmoustache import GoldenMoustacheIE
 from .golem import GolemIE
 from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
-from .gorillavid import GorillaVidIE
 from .goshgay import GoshgayIE
 from .groupon import GrouponIE
 from .hark import HarkIE
@@ -719,7 +723,6 @@ from .vh1 import VH1IE
 from .vice import ViceIE
 from .viddler import ViddlerIE
 from .videodetective import VideoDetectiveIE
-from .videolecturesnet import VideoLecturesNetIE
 from .videofyme import VideofyMeIE
 from .videomega import VideoMegaIE
 from .videopremium import VideoPremiumIE
@@ -729,6 +732,7 @@ from .vidme import VidmeIE
 from .vidzi import VidziIE
 from .vier import VierIE, VierVideosIE
 from .viewster import ViewsterIE
+from .viidea import ViideaIE
 from .vimeo import (
     VimeoIE,
     VimeoAlbumIE,
@@ -781,6 +785,7 @@ from .wrzuta import WrzutaIE
 from .wsj import WSJIE
 from .xbef import XBefIE
 from .xboxclips import XboxClipsIE
+from .xfileshare import XFileShareIE
 from .xhamster import (
     XHamsterIE,
     XHamsterEmbedIE,

youtube_dl/extractor/abc.py

@@ -12,7 +12,7 @@ from ..utils import (
 
 class ABCIE(InfoExtractor):
     IE_NAME = 'abc.net.au'
-    _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P<id>\d+)'
+    _VALID_URL = r'http://www\.abc\.net\.au/news/(?:[^/]+/){1,2}(?P<id>\d+)'
 
     _TESTS = [{
         'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
@@ -36,6 +36,18 @@ class ABCIE(InfoExtractor):
             'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill',
         },
         'add_ie': ['Youtube'],
+    }, {
+        'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080',
+        'md5': 'b96eee7c9edf4fc5a358a0252881cc1f',
+        'info_dict': {
+            'id': '6880080',
+            'ext': 'mp3',
+            'title': 'NAB lifts interest rates, following Westpac and CBA',
+            'description': 'md5:f13d8edc81e462fce4a0437c7dc04728',
+        },
+    }, {
+        'url': 'http://www.abc.net.au/news/2015-10-19/6866214',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -43,7 +55,7 @@ class ABCIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         mobj = re.search(
-            r'inline(?P<type>Video|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
+            r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
             webpage)
         if mobj is None:
             raise ExtractorError('Unable to extract video urls')
@@ -60,11 +72,13 @@ class ABCIE(InfoExtractor):
 
         formats = [{
             'url': url_info['url'],
+            'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none',
             'width': int_or_none(url_info.get('width')),
             'height': int_or_none(url_info.get('height')),
             'tbr': int_or_none(url_info.get('bitrate')),
             'filesize': int_or_none(url_info.get('filesize')),
         } for url_info in urls_info]
+
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/anitube.py

@@ -26,8 +26,8 @@ class AnitubeIE(InfoExtractor):
         video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id)
-        key = self._html_search_regex(
-            r'http://www\.anitube\.se/embed/([A-Za-z0-9_-]*)', webpage, 'key')
+        key = self._search_regex(
+            r'src=["\']https?://[^/]+/embed/([A-Za-z0-9_-]+)', webpage, 'key')
 
         config_xml = self._download_xml(
             'http://www.anitube.se/nuevo/econfig.php?key=%s' % key, key)

youtube_dl/extractor/ard.py

@@ -14,8 +14,8 @@ from ..utils import (
     parse_duration,
     unified_strdate,
     xpath_text,
-    parse_xml,
 )
+from ..compat import compat_etree_fromstring
 
 
 class ARDMediathekIE(InfoExtractor):
@@ -161,7 +161,7 @@ class ARDMediathekIE(InfoExtractor):
             raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True)
 
         if re.search(r'[\?&]rss($|[=&])', url):
-            doc = parse_xml(webpage)
+            doc = compat_etree_fromstring(webpage.encode('utf-8'))
             if doc.tag == 'rss':
                 return GenericIE()._extract_rss(url, video_id, doc)
 

youtube_dl/extractor/bbc.py

@@ -2,7 +2,6 @@
 from __future__ import unicode_literals
 
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -14,7 +13,10 @@ from ..utils import (
     remove_end,
     unescapeHTML,
 )
-from ..compat import compat_HTTPError
+from ..compat import (
+    compat_etree_fromstring,
+    compat_HTTPError,
+)
 
 
 class BBCCoUkIE(InfoExtractor):
@@ -344,7 +346,7 @@ class BBCCoUkIE(InfoExtractor):
                 url, programme_id, 'Downloading media selection XML')
         except ExtractorError as ee:
             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
-                media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
+                media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
             else:
                 raise
         return self._process_media_selector(media_selection, programme_id)

youtube_dl/extractor/beeg.py

@@ -33,6 +33,8 @@ class BeegIE(InfoExtractor):
 
         formats = []
         for format_id, video_url in video.items():
+            if not video_url:
+                continue
             height = self._search_regex(
                 r'^(\d+)[pP]$', format_id, 'height', default=None)
             if not height:

youtube_dl/extractor/bilibili.py

@@ -4,9 +4,11 @@ from __future__ import unicode_literals
 import re
 import itertools
 import json
-import xml.etree.ElementTree as ET
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_etree_fromstring,
+)
 from ..utils import (
     int_or_none,
     unified_strdate,
@@ -88,7 +90,7 @@ class BiliBiliIE(InfoExtractor):
         except ValueError:
             pass
 
-        lq_doc = ET.fromstring(lq_page)
+        lq_doc = compat_etree_fromstring(lq_page)
         lq_durls = lq_doc.findall('./durl')
 
         hq_doc = self._download_xml(

youtube_dl/extractor/brightcove.py

@@ -3,10 +3,10 @@ from __future__ import unicode_literals
 
 import re
 import json
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_etree_fromstring,
     compat_parse_qs,
     compat_str,
     compat_urllib_parse,
@@ -119,7 +119,7 @@ class BrightcoveIE(InfoExtractor):
         object_str = fix_xml_ampersands(object_str)
 
         try:
-            object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
+            object_doc = compat_etree_fromstring(object_str.encode('utf-8'))
         except compat_xml_parse_error:
             return
 

youtube_dl/extractor/cbsnews.py

@@ -67,9 +67,12 @@ class CBSNewsIE(InfoExtractor):
                 'format_id': format_id,
             }
             if uri.startswith('rtmp'):
+                play_path = re.sub(
+                    r'{slistFilePath}', '',
+                    uri.split('<break>')[-1].split('{break}')[-1])
                 fmt.update({
                     'app': 'ondemand?auth=cbs',
-                    'play_path': 'mp4:' + uri.split('<break>')[-1],
+                    'play_path': 'mp4:' + play_path,
                     'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf',
                     'page_url': 'http://www.cbsnews.com',
                     'ext': 'flv',

youtube_dl/extractor/clyp.py

@@ -0,0 +1,57 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    parse_iso8601,
+)
+
+
+class ClypIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?clyp\.it/(?P<id>[a-z0-9]+)'
+    _TEST = {
+        'url': 'https://clyp.it/ojz2wfah',
+        'md5': '1d4961036c41247ecfdcc439c0cddcbb',
+        'info_dict': {
+            'id': 'ojz2wfah',
+            'ext': 'mp3',
+            'title': 'Krisson80 - bits wip wip',
+            'description': '#Krisson80BitsWipWip #chiptune\n#wip',
+            'duration': 263.21,
+            'timestamp': 1443515251,
+            'upload_date': '20150929',
+        },
+    }
+
+    def _real_extract(self, url):
+        audio_id = self._match_id(url)
+
+        metadata = self._download_json(
+            'https://api.clyp.it/%s' % audio_id, audio_id)
+
+        formats = []
+        for secure in ('', 'Secure'):
+            for ext in ('Ogg', 'Mp3'):
+                format_id = '%s%s' % (secure, ext)
+                format_url = metadata.get('%sUrl' % format_id)
+                if format_url:
+                    formats.append({
+                        'url': format_url,
+                        'format_id': format_id,
+                        'vcodec': 'none',
+                    })
+        self._sort_formats(formats)
+
+        title = metadata['Title']
+        description = metadata.get('Description')
+        duration = float_or_none(metadata.get('Duration'))
+        timestamp = parse_iso8601(metadata.get('DateCreated'))
+
+        return {
+            'id': audio_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'timestamp': timestamp,
+            'formats': formats,
+        }

youtube_dl/extractor/cmt.py

@@ -4,7 +4,7 @@ from .mtv import MTVIE
 
 class CMTIE(MTVIE):
     IE_NAME = 'cmt.com'
-    _VALID_URL = r'https?://www\.cmt\.com/videos/.+?/(?P<videoid>[^/]+)\.jhtml'
+    _VALID_URL = r'https?://www\.cmt\.com/(?:videos|shows)/(?:[^/]+/)*(?P<videoid>\d+)'
     _FEED_URL = 'http://www.cmt.com/sitewide/apps/player/embed/rss/'
 
     _TESTS = [{
@@ -16,4 +16,7 @@ class CMTIE(MTVIE):
             'title': 'Garth Brooks - "The Call (featuring Trisha Yearwood)"',
             'description': 'Blame It All On My Roots',
         },
+    }, {
+        'url': 'http://www.cmt.com/shows/party-down-south/party-down-south-ep-407-gone-girl/1738172/playlist/#id=1738172',
+        'only_matching': True,
     }]

youtube_dl/extractor/common.py

@@ -10,13 +10,11 @@ import re
 import socket
 import sys
 import time
-import xml.etree.ElementTree
 
 from ..compat import (
     compat_cookiejar,
     compat_cookies,
     compat_getpass,
-    compat_HTTPError,
     compat_http_client,
     compat_urllib_error,
     compat_urllib_parse,
@@ -24,6 +22,7 @@ from ..compat import (
     compat_urllib_request,
     compat_urlparse,
     compat_str,
+    compat_etree_fromstring,
 )
 from ..utils import (
     NO_DEFAULT,
@@ -311,11 +310,11 @@ class InfoExtractor(object):
     @classmethod
     def ie_key(cls):
         """A string for getting the InfoExtractor with get_info_extractor"""
-        return cls.__name__[:-2]
+        return compat_str(cls.__name__[:-2])
 
     @property
     def IE_NAME(self):
-        return type(self).__name__[:-2]
+        return compat_str(type(self).__name__[:-2])
 
     def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
         """ Returns the response handle """
@@ -462,7 +461,7 @@ class InfoExtractor(object):
             return xml_string
         if transform_source:
             xml_string = transform_source(xml_string)
-        return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
+        return compat_etree_fromstring(xml_string.encode('utf-8'))
 
     def _download_json(self, url_or_request, video_id,
                        note='Downloading JSON metadata',
@@ -842,7 +841,7 @@ class InfoExtractor(object):
             self._request_webpage(url, video_id, 'Checking %s URL' % item)
             return True
         except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError):
+            if isinstance(e.cause, compat_urllib_error.URLError):
                 self.to_screen(
                     '%s: %s URL is invalid, skipping' % (video_id, item))
                 return False
@@ -944,13 +943,15 @@ class InfoExtractor(object):
             if re.match(r'^https?://', u)
             else compat_urlparse.urljoin(m3u8_url, u))
 
-        m3u8_doc = self._download_webpage(
+        res = self._download_webpage_handle(
             m3u8_url, video_id,
             note=note or 'Downloading m3u8 information',
             errnote=errnote or 'Failed to download m3u8 information',
             fatal=fatal)
-        if m3u8_doc is False:
-            return m3u8_doc
+        if res is False:
+            return res
+        m3u8_doc, urlh = res
+        m3u8_url = urlh.geturl()
         last_info = None
         last_media = None
         kv_rex = re.compile(

youtube_dl/extractor/crunchyroll.py

@@ -5,12 +5,12 @@ import re
 import json
 import base64
 import zlib
-import xml.etree.ElementTree
 
 from hashlib import sha1
 from math import pow, sqrt, floor
 from .common import InfoExtractor
 from ..compat import (
+    compat_etree_fromstring,
     compat_urllib_parse,
     compat_urllib_parse_unquote,
     compat_urllib_request,
@@ -21,6 +21,7 @@ from ..utils import (
     bytes_to_intlist,
     intlist_to_bytes,
     int_or_none,
+    lowercase_escape,
     remove_end,
     unified_strdate,
     urlencode_postdata,
@@ -104,7 +105,7 @@ class CrunchyrollIE(CrunchyrollBaseIE):
             'id': '589804',
             'ext': 'flv',
             'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
-            'description': 'md5:fe2743efedb49d279552926d0bd0cd9e',
+            'description': 'md5:2fbc01f90b87e8e9137296f37b461c12',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': 'Danny Choo Network',
             'upload_date': '20120213',
@@ -234,7 +235,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         return output
 
     def _extract_subtitles(self, subtitle):
-        sub_root = xml.etree.ElementTree.fromstring(subtitle)
+        sub_root = compat_etree_fromstring(subtitle)
         return [{
             'ext': 'srt',
             'data': self._convert_subtitles_to_srt(sub_root),
@@ -287,11 +288,15 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         if 'To view this, please log in to verify you are 18 or older.' in webpage:
             self.raise_login_required()
 
-        video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
+        video_title = self._html_search_regex(
+            r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
+            webpage, 'video_title')
         video_title = re.sub(r' {2,}', ' ', video_title)
-        video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
-        if not video_description:
-            video_description = None
+        video_description = self._html_search_regex(
+            r'<script[^>]*>\s*.+?\[media_id=%s\].+?"description"\s*:\s*"([^"]+)' % video_id,
+            webpage, 'description', default=None)
+        if video_description:
+            video_description = lowercase_escape(video_description.replace(r'\r\n', '\n'))
         video_upload_date = self._html_search_regex(
             [r'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'],
             webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)

youtube_dl/extractor/dailymotion.py

@@ -141,9 +141,17 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                     type_ = media.get('type')
                     if type_ == 'application/vnd.lumberjack.manifest':
                         continue
-                    if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8':
-                        formats.extend(self._extract_m3u8_formats(
-                            media_url, video_id, 'mp4', m3u8_id='hls'))
+                    ext = determine_ext(media_url)
+                    if type_ == 'application/x-mpegURL' or ext == 'm3u8':
+                        m3u8_formats = self._extract_m3u8_formats(
+                            media_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
+                        if m3u8_formats:
+                            formats.extend(m3u8_formats)
+                    elif type_ == 'application/f4m' or ext == 'f4m':
+                        f4m_formats = self._extract_f4m_formats(
+                            media_url, video_id, preference=-1, f4m_id='hds', fatal=False)
+                        if f4m_formats:
+                            formats.extend(f4m_formats)
                     else:
                         f = {
                             'url': media_url,

youtube_dl/extractor/democracynow.py

@@ -0,0 +1,88 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+import os.path
+
+from .common import InfoExtractor
+from ..compat import compat_urlparse
+from ..utils import (
+    url_basename,
+    remove_start,
+)
+
+
+class DemocracynowIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?democracynow.org/(?P<id>[^\?]*)'
+    IE_NAME = 'democracynow'
+    _TESTS = [{
+        'url': 'http://www.democracynow.org/shows/2015/7/3',
+        'md5': 'fbb8fe3d7a56a5e12431ce2f9b2fab0d',
+        'info_dict': {
+            'id': '2015-0703-001',
+            'ext': 'mp4',
+            'title': 'July 03, 2015 - Democracy Now!',
+            'description': 'A daily independent global news hour with Amy Goodman & Juan González "What to the Slave is 4th of July?": James Earl Jones Reads Frederick Douglass\u2019 Historic Speech : "This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag : "We Shall Overcome": Remembering Folk Icon, Activist Pete Seeger in His Own Words & Songs',
+        },
+    }, {
+        'url': 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree',
+        'md5': 'fbb8fe3d7a56a5e12431ce2f9b2fab0d',
+        'info_dict': {
+            'id': '2015-0703-001',
+            'ext': 'mp4',
+            'title': '"This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag',
+            'description': 'md5:4d2bc4f0d29f5553c2210a4bc7761a21',
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        description = self._og_search_description(webpage)
+
+        json_data = self._parse_json(self._search_regex(
+            r'<script[^>]+type="text/json"[^>]*>\s*({[^>]+})', webpage, 'json'),
+            display_id)
+        video_id = None
+        formats = []
+
+        default_lang = 'en'
+
+        subtitles = {}
+
+        def add_subtitle_item(lang, info_dict):
+            if lang not in subtitles:
+                subtitles[lang] = []
+            subtitles[lang].append(info_dict)
+
+        # chapter_file are not subtitles
+        if 'caption_file' in json_data:
+            add_subtitle_item(default_lang, {
+                'url': compat_urlparse.urljoin(url, json_data['caption_file']),
+            })
+
+        for subtitle_item in json_data.get('captions', []):
+            lang = subtitle_item.get('language', '').lower() or default_lang
+            add_subtitle_item(lang, {
+                'url': compat_urlparse.urljoin(url, subtitle_item['url']),
+            })
+
+        for key in ('file', 'audio', 'video'):
+            media_url = json_data.get(key, '')
+            if not media_url:
+                continue
+            media_url = re.sub(r'\?.*', '', compat_urlparse.urljoin(url, media_url))
+            video_id = video_id or remove_start(os.path.splitext(url_basename(media_url))[0], 'dn')
+            formats.append({
+                'url': media_url,
+            })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id or display_id,
+            'title': json_data['title'],
+            'description': description,
+            'subtitles': subtitles,
+            'formats': formats,
+        }

youtube_dl/extractor/eitb.py

@@ -1,39 +1,92 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from .brightcove import BrightcoveIE
-from ..utils import ExtractorError
+from ..compat import compat_urllib_request
+from ..utils import (
+    float_or_none,
+    int_or_none,
+    parse_iso8601,
+)
 
 
 class EitbIE(InfoExtractor):
     IE_NAME = 'eitb.tv'
-    _VALID_URL = r'https?://www\.eitb\.tv/(eu/bideoa|es/video)/[^/]+/(?P<playlist_id>\d+)/(?P<chapter_id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?eitb\.tv/(?:eu/bideoa|es/video)/[^/]+/\d+/(?P<id>\d+)'
 
     _TEST = {
-        'add_ie': ['Brightcove'],
-        'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/2677100210001/2743577154001/lasa-y-zabala-30-anos/',
+        'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/4104995148001/4090227752001/lasa-y-zabala-30-anos/',
         'md5': 'edf4436247185adee3ea18ce64c47998',
         'info_dict': {
-            'id': '2743577154001',
+            'id': '4090227752001',
             'ext': 'mp4',
             'title': '60 minutos (Lasa y Zabala, 30 años)',
-            # All videos from eitb has this description in the brightcove info
-            'description': '.',
-            'uploader': 'Euskal Telebista',
+            'description': 'Programa de reportajes de actualidad.',
+            'duration': 3996.76,
+            'timestamp': 1381789200,
+            'upload_date': '20131014',
+            'tags': list,
         },
     }
 
     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('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())
+        video_id = self._match_id(url)
+
+        video = self._download_json(
+            'http://mam.eitb.eus/mam/REST/ServiceMultiweb/Video/MULTIWEBTV/%s/' % video_id,
+            video_id, 'Downloading video JSON')
+
+        media = video['web_media'][0]
+
+        formats = []
+        for rendition in media['RENDITIONS']:
+            video_url = rendition.get('PMD_URL')
+            if not video_url:
+                continue
+            tbr = float_or_none(rendition.get('ENCODING_RATE'), 1000)
+            format_id = 'http'
+            if tbr:
+                format_id += '-%d' % int(tbr)
+            formats.append({
+                'url': rendition['PMD_URL'],
+                'format_id': format_id,
+                'width': int_or_none(rendition.get('FRAME_WIDTH')),
+                'height': int_or_none(rendition.get('FRAME_HEIGHT')),
+                'tbr': tbr,
+            })
+
+        hls_url = media.get('HLS_SURL')
+        if hls_url:
+            request = compat_urllib_request.Request(
+                'http://mam.eitb.eus/mam/REST/ServiceMultiweb/DomainRestrictedSecurity/TokenAuth/',
+                headers={'Referer': url})
+            token_data = self._download_json(
+                request, video_id, 'Downloading auth token', fatal=False)
+            if token_data:
+                token = token_data.get('token')
+                if token:
+                    m3u8_formats = self._extract_m3u8_formats(
+                        '%s?hdnts=%s' % (hls_url, token), video_id, m3u8_id='hls', fatal=False)
+                    if m3u8_formats:
+                        formats.extend(m3u8_formats)
+
+        hds_url = media.get('HDS_SURL')
+        if hds_url:
+            f4m_formats = self._extract_f4m_formats(
+                '%s?hdcore=3.7.0' % hds_url.replace('euskalsvod', 'euskalvod'),
+                video_id, f4m_id='hds', fatal=False)
+            if f4m_formats:
+                formats.extend(f4m_formats)
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': media.get('NAME_ES') or media.get('name') or media['NAME_EU'],
+            'description': media.get('SHORT_DESC_ES') or video.get('desc_group') or media.get('SHORT_DESC_EU'),
+            'thumbnail': media.get('STILL_URL') or media.get('THUMBNAIL_URL'),
+            'duration': float_or_none(media.get('LENGTH'), 1000),
+            'timestamp': parse_iso8601(media.get('BROADCST_DATE'), ' '),
+            'tags': media.get('TAGS'),
+            'formats': formats,
+        }

youtube_dl/extractor/extremetube.py

@@ -3,23 +3,20 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_parse_qs,
-    compat_urllib_request,
-)
+from ..compat import compat_urllib_request
 from ..utils import (
-    qualities,
+    int_or_none,
     str_to_int,
 )
 
 
 class ExtremeTubeIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?(?P<url>extremetube\.com/.*?video/.+?(?P<id>[0-9]+))(?:[/?&]|$)'
+    _VALID_URL = r'https?://(?:www\.)?extremetube\.com/(?:[^/]+/)?video/(?P<id>[^/#?&]+)'
     _TESTS = [{
         'url': 'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431',
         'md5': '344d0c6d50e2f16b06e49ca011d8ac69',
         'info_dict': {
-            'id': '652431',
+            'id': 'music-video-14-british-euro-brit-european-cumshots-swallow-652431',
             'ext': 'mp4',
             'title': 'Music Video 14 british euro brit european cumshots swallow',
             'uploader': 'unknown',
@@ -29,12 +26,16 @@ class ExtremeTubeIE(InfoExtractor):
     }, {
         'url': 'http://www.extremetube.com/gay/video/abcde-1234',
         'only_matching': True,
+    }, {
+        'url': 'http://www.extremetube.com/video/latina-slut-fucked-by-fat-black-dick',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.extremetube.com/video/652431',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        url = 'http://www.' + mobj.group('url')
+        video_id = self._match_id(url)
 
         req = compat_urllib_request.Request(url)
         req.add_header('Cookie', 'age_verified=1')
@@ -49,20 +50,36 @@ class ExtremeTubeIE(InfoExtractor):
             r'Views:\s*</strong>\s*<span>([\d,\.]+)</span>',
             webpage, 'view count', fatal=False))
 
-        flash_vars = compat_parse_qs(self._search_regex(
-            r'<param[^>]+?name="flashvars"[^>]+?value="([^"]+)"', webpage, 'flash vars'))
+        flash_vars = self._parse_json(
+            self._search_regex(
+                r'var\s+flashvars\s*=\s*({.+?});', webpage, 'flash vars'),
+            video_id)
 
         formats = []
-        quality = qualities(['180p', '240p', '360p', '480p', '720p', '1080p'])
-        for k, vals in flash_vars.items():
-            m = re.match(r'quality_(?P<quality>[0-9]+p)$', k)
-            if m is not None:
-                formats.append({
-                    'format_id': m.group('quality'),
-                    'quality': quality(m.group('quality')),
-                    'url': vals[0],
+        for quality_key, video_url in flash_vars.items():
+            height = int_or_none(self._search_regex(
+                r'quality_(\d+)[pP]$', quality_key, 'height', default=None))
+            if not height:
+                continue
+            f = {
+                'url': video_url,
+            }
+            mobj = re.search(
+                r'/(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', video_url)
+            if mobj:
+                height = int(mobj.group('height'))
+                bitrate = int(mobj.group('bitrate'))
+                f.update({
+                    'format_id': '%dp-%dk' % (height, bitrate),
+                    'height': height,
+                    'tbr': bitrate,
                 })
-
+            else:
+                f.update({
+                    'format_id': '%dp' % height,
+                    'height': height,
+                })
+            formats.append(f)
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/francetv.py

@@ -83,6 +83,14 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
         if subtitle:
             title += ' - %s' % subtitle
 
+        subtitles = {}
+        subtitles_list = [{
+            'url': subformat['url'],
+            'ext': subformat.get('format'),
+        } for subformat in info.get('subtitles', []) if subformat.get('url')]
+        if subtitles_list:
+            subtitles['fr'] = subtitles_list
+
         return {
             'id': video_id,
             'title': title,
@@ -91,20 +99,27 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
             'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
             'timestamp': int_or_none(info['diffusion']['timestamp']),
             'formats': formats,
+            'subtitles': subtitles,
         }
 
 
 class PluzzIE(FranceTVBaseInfoExtractor):
     IE_NAME = 'pluzz.francetv.fr'
-    _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html'
+    _VALID_URL = r'https?://(?:m\.)?pluzz\.francetv\.fr/videos/(?P<id>.+?)\.html'
 
     # Can't use tests, videos expire in 7 days
 
     def _real_extract(self, url):
-        title = re.match(self._VALID_URL, url).group(1)
-        webpage = self._download_webpage(url, title)
-        video_id = self._search_regex(
-            r'data-diffusion="(\d+)"', webpage, 'ID')
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        video_id = self._html_search_meta(
+            'id_video', webpage, 'video id', default=None)
+        if not video_id:
+            video_id = self._search_regex(
+                r'data-diffusion=["\'](\d+)', webpage, 'video id')
+
         return self._extract_video(video_id, 'Pluzz')
 
 
@@ -120,6 +135,9 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
             'title': 'Soir 3',
             'upload_date': '20130826',
             'timestamp': 1377548400,
+            'subtitles': {
+                'fr': 'mincount:2',
+            },
         },
     }, {
         'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',

youtube_dl/extractor/funnyordie.py

@@ -45,11 +45,20 @@ class FunnyOrDieIE(InfoExtractor):
 
         links.sort(key=lambda link: 1 if link[1] == 'mp4' else 0)
 
-        bitrates = self._html_search_regex(r'<source src="[^"]+/v,((?:\d+,)+)\.mp4\.csmil', webpage, 'video bitrates')
-        bitrates = [int(b) for b in bitrates.rstrip(',').split(',')]
-        bitrates.sort()
+        m3u8_url = self._search_regex(
+            r'<source[^>]+src=(["\'])(?P<url>.+?/master\.m3u8)\1',
+            webpage, 'm3u8 url', default=None, group='url')
 
         formats = []
+
+        m3u8_formats = self._extract_m3u8_formats(
+            m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
+        if m3u8_formats:
+            formats.extend(m3u8_formats)
+
+        bitrates = [int(bitrate) for bitrate in re.findall(r'[,/]v(\d+)[,/]', m3u8_url)]
+        bitrates.sort()
+
         for bitrate in bitrates:
             for link in links:
                 formats.append({

youtube_dl/extractor/generic.py

@@ -9,6 +9,7 @@ import sys
 from .common import InfoExtractor
 from .youtube import YoutubeIE
 from ..compat import (
+    compat_etree_fromstring,
     compat_urllib_parse_unquote,
     compat_urllib_request,
     compat_urlparse,
@@ -21,7 +22,6 @@ from ..utils import (
     HEADRequest,
     is_html,
     orderedSet,
-    parse_xml,
     smuggle_url,
     unescapeHTML,
     unified_strdate,
@@ -141,6 +141,7 @@ class GenericIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Automatics, robotics and biocybernetics',
                 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
+                'upload_date': '20130627',
                 'formats': 'mincount:16',
                 'subtitles': 'mincount:1',
             },
@@ -1237,7 +1238,7 @@ class GenericIE(InfoExtractor):
 
         # Is it an RSS feed, a SMIL file or a XSPF playlist?
         try:
-            doc = parse_xml(webpage)
+            doc = compat_etree_fromstring(webpage.encode('utf-8'))
             if doc.tag == 'rss':
                 return self._extract_rss(url, video_id, doc)
             elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
@@ -1671,8 +1672,8 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'), 'Zapiks')
 
         # Look for Kaltura embeds
-        mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage) or
-                re.search(r'(?s)(["\'])(?:https?:)?//cdnapisec\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?\1.*?entry_id\s*:\s*(["\'])(?P<id>[^\2]+?)\2', webpage))
+        mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or
+                re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage))
         if mobj is not None:
             return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
 

youtube_dl/extractor/globo.py

@@ -14,79 +14,58 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     int_or_none,
+    str_or_none,
 )
 
 
 class GloboIE(InfoExtractor):
-    _VALID_URL = 'https?://.+?\.globo\.com/(?P<id>.+)'
+    _VALID_URL = '(?:globo:|https?://.+?\.globo\.com/(?:[^/]+/)*(?:v/(?:[^/]+/)?|videos/))(?P<id>\d{7,})'
 
     _API_URL_TEMPLATE = 'http://api.globovideos.com/videos/%s/playlist'
     _SECURITY_URL_TEMPLATE = 'http://security.video.globo.com/videos/%s/hash?player=flash&version=17.0.0.132&resource_id=%s'
 
-    _VIDEOID_REGEXES = [
-        r'\bdata-video-id="(\d+)"',
-        r'\bdata-player-videosids="(\d+)"',
-        r'<div[^>]+\bid="(\d+)"',
-    ]
-
     _RESIGN_EXPIRATION = 86400
 
-    _TESTS = [
-        {
-            'url': 'http://globotv.globo.com/sportv/futebol-nacional/v/os-gols-de-atletico-mg-3-x-2-santos-pela-24a-rodada-do-brasileirao/3654973/',
-            'md5': '03ebf41cb7ade43581608b7d9b71fab0',
-            'info_dict': {
-                'id': '3654973',
-                'ext': 'mp4',
-                'title': 'Os gols de Atlético-MG 3 x 2 Santos pela 24ª rodada do Brasileirão',
-                'duration': 251.585,
-                'uploader': 'SporTV',
-                'uploader_id': 698,
-                'like_count': int,
-            }
+    _TESTS = [{
+        'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/',
+        'md5': 'b3ccc801f75cd04a914d51dadb83a78d',
+        'info_dict': {
+            'id': '3607726',
+            'ext': 'mp4',
+            'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa',
+            'duration': 103.204,
+            'uploader': 'Globo.com',
+            'uploader_id': '265',
         },
-        {
-            'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/',
-            'md5': 'b3ccc801f75cd04a914d51dadb83a78d',
-            'info_dict': {
-                'id': '3607726',
-                'ext': 'mp4',
-                'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa',
-                'duration': 103.204,
-                'uploader': 'Globo.com',
-                'uploader_id': 265,
-                'like_count': int,
-            }
+    }, {
+        'url': 'http://globoplay.globo.com/v/4581987/',
+        'md5': 'f36a1ecd6a50da1577eee6dd17f67eff',
+        'info_dict': {
+            'id': '4581987',
+            'ext': 'mp4',
+            'title': 'Acidentes de trânsito estão entre as maiores causas de queda de energia em SP',
+            'duration': 137.973,
+            'uploader': 'Rede Globo',
+            'uploader_id': '196',
         },
-        {
-            'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html',
-            'md5': '307fdeae4390ccfe6ba1aa198cf6e72b',
-            'info_dict': {
-                'id': '3652183',
-                'ext': 'mp4',
-                'title': 'Receita Federal explica como vai fiscalizar bagagens de quem retorna ao Brasil de avião',
-                'duration': 110.711,
-                'uploader': 'Rede Globo',
-                'uploader_id': 196,
-                'like_count': int,
-            }
-        },
-        {
-            'url': 'http://globotv.globo.com/canal-brasil/sangue-latino/t/todos-os-videos/v/ator-e-diretor-argentino-ricado-darin-fala-sobre-utopias-e-suas-perdas/3928201/',
-            'md5': 'c1defca721ce25b2354e927d3e4b3dec',
-            'info_dict': {
-                'id': '3928201',
-                'ext': 'mp4',
-                'title': 'Ator e diretor argentino, Ricado Darín fala sobre utopias e suas perdas',
-                'duration': 1472.906,
-                'uploader': 'Canal Brasil',
-                'uploader_id': 705,
-                'like_count': int,
-            }
-        },
-    ]
+    }, {
+        'url': 'http://canalbrasil.globo.com/programas/sangue-latino/videos/3928201.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://globosatplay.globo.com/globonews/v/4472924/',
+        'only_matching': True,
+    }, {
+        'url': 'http://globotv.globo.com/t/programa/v/clipe-sexo-e-as-negas-adeus/3836166/',
+        'only_matching': True,
+    }, {
+        'url': 'http://globotv.globo.com/canal-brasil/sangue-latino/t/todos-os-videos/v/ator-e-diretor-argentino-ricado-darin-fala-sobre-utopias-e-suas-perdas/3928201/',
+        'only_matching': True,
+    }, {
+        'url': 'http://canaloff.globo.com/programas/desejar-profundo/videos/4518560.html',
+        'only_matching': True,
+    }]
 
-    class MD5():
+    class MD5:
         HEX_FORMAT_LOWERCASE = 0
         HEX_FORMAT_UPPERCASE = 1
         BASE64_PAD_CHARACTER_DEFAULT_COMPLIANCE = ''
@@ -353,9 +332,6 @@ class GloboIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
-        video_id = self._search_regex(self._VIDEOID_REGEXES, webpage, 'video id')
-
         video = self._download_json(
             self._API_URL_TEMPLATE % video_id, video_id)['videos'][0]
 
@@ -364,7 +340,7 @@ class GloboIE(InfoExtractor):
         formats = []
         for resource in video['resources']:
             resource_id = resource.get('_id')
-            if not resource_id:
+            if not resource_id or resource_id.endswith('manifest'):
                 continue
 
             security = self._download_json(
@@ -393,20 +369,23 @@ class GloboIE(InfoExtractor):
             resource_url = resource['url']
             signed_url = '%s?h=%s&k=%s' % (resource_url, signed_hash, 'flash')
             if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'):
-                formats.extend(self._extract_m3u8_formats(signed_url, resource_id, 'mp4'))
+                m3u8_formats = self._extract_m3u8_formats(
+                    signed_url, resource_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False)
+                if m3u8_formats:
+                    formats.extend(m3u8_formats)
             else:
                 formats.append({
                     'url': signed_url,
-                    'format_id': resource_id,
-                    'height': resource.get('height'),
+                    'format_id': 'http-%s' % resource_id,
+                    'height': int_or_none(resource.get('height')),
                 })
 
         self._sort_formats(formats)
 
         duration = float_or_none(video.get('duration'), 1000)
-        like_count = int_or_none(video.get('likes'))
         uploader = video.get('channel')
-        uploader_id = video.get('channel_id')
+        uploader_id = str_or_none(video.get('channel_id'))
 
         return {
             'id': video_id,
@@ -414,6 +393,46 @@ class GloboIE(InfoExtractor):
             'duration': duration,
             'uploader': uploader,
             'uploader_id': uploader_id,
-            'like_count': like_count,
             'formats': formats
         }
+
+
+class GloboArticleIE(InfoExtractor):
+    _VALID_URL = 'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/]+)\.html'
+
+    _VIDEOID_REGEXES = [
+        r'\bdata-video-id=["\'](\d{7,})',
+        r'\bdata-player-videosids=["\'](\d{7,})',
+        r'\bvideosIDs\s*:\s*["\'](\d{7,})',
+        r'\bdata-id=["\'](\d{7,})',
+        r'<div[^>]+\bid=["\'](\d{7,})',
+    ]
+
+    _TESTS = [{
+        'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html',
+        'md5': '307fdeae4390ccfe6ba1aa198cf6e72b',
+        'info_dict': {
+            'id': '3652183',
+            'ext': 'mp4',
+            'title': 'Receita Federal explica como vai fiscalizar bagagens de quem retorna ao Brasil de avião',
+            'duration': 110.711,
+            'uploader': 'Rede Globo',
+            'uploader_id': '196',
+        }
+    }, {
+        'url': 'http://gq.globo.com/Prazeres/Poder/noticia/2015/10/all-o-desafio-assista-ao-segundo-capitulo-da-serie.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://gshow.globo.com/programas/tv-xuxa/O-Programa/noticia/2014/01/xuxa-e-junno-namoram-muuuito-em-luau-de-zeze-di-camargo-e-luciano.html',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if GloboIE.suitable(url) else super(GloboArticleIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_id = self._search_regex(self._VIDEOID_REGEXES, webpage, 'video id')
+        return self.url_result('globo:%s' % video_id, 'Globo')

youtube_dl/extractor/googleplus.py

@@ -61,7 +61,7 @@ class GooglePlusIE(InfoExtractor):
             'width': int(width),
             'height': int(height),
         } for width, height, video_url in re.findall(
-            r'\d+,(\d+),(\d+),"(https?://redirector\.googlevideo\.com.*?)"', webpage)]
+            r'\d+,(\d+),(\d+),"(https?://[^.]+\.googleusercontent.com.*?)"', webpage)]
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/kaltura.py

@@ -16,7 +16,7 @@ class KalturaIE(InfoExtractor):
                 (?:
                     kaltura:(?P<partner_id_s>\d+):(?P<id_s>[0-9a-z_]+)|
                     https?://
-                        (:?(?:www|cdnapisec)\.)?kaltura\.com/
+                        (:?(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/
                         (?:
                             (?:
                                 # flash player

youtube_dl/extractor/lynda.py

@@ -82,6 +82,11 @@ class LyndaBaseIE(InfoExtractor):
                         expected=True)
             raise ExtractorError('Unable to log in')
 
+    def _logout(self):
+        self._download_webpage(
+            'http://www.lynda.com/ajax/logout.aspx', None,
+            'Logging out', 'Unable to log out', fatal=False)
+
 
 class LyndaIE(LyndaBaseIE):
     IE_NAME = 'lynda'
@@ -108,51 +113,47 @@ class LyndaIE(LyndaBaseIE):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        page = self._download_webpage(
+        video = self._download_json(
             'http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
             video_id, 'Downloading video JSON')
-        video_json = json.loads(page)
 
-        if 'Status' in video_json:
+        if 'Status' in video:
             raise ExtractorError(
-                'lynda returned error: %s' % video_json['Message'], expected=True)
+                'lynda returned error: %s' % video['Message'], expected=True)
 
-        if video_json['HasAccess'] is False:
+        if video.get('HasAccess') is False:
             self.raise_login_required('Video %s is only available for members' % video_id)
 
-        video_id = compat_str(video_json['ID'])
-        duration = video_json['DurationInSeconds']
-        title = video_json['Title']
+        video_id = compat_str(video.get('ID') or video_id)
+        duration = int_or_none(video.get('DurationInSeconds'))
+        title = video['Title']
 
         formats = []
 
-        fmts = video_json.get('Formats')
+        fmts = video.get('Formats')
         if fmts:
-            formats.extend([
-                {
-                    'url': fmt['Url'],
-                    'ext': fmt['Extension'],
-                    'width': fmt['Width'],
-                    'height': fmt['Height'],
-                    'filesize': fmt['FileSize'],
-                    'format_id': str(fmt['Resolution'])
-                } for fmt in fmts])
+            formats.extend([{
+                'url': f['Url'],
+                'ext': f.get('Extension'),
+                'width': int_or_none(f.get('Width')),
+                'height': int_or_none(f.get('Height')),
+                'filesize': int_or_none(f.get('FileSize')),
+                'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None,
+            } for f in fmts if f.get('Url')])
 
-        prioritized_streams = video_json.get('PrioritizedStreams')
+        prioritized_streams = video.get('PrioritizedStreams')
         if prioritized_streams:
             for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
-                formats.extend([
-                    {
-                        'url': video_url,
-                        'width': int_or_none(format_id),
-                        'format_id': '%s-%s' % (prioritized_stream_id, format_id),
-                    } for format_id, video_url in prioritized_stream.items()
-                ])
+                formats.extend([{
+                    'url': video_url,
+                    'width': int_or_none(format_id),
+                    'format_id': '%s-%s' % (prioritized_stream_id, format_id),
+                } for format_id, video_url in prioritized_stream.items()])
 
         self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
-        subtitles = self.extract_subtitles(video_id, page)
+        subtitles = self.extract_subtitles(video_id)
 
         return {
             'id': video_id,
@@ -183,7 +184,7 @@ class LyndaIE(LyndaBaseIE):
         if srt:
             return srt
 
-    def _get_subtitles(self, video_id, webpage):
+    def _get_subtitles(self, video_id):
         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
         subs = self._download_json(url, None, False)
         if subs:
@@ -205,12 +206,13 @@ class LyndaCourseIE(LyndaBaseIE):
         course_path = mobj.group('coursepath')
         course_id = mobj.group('courseid')
 
-        page = self._download_webpage(
+        course = self._download_json(
             'http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
             course_id, 'Downloading course JSON')
-        course_json = json.loads(page)
 
-        if 'Status' in course_json and course_json['Status'] == 'NotFound':
+        self._logout()
+
+        if course.get('Status') == 'NotFound':
             raise ExtractorError(
                 'Course %s does not exist' % course_id, expected=True)
 
@@ -220,12 +222,13 @@ class LyndaCourseIE(LyndaBaseIE):
         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
         # by single video API anymore
 
-        for chapter in course_json['Chapters']:
-            for video in chapter['Videos']:
-                if video['HasAccess'] is False:
+        for chapter in course['Chapters']:
+            for video in chapter.get('Videos', []):
+                if video.get('HasAccess') is False:
                     unaccessible_videos += 1
                     continue
-                videos.append(video['ID'])
+                if video.get('ID'):
+                    videos.append(video['ID'])
 
         if unaccessible_videos > 0:
             self._downloader.report_warning(
@@ -238,6 +241,6 @@ class LyndaCourseIE(LyndaBaseIE):
                 'Lynda')
             for video_id in videos]
 
-        course_title = course_json['Title']
+        course_title = course.get('Title')
 
         return self.playlist_result(entries, course_id, course_title)

youtube_dl/extractor/mdr.py

@@ -1,64 +1,169 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..compat import compat_urlparse
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+    xpath_text,
+)
 
 
 class MDRIE(InfoExtractor):
-    _VALID_URL = r'^(?P<domain>https?://(?:www\.)?mdr\.de)/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)(?:_|\.html)'
+    IE_DESC = 'MDR.DE and KiKA'
+    _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z]+(?P<id>\d+)(?:_.+?)?\.html'
 
-    # No tests, MDR regularily deletes its videos
-    _TEST = {
+    _TESTS = [{
+        # MDR regularily deletes its videos
         'url': 'http://www.mdr.de/fakt/video189002.html',
         'only_matching': True,
-    }
+    }, {
+        # audio
+        'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html',
+        'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa',
+        'info_dict': {
+            'id': '1312272',
+            'ext': 'mp3',
+            'title': 'Feuilleton vom 30. Oktober 2015',
+            'duration': 250,
+            'uploader': 'MITTELDEUTSCHER RUNDFUNK',
+        },
+    }, {
+        'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
+        'md5': '4930515e36b06c111213e80d1e4aad0e',
+        'info_dict': {
+            'id': '19636',
+            'ext': 'mp4',
+            'title': 'Baumhaus vom 30. Oktober 2015',
+            'duration': 134,
+            'uploader': 'KIKA',
+        },
+    }, {
+        'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
+        'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
+        'info_dict': {
+            'id': '8182',
+            'ext': 'mp4',
+            'title': 'Beutolomäus und der geheime Weihnachtswunsch',
+            'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
+            'timestamp': 1419047100,
+            'upload_date': '20141220',
+            'duration': 4628,
+            'uploader': 'KIKA',
+        },
+    }, {
+        'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('video_id')
-        domain = m.group('domain')
+        video_id = self._match_id(url)
 
-        # determine title and media streams from webpage
-        html = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, video_id)
 
-        title = self._html_search_regex(r'<h[12]>(.*?)</h[12]>', html, 'title')
-        xmlurl = self._search_regex(
-            r'dataURL:\'(/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, 'XML URL')
+        data_url = self._search_regex(
+            r'dataURL\s*:\s*(["\'])(?P<url>/.+/(?:video|audio)[0-9]+-avCustom\.xml)\1',
+            webpage, 'data url', group='url')
+
+        doc = self._download_xml(
+            compat_urlparse.urljoin(url, data_url), video_id)
+
+        title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)
 
-        doc = self._download_xml(domain + xmlurl, video_id)
         formats = []
-        for a in doc.findall('./assets/asset'):
-            url_el = a.find('./progressiveDownloadUrl')
-            if url_el is None:
-                continue
-            abr = int(a.find('bitrateAudio').text) // 1000
-            media_type = a.find('mediaType').text
-            format = {
-                'abr': abr,
-                'filesize': int(a.find('fileSize').text),
-                'url': url_el.text,
-            }
+        processed_urls = []
+        for asset in doc.findall('./assets/asset'):
+            for source in (
+                    'progressiveDownload',
+                    'dynamicHttpStreamingRedirector',
+                    'adaptiveHttpStreamingRedirector'):
+                url_el = asset.find('./%sUrl' % source)
+                if url_el is None:
+                    continue
+
+                video_url = url_el.text
+                if video_url in processed_urls:
+                    continue
+
+                processed_urls.append(video_url)
+
+                vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
+                abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
+
+                ext = determine_ext(url_el.text)
+                if ext == 'm3u8':
+                    url_formats = self._extract_m3u8_formats(
+                        video_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                        preference=0, m3u8_id='HLS', fatal=False)
+                elif ext == 'f4m':
+                    url_formats = self._extract_f4m_formats(
+                        video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
+                        preference=0, f4m_id='HDS', fatal=False)
+                else:
+                    media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
+                    vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
+                    abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
+                    filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
+
+                    f = {
+                        'url': video_url,
+                        'format_id': '%s-%d' % (media_type, vbr or abr),
+                        'filesize': filesize,
+                        'abr': abr,
+                        'preference': 1,
+                    }
+
+                    if vbr:
+                        width = int_or_none(xpath_text(asset, './frameWidth', 'width'))
+                        height = int_or_none(xpath_text(asset, './frameHeight', 'height'))
+                        f.update({
+                            'vbr': vbr,
+                            'width': width,
+                            'height': height,
+                        })
+
+                    url_formats = [f]
+
+                if not url_formats:
+                    continue
+
+                if not vbr:
+                    for f in url_formats:
+                        abr = f.get('tbr') or abr
+                        if 'tbr' in f:
+                            del f['tbr']
+                        f.update({
+                            'abr': abr,
+                            'vcodec': 'none',
+                        })
+
+                formats.extend(url_formats)
 
-            vbr_el = a.find('bitrateVideo')
-            if vbr_el is None:
-                format.update({
-                    'vcodec': 'none',
-                    'format_id': '%s-%d' % (media_type, abr),
-                })
-            else:
-                vbr = int(vbr_el.text) // 1000
-                format.update({
-                    'vbr': vbr,
-                    'width': int(a.find('frameWidth').text),
-                    'height': int(a.find('frameHeight').text),
-                    'format_id': '%s-%d' % (media_type, vbr),
-                })
-            formats.append(format)
         self._sort_formats(formats)
 
+        description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
+        timestamp = parse_iso8601(
+            xpath_text(
+                doc, [
+                    './broadcast/broadcastDate',
+                    './broadcast/broadcastStartDate',
+                    './broadcast/broadcastEndDate'],
+                'timestamp', default=None))
+        duration = parse_duration(xpath_text(doc, './duration', 'duration'))
+        uploader = xpath_text(doc, './rights', 'uploader')
+
         return {
             'id': video_id,
             'title': title,
+            'description': description,
+            'timestamp': timestamp,
+            'duration': duration,
+            'uploader': uploader,
             'formats': formats,
         }

youtube_dl/extractor/miomio.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import random
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_request
 from ..utils import (
     xpath_text,
     int_or_none,
@@ -51,6 +52,8 @@ class MioMioIE(InfoExtractor):
         mioplayer_path = self._search_regex(
             r'src="(/mioplayer/[^"]+)"', webpage, 'ref_path')
 
+        http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}
+
         xml_config = self._search_regex(
             r'flashvars="type=(?:sina|video)&(.+?)&',
             webpage, 'xml config')
@@ -60,14 +63,12 @@ class MioMioIE(InfoExtractor):
             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)),
             video_id)
 
-        # the following xml contains the actual configuration information on the video file(s)
-        vid_config = self._download_xml(
+        vid_config_request = compat_urllib_request.Request(
             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config),
-            video_id)
+            headers=http_headers)
 
-        http_headers = {
-            'Referer': 'http://www.miomio.tv%s' % mioplayer_path,
-        }
+        # the following xml contains the actual configuration information on the video file(s)
+        vid_config = self._download_xml(vid_config_request, video_id)
 
         if not int_or_none(xpath_text(vid_config, 'timelength')):
             raise ExtractorError('Unable to load videos!', expected=True)

youtube_dl/extractor/mit.py

@@ -86,7 +86,7 @@ class MITIE(TechTVMITIE):
         webpage = self._download_webpage(url, page_title)
         embed_url = self._search_regex(
             r'<iframe .*?src="(.+?)"', webpage, 'embed url')
-        return self.url_result(embed_url, ie='TechTVMIT')
+        return self.url_result(embed_url)
 
 
 class OCWMITIE(InfoExtractor):

youtube_dl/extractor/mitele.py

@@ -1,7 +1,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse
+from ..compat import (
+    compat_urllib_parse,
+    compat_urlparse,
+)
 from ..utils import (
     encode_dict,
     get_element_by_attribute,
@@ -15,7 +18,7 @@ class MiTeleIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
-        'md5': 'ace7635b2a0b286aaa37d3ff192d2a8a',
+        'md5': '0ff1a13aebb35d9bc14081ff633dd324',
         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',
@@ -34,6 +37,7 @@ class MiTeleIE(InfoExtractor):
 
         config_url = self._search_regex(
             r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url')
+        config_url = compat_urlparse.urljoin(url, config_url)
 
         config = self._download_json(
             config_url, display_id, 'Downloading config JSON')
@@ -56,7 +60,7 @@ class MiTeleIE(InfoExtractor):
                 'sta': '0',
             }
             media = self._download_json(
-                '%s/?%s' % (gat, compat_urllib_parse.urlencode(encode_dict(token_data)).encode('utf-8')),
+                '%s/?%s' % (gat, compat_urllib_parse.urlencode(encode_dict(token_data))),
                 display_id, 'Downloading %s JSON' % location['loc'])
             file_ = media.get('file')
             if not file_:

youtube_dl/extractor/moniker.py

@@ -17,7 +17,7 @@ from ..utils import (
 
 class MonikerIE(InfoExtractor):
     IE_DESC = 'allmyvideos.net and vidspot.net'
-    _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?P<id>[a-zA-Z0-9_-]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?:(?:2|v)/v-)?(?P<id>[a-zA-Z0-9_-]+)'
 
     _TESTS = [{
         'url': 'http://allmyvideos.net/jih3nce3x6wn',
@@ -46,6 +46,18 @@ class MonikerIE(InfoExtractor):
     }, {
         'url': 'https://www.vidspot.net/l2ngsmhs8ci5',
         'only_matching': True,
+    }, {
+        'url': 'http://vidspot.net/2/v-ywDf99',
+        'md5': '5f8254ce12df30479428b0152fb8e7ba',
+        'info_dict': {
+            'id': 'ywDf99',
+            'ext': 'mp4',
+            'title': 'IL FAIT LE MALIN EN PORSHE CAYENNE ( mais pas pour longtemps)',
+            'description': 'IL FAIT LE MALIN EN PORSHE CAYENNE.',
+        },
+    }, {
+        'url': 'http://allmyvideos.net/v/v-HXZm5t',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -64,18 +76,30 @@ class MonikerIE(InfoExtractor):
             raise ExtractorError(
                 '%s returned error: %s' % (self.IE_NAME, error), expected=True)
 
-        fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage)
-        data = dict(fields)
+        builtin_url = self._search_regex(
+            r'<iframe[^>]+src=(["\'])(?P<url>.+?/builtin-.+?)\1',
+            orig_webpage, 'builtin URL', default=None, group='url')
 
-        post = compat_urllib_parse.urlencode(data)
-        headers = {
-            b'Content-Type': b'application/x-www-form-urlencoded',
-        }
-        req = compat_urllib_request.Request(url, post, headers)
-        webpage = self._download_webpage(
-            req, video_id, note='Downloading video page ...')
+        if builtin_url:
+            req = compat_urllib_request.Request(builtin_url)
+            req.add_header('Referer', url)
+            webpage = self._download_webpage(req, video_id, 'Downloading builtin page')
+            title = self._og_search_title(orig_webpage).strip()
+            description = self._og_search_description(orig_webpage).strip()
+        else:
+            fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage)
+            data = dict(fields)
 
-        title = os.path.splitext(data['fname'])[0]
+            post = compat_urllib_parse.urlencode(data)
+            headers = {
+                b'Content-Type': b'application/x-www-form-urlencoded',
+            }
+            req = compat_urllib_request.Request(url, post, headers)
+            webpage = self._download_webpage(
+                req, video_id, note='Downloading video page ...')
+
+            title = os.path.splitext(data['fname'])[0]
+            description = None
 
         # Could be several links with different quality
         links = re.findall(r'"file" : "?(.+?)",', webpage)
@@ -89,5 +113,6 @@ class MonikerIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
+            'description': description,
             'formats': formats,
         }

youtube_dl/extractor/movieclips.py

@@ -1,80 +1,42 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..compat import (
-    compat_str,
-)
-from ..utils import (
-    ExtractorError,
-    clean_html,
+    compat_urllib_request,
 )
 
 
 class MovieClipsIE(InfoExtractor):
-    _VALID_URL = r'https?://movieclips\.com/(?P<id>[\da-zA-Z]+)(?:-(?P<display_id>[\da-z-]+))?'
+    _VALID_URL = r'https?://(?:www.)?movieclips\.com/videos/(?P<id>[^/?#]+)'
     _TEST = {
-        'url': 'http://movieclips.com/Wy7ZU-my-week-with-marilyn-movie-do-you-love-me/',
+        'url': 'http://www.movieclips.com/videos/warcraft-trailer-1-561180739597?autoPlay=true&playlistId=5',
         'info_dict': {
-            'id': 'Wy7ZU',
-            'display_id': 'my-week-with-marilyn-movie-do-you-love-me',
+            'id': 'pKIGmG83AqD9',
+            'display_id': 'warcraft-trailer-1-561180739597',
             'ext': 'mp4',
-            'title': 'My Week with Marilyn - Do You Love Me?',
-            'description': 'md5:e86795bd332fe3cff461e7c8dc542acb',
+            'title': 'Warcraft Trailer 1',
+            'description': 'Watch Trailer 1 from Warcraft (2016). Legendary’s WARCRAFT is a 3D epic adventure of world-colliding conflict based.',
             'thumbnail': 're:^https?://.*\.jpg$',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        }
+        'add_ie': ['ThePlatform'],
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id')
-        show_id = display_id or video_id
+        display_id = self._match_id(url)
 
-        config = self._download_xml(
-            'http://config.movieclips.com/player/config/%s' % video_id,
-            show_id, 'Downloading player config')
-
-        if config.find('./country-region').text == 'false':
-            raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, config.find('./region_alert').text), expected=True)
-
-        properties = config.find('./video/properties')
-        smil_file = properties.attrib['smil_file']
-
-        smil = self._download_xml(smil_file, show_id, 'Downloading SMIL')
-        base_url = smil.find('./head/meta').attrib['base']
-
-        formats = []
-        for video in smil.findall('./body/switch/video'):
-            vbr = int(video.attrib['system-bitrate']) / 1000
-            src = video.attrib['src']
-            formats.append({
-                'url': base_url,
-                'play_path': src,
-                'ext': src.split(':')[0],
-                'vbr': vbr,
-                'format_id': '%dk' % vbr,
-            })
-
-        self._sort_formats(formats)
-
-        title = '%s - %s' % (properties.attrib['clip_movie_title'], properties.attrib['clip_title'])
-        description = clean_html(compat_str(properties.attrib['clip_description']))
-        thumbnail = properties.attrib['image']
-        categories = properties.attrib['clip_categories'].split(',')
+        req = compat_urllib_request.Request(url)
+        # it doesn't work if it thinks the browser it's too old
+        req.add_header('User-Agent', 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/43.0 (Chrome)')
+        webpage = self._download_webpage(req, display_id)
+        theplatform_link = self._html_search_regex(r'src="(http://player.theplatform.com/p/.*?)"', webpage, 'theplatform link')
+        title = self._html_search_regex(r'<title[^>]*>([^>]+)-\s*\d+\s*|\s*Movieclips.com</title>', webpage, 'title')
+        description = self._html_search_meta('description', webpage)
 
         return {
-            'id': video_id,
-            'display_id': display_id,
+            '_type': 'url_transparent',
+            'url': theplatform_link,
             'title': title,
+            'display_id': display_id,
             'description': description,
-            'thumbnail': thumbnail,
-            'categories': categories,
-            'formats': formats,
         }

youtube_dl/extractor/ndr.py

@@ -14,7 +14,8 @@ from ..utils import (
 
 class NDRBaseIE(InfoExtractor):
     def _real_extract(self, url):
-        display_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        display_id = next(group for group in mobj.groups() if group)
         webpage = self._download_webpage(url, display_id)
         return self._extract_embed(webpage, display_id)
 
@@ -22,7 +23,7 @@ class NDRBaseIE(InfoExtractor):
 class NDRIE(NDRBaseIE):
     IE_NAME = 'ndr'
     IE_DESC = 'NDR.de - Norddeutscher Rundfunk'
-    _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)+(?P<id>[^/?#]+),[\da-z]+\.html'
+    _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)*(?P<id>[^/?#]+),[\da-z]+\.html'
     _TESTS = [{
         # httpVideo, same content id
         'url': 'http://www.ndr.de/fernsehen/Party-Poette-und-Parade,hafengeburtstag988.html',
@@ -77,6 +78,9 @@ class NDRIE(NDRBaseIE):
         'params': {
             'skip_download': True,
         },
+    }, {
+        'url': 'https://www.ndr.de/Fettes-Brot-Ferris-MC-und-Thees-Uhlmann-live-on-stage,festivalsommer116.html',
+        'only_matching': True,
     }]
 
     def _extract_embed(self, webpage, display_id):
@@ -101,7 +105,7 @@ class NDRIE(NDRBaseIE):
 class NJoyIE(NDRBaseIE):
     IE_NAME = 'njoy'
     IE_DESC = 'N-JOY'
-    _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)+(?P<id>[^/?#]+),[\da-z]+\.html'
+    _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)*(?:(?P<display_id>[^/?#]+),)?(?P<id>[\da-z]+)\.html'
     _TESTS = [{
         # httpVideo, same content id
         'url': 'http://www.n-joy.de/entertainment/comedy/comedy_contest/Benaissa-beim-NDR-Comedy-Contest,comedycontest2480.html',
@@ -136,6 +140,9 @@ class NJoyIE(NDRBaseIE):
         'params': {
             'skip_download': True,
         },
+    }, {
+        'url': 'http://www.n-joy.de/radio/webradio/morningshow209.html',
+        'only_matching': True,
     }]
 
     def _extract_embed(self, webpage, display_id):
@@ -231,7 +238,7 @@ class NDREmbedBaseIE(InfoExtractor):
 
 class NDREmbedIE(NDREmbedBaseIE):
     IE_NAME = 'ndr:embed'
-    _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)+(?P<id>[\da-z]+)-(?:player|externalPlayer)\.html'
+    _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)*(?P<id>[\da-z]+)-(?:player|externalPlayer)\.html'
     _TESTS = [{
         'url': 'http://www.ndr.de/fernsehen/sendungen/ndr_aktuell/ndraktuell28488-player.html',
         'md5': '8b9306142fe65bbdefb5ce24edb6b0a9',
@@ -325,7 +332,7 @@ class NDREmbedIE(NDREmbedBaseIE):
 
 class NJoyEmbedIE(NDREmbedBaseIE):
     IE_NAME = 'njoy:embed'
-    _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)+(?P<id>[\da-z]+)-(?:player|externalPlayer)_[^/]+\.html'
+    _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)*(?P<id>[\da-z]+)-(?:player|externalPlayer)_[^/]+\.html'
     _TESTS = [{
         # httpVideo
         'url': 'http://www.n-joy.de/events/reeperbahnfestival/doku948-player_image-bc168e87-5263-4d6d-bd27-bb643005a6de_theme-n-joy.html',

youtube_dl/extractor/novamov.py

@@ -4,10 +4,14 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_urllib_request,
     compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
+    NO_DEFAULT,
+    encode_dict,
+    urlencode_postdata,
 )
 
 
@@ -38,19 +42,40 @@ class NovaMovIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
-        page = self._download_webpage(
-            'http://%s/video/%s' % (self._HOST, video_id), video_id, 'Downloading video page')
+        url = 'http://%s/video/%s' % (self._HOST, video_id)
 
-        if re.search(self._FILE_DELETED_REGEX, page) is not None:
+        webpage = self._download_webpage(
+            url, video_id, 'Downloading video page')
+
+        if re.search(self._FILE_DELETED_REGEX, webpage) is not None:
             raise ExtractorError('Video %s does not exist' % video_id, expected=True)
 
-        filekey = self._search_regex(self._FILEKEY_REGEX, page, 'filekey')
+        def extract_filekey(default=NO_DEFAULT):
+            return self._search_regex(
+                self._FILEKEY_REGEX, webpage, 'filekey', default=default)
 
-        title = self._html_search_regex(self._TITLE_REGEX, page, 'title', fatal=False)
-        description = self._html_search_regex(self._DESCRIPTION_REGEX, page, 'description', default='', fatal=False)
+        filekey = extract_filekey(default=None)
+
+        if not filekey:
+            fields = self._hidden_inputs(webpage)
+            post_url = self._search_regex(
+                r'<form[^>]+action=(["\'])(?P<url>.+?)\1', webpage,
+                'post url', default=url, group='url')
+            if not post_url.startswith('http'):
+                post_url = compat_urlparse.urljoin(url, post_url)
+            request = compat_urllib_request.Request(
+                post_url, urlencode_postdata(encode_dict(fields)))
+            request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+            request.add_header('Referer', post_url)
+            webpage = self._download_webpage(
+                request, video_id, 'Downloading continue to the video page')
+
+        filekey = extract_filekey()
+
+        title = self._html_search_regex(self._TITLE_REGEX, webpage, 'title', fatal=False)
+        description = self._html_search_regex(self._DESCRIPTION_REGEX, webpage, 'description', default='', fatal=False)
 
         api_response = self._download_webpage(
             'http://%s/api/player.api.php?key=%s&file=%s' % (self._HOST, filekey, video_id), video_id,

youtube_dl/extractor/nowvideo.py

@@ -7,9 +7,9 @@ class NowVideoIE(NovaMovIE):
     IE_NAME = 'nowvideo'
     IE_DESC = 'NowVideo'
 
-    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:ch|ec|sx|eu|at|ag|co|li)'}
+    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:to|ch|ec|sx|eu|at|ag|co|li)'}
 
-    _HOST = 'www.nowvideo.ch'
+    _HOST = 'www.nowvideo.to'
 
     _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<'
     _FILEKEY_REGEX = r'var fkzd="([^"]+)";'

youtube_dl/extractor/pbs.py

@@ -8,6 +8,7 @@ from ..utils import (
     ExtractorError,
     determine_ext,
     int_or_none,
+    strip_jsonp,
     unified_strdate,
     US_RATINGS,
 )
@@ -153,6 +154,22 @@ class PBSIE(InfoExtractor):
             'params': {
                 'skip_download': True,  # requires ffmpeg
             },
+        },
+        {
+            # Frontline video embedded via flp2012.js
+            'url': 'http://www.pbs.org/wgbh/pages/frontline/the-atomic-artists',
+            'info_dict': {
+                'id': '2070868960',
+                'display_id': 'the-atomic-artists',
+                'ext': 'mp4',
+                'title': 'FRONTLINE - The Atomic Artists',
+                'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
+                'duration': 723,
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         }
     ]
     _ERRORS = {
@@ -191,9 +208,30 @@ class PBSIE(InfoExtractor):
             if media_id:
                 return media_id, presumptive_id, upload_date
 
-            url = self._search_regex(
-                r'(?s)<iframe[^>]+?(?:[a-z-]+?=["\'].*?["\'][^>]+?)*?\bsrc=["\']([^\'"]+partnerplayer[^\'"]+)["\']',
-                webpage, 'player URL')
+            # Fronline video embedded via flp
+            video_id = self._search_regex(
+                r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid', default=None)
+            if video_id:
+                # pkg_id calculation is reverse engineered from
+                # http://www.pbs.org/wgbh/pages/frontline/js/flp2012.js
+                prg_id = self._search_regex(
+                    r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid')[7:]
+                if 'q' in prg_id:
+                    prg_id = prg_id.split('q')[1]
+                prg_id = int(prg_id, 16)
+                getdir = self._download_json(
+                    'http://www.pbs.org/wgbh/pages/frontline/.json/getdir/getdir%d.json' % prg_id,
+                    presumptive_id, 'Downloading getdir JSON',
+                    transform_source=strip_jsonp)
+                return getdir['mid'], presumptive_id, upload_date
+
+            for iframe in re.findall(r'(?s)<iframe(.+?)></iframe>', webpage):
+                url = self._search_regex(
+                    r'src=(["\'])(?P<url>.+?partnerplayer.+?)\1', iframe,
+                    'player URL', default=None, group='url')
+                if url:
+                    break
+
             mobj = re.match(self._VALID_URL, url)
 
         player_id = mobj.group('player_id')

youtube_dl/extractor/periscope.py

@@ -12,7 +12,8 @@ from ..utils import parse_iso8601
 class PeriscopeIE(InfoExtractor):
     IE_DESC = 'Periscope'
     _VALID_URL = r'https?://(?:www\.)?periscope\.tv/w/(?P<id>[^/?#]+)'
-    _TEST = {
+    # Alive example URLs can be found here http://onperiscope.com/
+    _TESTS = [{
         'url': 'https://www.periscope.tv/w/aJUQnjY3MjA3ODF8NTYxMDIyMDl2zCg2pECBgwTqRpQuQD352EMPTKQjT4uqlM3cgWFA-g==',
         'md5': '65b57957972e503fcbbaeed8f4fa04ca',
         'info_dict': {
@@ -25,11 +26,15 @@ class PeriscopeIE(InfoExtractor):
             'uploader_id': '1465763',
         },
         'skip': 'Expires in 24 hours',
-    }
+    }, {
+        'url': 'https://www.periscope.tv/w/1ZkKzPbMVggJv',
+        'only_matching': True,
+    }]
 
-    def _call_api(self, method, token):
+    def _call_api(self, method, value):
+        attribute = 'token' if len(value) > 13 else 'broadcast_id'
         return self._download_json(
-            'https://api.periscope.tv/api/v2/%s?token=%s' % (method, token), token)
+            'https://api.periscope.tv/api/v2/%s?%s=%s' % (method, attribute, value), value)
 
     def _real_extract(self, url):
         token = self._match_id(url)

youtube_dl/extractor/prosiebensat1.py

@@ -20,7 +20,7 @@ from ..utils import (
 class ProSiebenSat1IE(InfoExtractor):
     IE_NAME = 'prosiebensat1'
     IE_DESC = 'ProSiebenSat.1 Digital'
-    _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at)|ran\.de|fem\.com)/(?P<id>.+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P<id>.+)'
 
     _TESTS = [
         {

youtube_dl/extractor/rtbf.py

@@ -9,8 +9,8 @@ from ..utils import (
 
 
 class RTBFIE(InfoExtractor):
-    _VALID_URL = r'https?://www.rtbf.be/video/[^\?]+\?id=(?P<id>\d+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?rtbf\.be/(?:video/[^?]+\?.*\bid=|ouftivi/(?:[^/]+/)*[^?]+\?.*\bvideoId=)(?P<id>\d+)'
+    _TESTS = [{
         'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274',
         'md5': '799f334ddf2c0a582ba80c44655be570',
         'info_dict': {
@@ -19,7 +19,14 @@ class RTBFIE(InfoExtractor):
             'title': 'Les Diables au coeur (épisode 2)',
             'duration': 3099,
         }
-    }
+    }, {
+        # geo restricted
+        'url': 'http://www.rtbf.be/ouftivi/heros/detail_scooby-doo-mysteres-associes?id=1097&videoId=2057442',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.rtbf.be/ouftivi/niouzz?videoId=2055858',
+        'only_matching': True,
+    }]
 
     _QUALITIES = [
         ('mobile', 'mobile'),

youtube_dl/extractor/ruutu.py

@@ -57,16 +57,21 @@ class RuutuIE(InfoExtractor):
                     extract_formats(child)
                 elif child.tag.endswith('File'):
                     video_url = child.text
-                    if not video_url or video_url in processed_urls or 'NOT_USED' in video_url:
+                    if (not video_url or video_url in processed_urls or
+                            any(p in video_url for p in ('NOT_USED', 'NOT-USED'))):
                         return
                     processed_urls.append(video_url)
                     ext = determine_ext(video_url)
                     if ext == 'm3u8':
-                        formats.extend(self._extract_m3u8_formats(
-                            video_url, video_id, 'mp4', m3u8_id='hls'))
+                        m3u8_formats = self._extract_m3u8_formats(
+                            video_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
+                        if m3u8_formats:
+                            formats.extend(m3u8_formats)
                     elif ext == 'f4m':
-                        formats.extend(self._extract_f4m_formats(
-                            video_url, video_id, f4m_id='hds'))
+                        f4m_formats = self._extract_f4m_formats(
+                            video_url, video_id, f4m_id='hds', fatal=False)
+                        if f4m_formats:
+                            formats.extend(f4m_formats)
                     else:
                         proto = compat_urllib_parse_urlparse(video_url).scheme
                         if not child.tag.startswith('HTTP') and proto != 'rtmp':

youtube_dl/extractor/senateisvp.py

@@ -121,9 +121,9 @@ class SenateISVPIE(InfoExtractor):
                 'url': compat_urlparse.urljoin(domain, filename) + '?v=3.1.0&fp=&r=&g=',
             }]
         else:
-            hdcore_sign = '?hdcore=3.1.0'
+            hdcore_sign = 'hdcore=3.1.0'
             url_params = (domain, video_id, stream_num)
-            f4m_url = '%s/z/%s_1@%s/manifest.f4m' % url_params + hdcore_sign
+            f4m_url = '%s/z/%s_1@%s/manifest.f4m?' % url_params + hdcore_sign
             m3u8_url = '%s/i/%s_1@%s/master.m3u8' % url_params
             for entry in self._extract_f4m_formats(f4m_url, video_id, f4m_id='f4m'):
                 # URLs without the extra param induce an 404 error

youtube_dl/extractor/spiegeltv.py

@@ -77,17 +77,21 @@ class SpiegeltvIE(InfoExtractor):
                     'rtmp_live': True,
                 })
             elif determine_ext(endpoint) == 'm3u8':
-                m3u8_formats = self._extract_m3u8_formats(
-                    endpoint.replace('[video]', play_path),
-                    video_id, 'm4v',
-                    preference=1,  # Prefer hls since it allows to workaround georestriction
-                    m3u8_id='hls', fatal=False)
-                if m3u8_formats is not False:
-                    formats.extend(m3u8_formats)
+                formats.append({
+                    'url': endpoint.replace('[video]', play_path),
+                    'ext': 'm4v',
+                    'format_id': 'hls',  # Prefer hls since it allows to workaround georestriction
+                    'protocol': 'm3u8',
+                    'preference': 1,
+                    'http_headers': {
+                        'Accept-Encoding': 'deflate',  # gzip causes trouble on the server side
+                    },
+                })
             else:
                 formats.append({
                     'url': endpoint,
                 })
+        self._check_formats(formats, video_id)
 
         thumbnails = []
         for image in media_json['images']:

youtube_dl/extractor/stitcher.py

@@ -64,7 +64,7 @@ class StitcherIE(InfoExtractor):
             'url': episode[episode_key],
             'ext': determine_ext(episode[episode_key]) or 'mp3',
             'vcodec': 'none',
-        } for episode_key in ('origEpisodeURL', 'episodeURL') if episode.get(episode_key)]
+        } for episode_key in ('episodeURL',) if episode.get(episode_key)]
         description = self._search_regex(
             r'Episode Info:\s*</span>([^<]+)<', webpage, 'description', fatal=False)
         duration = int_or_none(episode.get('duration'))

youtube_dl/extractor/tutv.py

@@ -10,10 +10,10 @@ class TutvIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?tu\.tv/videos/(?P<id>[^/?]+)'
     _TEST = {
         'url': 'http://tu.tv/videos/robots-futbolistas',
-        'md5': '627c7c124ac2a9b5ab6addb94e0e65f7',
+        'md5': '0cd9e28ad270488911b0d2a72323395d',
         'info_dict': {
             'id': '2973058',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Robots futbolistas',
         },
     }

youtube_dl/extractor/twitter.py

@@ -9,6 +9,8 @@ from ..utils import (
     float_or_none,
     xpath_text,
     remove_end,
+    int_or_none,
+    ExtractorError,
 )
 
 
@@ -120,7 +122,7 @@ class TwitterIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)'
     _TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'https://twitter.com/freethenipple/status/643211948184596480',
         'md5': '31cd83a116fc41f99ae3d909d4caf6a0',
         'info_dict': {
@@ -133,7 +135,19 @@ class TwitterIE(InfoExtractor):
             'uploader': 'FREE THE NIPPLE',
             'uploader_id': 'freethenipple',
         },
-    }
+    }, {
+        'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
+        'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
+        'info_dict': {
+            'id': '657991469417025536',
+            'ext': 'mp4',
+            'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai',
+            'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"',
+            'thumbnail': 're:^https?://.*\.png',
+            'uploader': 'Gifs',
+            'uploader_id': 'giphz',
+        },
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -150,17 +164,41 @@ class TwitterIE(InfoExtractor):
         mobj = re.match(r'“(.*)\s+(https?://[^ ]+)”', title)
         title, short_url = mobj.groups()
 
-        card_id = self._search_regex(
-            r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url')
-        card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
-
-        return {
-            '_type': 'url_transparent',
-            'ie_key': 'TwitterCard',
+        info = {
             'uploader_id': user_id,
             'uploader': username,
-            'url': card_url,
             'webpage_url': url,
             'description': '%s on Twitter: "%s %s"' % (username, title, short_url),
             'title': username + ' - ' + title,
         }
+
+        card_id = self._search_regex(
+            r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url', default=None)
+        if card_id:
+            card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
+            info.update({
+                '_type': 'url_transparent',
+                'ie_key': 'TwitterCard',
+                'url': card_url,
+            })
+            return info
+
+        mobj = re.search(r'''(?x)
+            <video[^>]+class="animated-gif"[^>]+
+                (?:data-height="(?P<height>\d+)")?[^>]+
+                (?:data-width="(?P<width>\d+)")?[^>]+
+                (?:poster="(?P<poster>[^"]+)")?[^>]*>\s*
+                <source[^>]+video-src="(?P<url>[^"]+)"
+        ''', webpage)
+
+        if mobj:
+            info.update({
+                'id': twid,
+                'url': mobj.group('url'),
+                'height': int_or_none(mobj.group('height')),
+                'width': int_or_none(mobj.group('width')),
+                'thumbnail': mobj.group('poster'),
+            })
+            return info
+
+        raise ExtractorError('There\'s not video in this tweet.')

youtube_dl/extractor/vevo.py

@@ -1,10 +1,10 @@
 from __future__ import unicode_literals
 
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_etree_fromstring,
     compat_urllib_request,
 )
 from ..utils import (
@@ -97,7 +97,7 @@ class VevoIE(InfoExtractor):
         if last_version['version'] == -1:
             raise ExtractorError('Unable to extract last version of the video')
 
-        renditions = xml.etree.ElementTree.fromstring(last_version['data'])
+        renditions = compat_etree_fromstring(last_version['data'])
         formats = []
         # Already sorted from worst to best quality
         for rend in renditions.findall('rendition'):
@@ -114,7 +114,7 @@ class VevoIE(InfoExtractor):
 
     def _formats_from_smil(self, smil_xml):
         formats = []
-        smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8'))
+        smil_doc = compat_etree_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']

youtube_dl/extractor/videofyme.py

@@ -2,8 +2,8 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
-    find_xpath_attr,
     int_or_none,
+    parse_iso8601,
 )
 
 
@@ -18,33 +18,35 @@ class VideofyMeIE(InfoExtractor):
             'id': '1100701',
             'ext': 'mp4',
             'title': 'This is VideofyMe',
-            'description': None,
+            'description': '',
+            'upload_date': '20130326',
+            'timestamp': 1364288959,
             'uploader': 'VideofyMe',
             'uploader_id': 'thisisvideofyme',
             'view_count': int,
+            'likes': int,
+            'comment_count': int,
         },
-
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        config = self._download_xml('http://sunshine.videofy.me/?videoId=%s' % video_id,
-                                    video_id)
-        video = config.find('video')
-        sources = video.find('sources')
-        url_node = next(node for node in [find_xpath_attr(sources, 'source', 'id', 'HQ %s' % key)
-                                          for key in ['on', 'av', 'off']] if node is not None)
-        video_url = url_node.find('url').text
-        view_count = int_or_none(self._search_regex(
-            r'([0-9]+)', video.find('views').text, 'view count', fatal=False))
+
+        config = self._download_json('http://vf-player-info-loader.herokuapp.com/%s.json' % video_id, video_id)['videoinfo']
+
+        video = config.get('video')
+        blog = config.get('blog', {})
 
         return {
             'id': video_id,
-            'title': video.find('title').text,
-            'url': video_url,
-            'thumbnail': video.find('thumb').text,
-            'description': video.find('description').text,
-            'uploader': config.find('blog/name').text,
-            'uploader_id': video.find('identifier').text,
-            'view_count': view_count,
+            'title': video['title'],
+            'url': video['sources']['source']['url'],
+            'thumbnail': video.get('thumb'),
+            'description': video.get('description'),
+            'timestamp': parse_iso8601(video.get('date')),
+            'uploader': blog.get('name'),
+            'uploader_id': blog.get('identifier'),
+            'view_count': int_or_none(self._search_regex(r'([0-9]+)', video.get('views'), 'view count', fatal=False)),
+            'likes': int_or_none(video.get('likes')),
+            'comment_count': int_or_none(video.get('nrOfComments')),
         }

youtube_dl/extractor/videolecturesnet.py

@@ -1,82 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..compat import (
-    compat_HTTPError,
-    compat_urlparse,
-)
-from ..utils import (
-    ExtractorError,
-    parse_duration,
-)
-
-
-class VideoLecturesNetIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/*(?:[#?].*)?$'
-    IE_NAME = 'videolectures.net'
-
-    _TESTS = [{
-        'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/',
-        'info_dict': {
-            'id': 'promogram_igor_mekjavic_eng',
-            'ext': 'mp4',
-            'title': 'Automatics, robotics and biocybernetics',
-            'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
-            'upload_date': '20130627',
-            'duration': 565,
-            'thumbnail': 're:http://.*\.jpg',
-        },
-    }, {
-        # video with invalid direct format links (HTTP 403)
-        'url': 'http://videolectures.net/russir2010_filippova_nlp/',
-        'info_dict': {
-            'id': 'russir2010_filippova_nlp',
-            'ext': 'flv',
-            'title': 'NLP at Google',
-            'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3',
-            'duration': 5352,
-            'thumbnail': 're:http://.*\.jpg',
-        },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://videolectures.net/deeplearning2015_montreal/',
-        'info_dict': {
-            'id': 'deeplearning2015_montreal',
-            'title': 'Deep Learning Summer School, Montreal 2015',
-            'description': 'md5:90121a40cc6926df1bf04dcd8563ed3b',
-        },
-        'playlist_count': 30,
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id
-
-        try:
-            smil = self._download_smil(smil_url, video_id)
-        except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
-                # Probably a playlist
-                webpage = self._download_webpage(url, video_id)
-                entries = [
-                    self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet')
-                    for _, video_url in re.findall(r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)]
-                playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True)
-                playlist_description = self._html_search_meta('description', webpage, 'description')
-                return self.playlist_result(entries, video_id, playlist_title, playlist_description)
-
-        info = self._parse_smil(smil, smil_url, video_id)
-
-        info['id'] = video_id
-
-        switch = smil.find('.//switch')
-        if switch is not None:
-            info['duration'] = parse_duration(switch.attrib.get('dur'))
-
-        return info

youtube_dl/extractor/vidme.py

@@ -101,6 +101,10 @@ class VidmeIE(InfoExtractor):
         # suspended
         'url': 'https://vid.me/Ox3G',
         'only_matching': True,
+    }, {
+        # deleted
+        'url': 'https://vid.me/KTPm',
+        'only_matching': True,
     }, {
         # no formats in the API response
         'url': 'https://vid.me/e5g',
@@ -143,6 +147,11 @@ class VidmeIE(InfoExtractor):
 
         video = response['video']
 
+        if video.get('state') == 'deleted':
+            raise ExtractorError(
+                'Vidme said: Sorry, this video has been deleted.',
+                expected=True)
+
         if video.get('state') in ('user-disabled', 'suspended'):
             raise ExtractorError(
                 'Vidme said: This video has been suspended either due to a copyright claim, '

youtube_dl/extractor/vidzi.py

@@ -20,8 +20,14 @@ class VidziIE(InfoExtractor):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
-        video_url = self._html_search_regex(
-            r'{\s*file\s*:\s*"([^"]+)"\s*}', webpage, 'video url')
+        video_host = self._html_search_regex(
+            r'id=\'vplayer\'><img src="http://(.*?)/i', webpage,
+            'video host')
+        video_hash = self._html_search_regex(
+            r'\|([a-z0-9]+)\|hls\|type', webpage, 'video_hash')
+        ext = self._html_search_regex(
+            r'\|tracks\|([a-z0-9]+)\|', webpage, 'video ext')
+        video_url = 'http://' + video_host + '/' + video_hash + '/v.' + ext
         title = self._html_search_regex(
             r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
 

youtube_dl/extractor/viidea.py

@@ -0,0 +1,188 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urlparse,
+    compat_str,
+)
+from ..utils import (
+    parse_duration,
+    js_to_json,
+    parse_iso8601,
+)
+
+
+class ViideaIE(InfoExtractor):
+    _VALID_URL = r'''(?x)http://(?:www\.)?(?:
+            videolectures\.net|
+            flexilearn\.viidea\.net|
+            presentations\.ocwconsortium\.org|
+            video\.travel-zoom\.si|
+            video\.pomp-forum\.si|
+            tv\.nil\.si|
+            video\.hekovnik.com|
+            video\.szko\.si|
+            kpk\.viidea\.com|
+            inside\.viidea\.net|
+            video\.kiberpipa\.org|
+            bvvideo\.si|
+            kongres\.viidea\.net|
+            edemokracija\.viidea\.com
+        )(?:/lecture)?/(?P<id>[^/]+)(?:/video/(?P<part>\d+))?/*(?:[#?].*)?$'''
+
+    _TESTS = [{
+        'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/',
+        'info_dict': {
+            'id': '20171',
+            'display_id': 'promogram_igor_mekjavic_eng',
+            'ext': 'mp4',
+            'title': 'Automatics, robotics and biocybernetics',
+            'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
+            'thumbnail': 're:http://.*\.jpg',
+            'timestamp': 1372349289,
+            'upload_date': '20130627',
+            'duration': 565,
+        },
+    }, {
+        # video with invalid direct format links (HTTP 403)
+        'url': 'http://videolectures.net/russir2010_filippova_nlp/',
+        'info_dict': {
+            'id': '14891',
+            'display_id': 'russir2010_filippova_nlp',
+            'ext': 'flv',
+            'title': 'NLP at Google',
+            'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3',
+            'thumbnail': 're:http://.*\.jpg',
+            'timestamp': 1284375600,
+            'upload_date': '20100913',
+            'duration': 5352,
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        # event playlist
+        'url': 'http://videolectures.net/deeplearning2015_montreal/',
+        'info_dict': {
+            'id': '23181',
+            'title': 'Deep Learning Summer School, Montreal 2015',
+            'description': 'md5:0533a85e4bd918df52a01f0e1ebe87b7',
+            'thumbnail': 're:http://.*\.jpg',
+            'timestamp': 1438560000,
+        },
+        'playlist_count': 30,
+    }, {
+        # multi part lecture
+        'url': 'http://videolectures.net/mlss09uk_bishop_ibi/',
+        'info_dict': {
+            'id': '9737',
+            'display_id': 'mlss09uk_bishop_ibi',
+            'title': 'Introduction To Bayesian Inference',
+            'thumbnail': 're:http://.*\.jpg',
+            'timestamp': 1251622800,
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': '9737_part1',
+                'display_id': 'mlss09uk_bishop_ibi_part1',
+                'ext': 'wmv',
+                'title': 'Introduction To Bayesian Inference (Part 1)',
+                'thumbnail': 're:http://.*\.jpg',
+                'duration': 4622,
+                'timestamp': 1251622800,
+                'upload_date': '20090830',
+            },
+        }, {
+            'info_dict': {
+                'id': '9737_part2',
+                'display_id': 'mlss09uk_bishop_ibi_part2',
+                'ext': 'wmv',
+                'title': 'Introduction To Bayesian Inference (Part 2)',
+                'thumbnail': 're:http://.*\.jpg',
+                'duration': 5641,
+                'timestamp': 1251622800,
+                'upload_date': '20090830',
+            },
+        }],
+        'playlist_count': 2,
+    }]
+
+    def _real_extract(self, url):
+        lecture_slug, explicit_part_id = re.match(self._VALID_URL, url).groups()
+
+        webpage = self._download_webpage(url, lecture_slug)
+
+        cfg = self._parse_json(self._search_regex(
+            [r'cfg\s*:\s*({.+?})\s*,\s*[\da-zA-Z_]+\s*:\s*\(?\s*function',
+             r'cfg\s*:\s*({[^}]+})'],
+            webpage, 'cfg'), lecture_slug, js_to_json)
+
+        lecture_id = compat_str(cfg['obj_id'])
+
+        base_url = self._proto_relative_url(cfg['livepipe'], 'http:')
+
+        lecture_data = self._download_json(
+            '%s/site/api/lecture/%s?format=json' % (base_url, lecture_id),
+            lecture_id)['lecture'][0]
+
+        lecture_info = {
+            'id': lecture_id,
+            'display_id': lecture_slug,
+            'title': lecture_data['title'],
+            'timestamp': parse_iso8601(lecture_data.get('time')),
+            'description': lecture_data.get('description_wiki'),
+            'thumbnail': lecture_data.get('thumb'),
+        }
+
+        playlist_entries = []
+        lecture_type = lecture_data.get('type')
+        parts = [compat_str(video) for video in cfg.get('videos', [])]
+        if parts:
+            multipart = len(parts) > 1
+
+            def extract_part(part_id):
+                smil_url = '%s/%s/video/%s/smil.xml' % (base_url, lecture_slug, part_id)
+                smil = self._download_smil(smil_url, lecture_id)
+                info = self._parse_smil(smil, smil_url, lecture_id)
+                info['id'] = lecture_id if not multipart else '%s_part%s' % (lecture_id, part_id)
+                info['display_id'] = lecture_slug if not multipart else '%s_part%s' % (lecture_slug, part_id)
+                if multipart:
+                    info['title'] += ' (Part %s)' % part_id
+                switch = smil.find('.//switch')
+                if switch is not None:
+                    info['duration'] = parse_duration(switch.attrib.get('dur'))
+                item_info = lecture_info.copy()
+                item_info.update(info)
+                return item_info
+
+            if explicit_part_id or not multipart:
+                result = extract_part(explicit_part_id or parts[0])
+            else:
+                result = {
+                    '_type': 'multi_video',
+                    'entries': [extract_part(part) for part in parts],
+                }
+                result.update(lecture_info)
+
+            # Immediately return explicitly requested part or non event item
+            if explicit_part_id or lecture_type != 'evt':
+                return result
+
+            playlist_entries.append(result)
+
+        # It's probably a playlist
+        if not parts or lecture_type == 'evt':
+            playlist_webpage = self._download_webpage(
+                '%s/site/ajax/drilldown/?id=%s' % (base_url, lecture_id), lecture_id)
+            entries = [
+                self.url_result(compat_urlparse.urljoin(url, video_url), 'Viidea')
+                for _, video_url in re.findall(
+                    r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', playlist_webpage)]
+            playlist_entries.extend(entries)
+
+        playlist = self.playlist_result(playlist_entries, lecture_id)
+        playlist.update(lecture_info)
+        return playlist

youtube_dl/extractor/vimeo.py

@@ -8,11 +8,11 @@ import itertools
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
-    compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
 )
 from ..utils import (
+    encode_dict,
     ExtractorError,
     InAdvancePagedList,
     int_or_none,
@@ -40,17 +40,17 @@ class VimeoBaseInfoExtractor(InfoExtractor):
         self.report_login()
         webpage = self._download_webpage(self._LOGIN_URL, None, False)
         token, vuid = self._extract_xsrft_and_vuid(webpage)
-        data = urlencode_postdata({
+        data = urlencode_postdata(encode_dict({
             'action': 'login',
             'email': username,
             'password': password,
             'service': 'vimeo',
             'token': token,
-        })
+        }))
         login_request = compat_urllib_request.Request(self._LOGIN_URL, data)
         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        login_request.add_header('Cookie', 'vuid=%s' % vuid)
         login_request.add_header('Referer', self._LOGIN_URL)
+        self._set_vimeo_cookie('vuid', vuid)
         self._download_webpage(login_request, None, False, 'Wrong login info')
 
     def _extract_xsrft_and_vuid(self, webpage):
@@ -62,6 +62,9 @@ class VimeoBaseInfoExtractor(InfoExtractor):
             webpage, 'vuid', group='vuid')
         return xsrft, vuid
 
+    def _set_vimeo_cookie(self, name, value):
+        self._set_cookie('vimeo.com', name, value)
+
 
 class VimeoIE(VimeoBaseInfoExtractor):
     """Information extractor for vimeo.com."""
@@ -208,17 +211,17 @@ class VimeoIE(VimeoBaseInfoExtractor):
         if password is None:
             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
         token, vuid = self._extract_xsrft_and_vuid(webpage)
-        data = urlencode_postdata({
+        data = urlencode_postdata(encode_dict({
             'password': password,
             'token': token,
-        })
+        }))
         if url.startswith('http://'):
             # vimeo only supports https now, but the user can give an http url
             url = url.replace('http://', 'https://')
         password_request = compat_urllib_request.Request(url + '/password', data)
         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        password_request.add_header('Cookie', 'clip_test2=1; vuid=%s' % vuid)
         password_request.add_header('Referer', url)
+        self._set_vimeo_cookie('vuid', vuid)
         return self._download_webpage(
             password_request, video_id,
             'Verifying the password', 'Wrong password')
@@ -227,7 +230,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
         password = self._downloader.params.get('videopassword', None)
         if password is None:
             raise ExtractorError('This video is protected by a password, use the --video-password option')
-        data = compat_urllib_parse.urlencode({'password': password})
+        data = urlencode_postdata(encode_dict({'password': password}))
         pass_url = url + '/check-password'
         password_request = compat_urllib_request.Request(pass_url, data)
         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
@@ -384,47 +387,29 @@ class VimeoIE(VimeoBaseInfoExtractor):
             like_count = None
             comment_count = None
 
-        # Vimeo specific: extract request signature and timestamp
-        sig = config['request']['signature']
-        timestamp = config['request']['timestamp']
-
-        # Vimeo specific: extract video codec and quality information
-        # First consider quality, then codecs, then take everything
-        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:
-            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[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, quality, codec_name.upper())
-
-                files[key].append({
-                    'ext': codec_extension,
-                    'url': video_url,
-                    'format_id': format_id,
-                    'width': int_or_none(file_info.get('width')),
-                    'height': int_or_none(file_info.get('height')),
-                    'tbr': int_or_none(file_info.get('bitrate')),
-                })
         formats = []
-        m3u8_url = config_files.get('hls', {}).get('all')
+        config_files = config['video'].get('files') or config['request'].get('files', {})
+        for f in config_files.get('progressive', []):
+            video_url = f.get('url')
+            if not video_url:
+                continue
+            formats.append({
+                'url': video_url,
+                'format_id': 'http-%s' % f.get('quality'),
+                'width': int_or_none(f.get('width')),
+                'height': int_or_none(f.get('height')),
+                'fps': int_or_none(f.get('fps')),
+                'tbr': int_or_none(f.get('bitrate')),
+            })
+        m3u8_url = config_files.get('hls', {}).get('url')
         if m3u8_url:
             m3u8_formats = self._extract_m3u8_formats(
                 m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False)
             if m3u8_formats:
                 formats.extend(m3u8_formats)
-        for key in ('other', 'sd', 'hd'):
-            formats += files[key]
-        self._sort_formats(formats)
+        # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
+        # at the same time without actual units specified. This lead to wrong sorting.
+        self._sort_formats(formats, field_preference=('height', 'width', 'fps', 'format_id'))
 
         subtitles = {}
         text_tracks = config['request'].get('text_tracks')
@@ -488,14 +473,14 @@ class VimeoChannelIE(VimeoBaseInfoExtractor):
         token, vuid = self._extract_xsrft_and_vuid(webpage)
         fields['token'] = token
         fields['password'] = password
-        post = urlencode_postdata(fields)
+        post = urlencode_postdata(encode_dict(fields))
         password_path = self._search_regex(
             r'action="([^"]+)"', login_form, 'password URL')
         password_url = compat_urlparse.urljoin(page_url, password_path)
         password_request = compat_urllib_request.Request(password_url, post)
         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
-        password_request.add_header('Cookie', 'vuid=%s' % vuid)
-        self._set_cookie('vimeo.com', 'xsrft', token)
+        self._set_vimeo_cookie('vuid', vuid)
+        self._set_vimeo_cookie('xsrft', token)
 
         return self._download_webpage(
             password_request, list_id,

youtube_dl/extractor/vk.py

@@ -281,9 +281,13 @@ class VKIE(InfoExtractor):
             mobj.group(1) + ' ' + mobj.group(2)
             upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
 
-        view_count = str_to_int(self._search_regex(
-            r'"mv_views_count_number"[^>]*>([\d,.]+) views<',
-            info_page, 'view count', fatal=False))
+        view_count = None
+        views = self._html_search_regex(
+            r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',
+            info_page, 'view count', fatal=False)
+        if views:
+            view_count = str_to_int(self._search_regex(
+                r'([\d,.]+)', views, 'view count', fatal=False))
 
         formats = [{
             'format_id': k,

youtube_dl/extractor/wsj.py

@@ -84,6 +84,5 @@ class WSJIE(InfoExtractor):
             'duration': duration,
             'upload_date': upload_date,
             'title': title,
-            'formats': formats,
             'categories': categories,
         }

youtube_dl/extractor/xfileshare.py

@@ -1,4 +1,4 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -15,11 +15,11 @@ from ..utils import (
 )
 
 
-class GorillaVidIE(InfoExtractor):
-    IE_DESC = 'GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net and filehoot.com'
+class XFileShareIE(InfoExtractor):
+    IE_DESC = 'XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me'
     _VALID_URL = r'''(?x)
         https?://(?P<host>(?:www\.)?
-            (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com))/
+            (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com|vidto.\me))/
         (?:embed-)?(?P<id>[0-9a-zA-Z]+)(?:-[0-9]+x[0-9]+\.html)?
     '''
 
@@ -76,6 +76,13 @@ class GorillaVidIE(InfoExtractor):
             'title': 'youtube-dl test video \'äBaW_jenozKc.mp4.mp4',
             'thumbnail': 're:http://.*\.jpg',
         }
+    }, {
+        'url': 'http://vidto.me/ku5glz52nqe1.html',
+        'info_dict': {
+            'id': 'ku5glz52nqe1',
+            'ext': 'mp4',
+            'title': 'test'
+        }
     }]
 
     def _real_extract(self, url):
@@ -104,13 +111,18 @@ class GorillaVidIE(InfoExtractor):
 
             webpage = self._download_webpage(req, video_id, 'Downloading video page')
 
-        title = self._search_regex(
-            [r'style="z-index: [0-9]+;">([^<]+)</span>', r'<td nowrap>([^<]+)</td>', r'>Watch (.+) '],
-            webpage, 'title', default=None) or self._og_search_title(webpage)
+        title = (self._search_regex(
+            [r'style="z-index: [0-9]+;">([^<]+)</span>',
+             r'<td nowrap>([^<]+)</td>',
+             r'>Watch (.+) ',
+             r'<h2 class="video-page-head">([^<]+)</h2>'],
+            webpage, 'title', default=None) or self._og_search_title(webpage)).strip()
         video_url = self._search_regex(
-            r'file\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'file url')
+            [r'file\s*:\s*["\'](http[^"\']+)["\'],',
+             r'file_link\s*=\s*\'(https?:\/\/[0-9a-zA-z.\/\-_]+)'],
+            webpage, 'file url')
         thumbnail = self._search_regex(
-            r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', fatal=False)
+            r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', default=None)
 
         formats = [{
             'format_id': 'sd',

youtube_dl/extractor/youporn.py

@@ -1,121 +1,171 @@
 from __future__ import unicode_literals
 
-
-import json
 import re
-import sys
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse_urlparse,
-    compat_urllib_request,
-)
+from ..compat import compat_urllib_request
 from ..utils import (
-    ExtractorError,
+    int_or_none,
+    str_to_int,
     unescapeHTML,
     unified_strdate,
 )
-from ..aes import (
-    aes_decrypt_text
-)
+from ..aes import aes_decrypt_text
 
 
 class YouPornIE(InfoExtractor):
-    _VALID_URL = r'^(?P<proto>https?://)(?:www\.)?(?P<url>youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+))'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?youporn\.com/watch/(?P<id>\d+)/(?P<display_id>[^/?#&]+)'
+    _TESTS = [{
         'url': 'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/',
+        'md5': '71ec5fcfddacf80f495efa8b6a8d9a89',
         'info_dict': {
             'id': '505835',
+            'display_id': 'sex-ed-is-it-safe-to-masturbate-daily',
             'ext': 'mp4',
-            'upload_date': '20101221',
-            'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?',
-            'uploader': 'Ask Dan And Jennifer',
             'title': 'Sex Ed: Is It Safe To Masturbate Daily?',
+            'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Ask Dan And Jennifer',
+            'upload_date': '20101221',
+            'average_rating': int,
+            'view_count': int,
+            'comment_count': int,
+            'categories': list,
+            'tags': list,
             'age_limit': 18,
-        }
-    }
+        },
+    }, {
+        # Anonymous User uploader
+        'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4',
+        'info_dict': {
+            'id': '561726',
+            'display_id': 'big-tits-awesome-brunette-on-amazing-webcam-show',
+            'ext': 'mp4',
+            'title': 'Big Tits Awesome Brunette On amazing webcam show',
+            'description': 'http://sweetlivegirls.com Big Tits Awesome Brunette On amazing webcam show.mp4',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Anonymous User',
+            'upload_date': '20111125',
+            'average_rating': int,
+            'view_count': int,
+            'comment_count': int,
+            'categories': list,
+            'tags': list,
+            'age_limit': 18,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
-        url = mobj.group('proto') + 'www.' + mobj.group('url')
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
 
-        req = compat_urllib_request.Request(url)
-        req.add_header('Cookie', 'age_verified=1')
-        webpage = self._download_webpage(req, video_id)
-        age_limit = self._rta_search(webpage)
+        request = compat_urllib_request.Request(url)
+        request.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(request, display_id)
 
-        # Get JSON parameters
-        json_params = self._search_regex(
-            [r'videoJa?son\s*=\s*({.+})',
-             r'var\s+currentVideo\s*=\s*new\s+Video\((.+?)\)[,;]'],
-            webpage, 'JSON parameters')
-        try:
-            params = json.loads(json_params)
-        except ValueError:
-            raise ExtractorError('Invalid JSON')
+        title = self._search_regex(
+            [r'(?:video_titles|videoTitle)\s*[:=]\s*(["\'])(?P<title>.+?)\1',
+             r'<h1[^>]+class=["\']heading\d?["\'][^>]*>([^<])<'],
+            webpage, 'title', group='title')
 
-        self.report_extraction(video_id)
-        try:
-            video_title = params['title']
-            upload_date = unified_strdate(params['release_date_f'])
-            video_description = params['description']
-            video_uploader = params['submitted_by']
-            thumbnail = params['thumbnails'][0]['image']
-        except KeyError:
-            raise ExtractorError('Missing JSON parameter: ' + sys.exc_info()[1])
+        links = []
 
-        # Get all of the links from the page
-        DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>'
-        download_list_html = self._search_regex(DOWNLOAD_LIST_RE,
-                                                webpage, 'download list').strip()
-        LINK_RE = r'<a href="([^"]+)">'
-        links = re.findall(LINK_RE, download_list_html)
+        sources = self._search_regex(
+            r'sources\s*:\s*({.+?})', webpage, 'sources', default=None)
+        if sources:
+            for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources):
+                links.append(link)
 
-        # Get all encrypted links
-        encrypted_links = re.findall(r'var encryptedQuality[0-9]{3}URL = \'([a-zA-Z0-9+/]+={0,2})\';', webpage)
-        for encrypted_link in encrypted_links:
-            link = aes_decrypt_text(encrypted_link, video_title, 32).decode('utf-8')
+        # Fallback #1
+        for _, link in re.findall(
+                r'(?:videoUrl|videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage):
             links.append(link)
 
+        # Fallback #2, this also contains extra low quality 180p format
+        for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage):
+            links.append(link)
+
+        # Fallback #3, encrypted links
+        for _, encrypted_link in re.findall(
+                r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage):
+            links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8'))
+
         formats = []
-        for link in links:
-            # A link looks like this:
-            # http://cdn1.download.youporn.phncdn.com/201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4?nvb=20121113051249&nva=20121114051249&ir=1200&sr=1200&hash=014b882080310e95fb6a0
-            # A path looks like this:
-            # /201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4
-            video_url = unescapeHTML(link)
-            path = compat_urllib_parse_urlparse(video_url).path
-            format_parts = path.split('/')[4].split('_')[:2]
-
-            dn = compat_urllib_parse_urlparse(video_url).netloc.partition('.')[0]
-
-            resolution = format_parts[0]
-            height = int(resolution[:-len('p')])
-            bitrate = int(format_parts[1][:-len('k')])
-            format = '-'.join(format_parts) + '-' + dn
-
-            formats.append({
+        for video_url in set(unescapeHTML(link) for link in links):
+            f = {
                 'url': video_url,
-                'format': format,
-                'format_id': format,
-                'height': height,
-                'tbr': bitrate,
-                'resolution': resolution,
-            })
-
+            }
+            # Video URL's path looks like this:
+            #  /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
+            # We will benefit from it by extracting some metadata
+            mobj = re.search(r'/(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+/', video_url)
+            if mobj:
+                height = int(mobj.group('height'))
+                bitrate = int(mobj.group('bitrate'))
+                f.update({
+                    'format_id': '%dp-%dk' % (height, bitrate),
+                    'height': height,
+                    'tbr': bitrate,
+                })
+            formats.append(f)
         self._sort_formats(formats)
 
-        if not formats:
-            raise ExtractorError('ERROR: no known formats available for video')
+        description = self._html_search_regex(
+            r'(?s)<div[^>]+class=["\']video-description["\'][^>]*>(.+?)</div>',
+            webpage, 'description', default=None)
+        thumbnail = self._search_regex(
+            r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1',
+            webpage, 'thumbnail', fatal=False, group='thumbnail')
+
+        uploader = self._html_search_regex(
+            r'(?s)<div[^>]+class=["\']videoInfoBy["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>',
+            webpage, 'uploader', fatal=False)
+        upload_date = unified_strdate(self._html_search_regex(
+            r'(?s)<div[^>]+class=["\']videoInfoTime["\'][^>]*>(.+?)</div>',
+            webpage, 'upload date', fatal=False))
+
+        age_limit = self._rta_search(webpage)
+
+        average_rating = int_or_none(self._search_regex(
+            r'<div[^>]+class=["\']videoInfoRating["\'][^>]*>\s*<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>',
+            webpage, 'average rating', fatal=False))
+
+        view_count = str_to_int(self._search_regex(
+            r'(?s)<div[^>]+class=["\']videoInfoViews["\'][^>]*>.*?([\d,.]+)\s*</div>',
+            webpage, 'view count', fatal=False))
+        comment_count = str_to_int(self._search_regex(
+            r'>All [Cc]omments? \(([\d,.]+)\)',
+            webpage, 'comment count', fatal=False))
+
+        def extract_tag_box(title):
+            tag_box = self._search_regex(
+                (r'<div[^>]+class=["\']tagBoxTitle["\'][^>]*>\s*%s\b.*?</div>\s*'
+                 '<div[^>]+class=["\']tagBoxContent["\']>(.+?)</div>') % re.escape(title),
+                webpage, '%s tag box' % title, default=None)
+            if not tag_box:
+                return []
+            return re.findall(r'<a[^>]+href=[^>]+>([^<]+)', tag_box)
+
+        categories = extract_tag_box('Category')
+        tags = extract_tag_box('Tags')
 
         return {
             'id': video_id,
-            'uploader': video_uploader,
-            'upload_date': upload_date,
-            'title': video_title,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
             'thumbnail': thumbnail,
-            'description': video_description,
+            'uploader': uploader,
+            'upload_date': upload_date,
+            'average_rating': average_rating,
+            'view_count': view_count,
+            'comment_count': comment_count,
+            'categories': categories,
+            'tags': tags,
             'age_limit': age_limit,
             'formats': formats,
         }

youtube_dl/extractor/youtube.py

@@ -703,7 +703,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
     def _extract_signature_function(self, video_id, player_url, example_sig):
         id_m = re.match(
-            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?)?\.(?P<ext>[a-z]+)$',
+            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
             player_url)
         if not id_m:
             raise ExtractorError('Cannot identify player %r' % player_url)
@@ -1107,6 +1107,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     if not video_info:
                         video_info = get_video_info
                     if 'token' in get_video_info:
+                        # Different get_video_info requests may report different results, e.g.
+                        # some may report video unavailability, but some may serve it without
+                        # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
+                        # the original webpage as well as el=info and el=embedded get_video_info
+                        # requests report video unavailability due to geo restriction while
+                        # el=detailpage succeeds and returns valid data). This is probably
+                        # due to YouTube measures against IP ranges of hosting providers.
+                        # Working around by preferring the first succeeded video_info containing
+                        # the token if no such video_info yet was found.
+                        if 'token' not in video_info:
+                            video_info = get_video_info
                         break
         if 'token' not in video_info:
             if 'reason' in video_info:
@@ -1332,7 +1343,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                                 player_desc = 'flash player %s' % player_version
                             else:
                                 player_version = self._search_regex(
-                                    r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
+                                    [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
                                     player_url,
                                     'html5 player', fatal=False)
                                 player_desc = 'html5 player %s' % player_version
@@ -1644,8 +1655,18 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
         'playlist_mincount': 91,
         'info_dict': {
-            'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+            'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
+            'title': 'Uploads from lex will',
         }
+    }, {
+        'note': 'Age restricted channel',
+        # from https://www.youtube.com/user/DeusExOfficial
+        'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
+        'playlist_mincount': 64,
+        'info_dict': {
+            'id': 'UUs0ifCMCm1icqRbqhUINa0w',
+            'title': 'Uploads from Deus Ex',
+        },
     }]
 
     def _real_extract(self, url):
@@ -1666,7 +1687,7 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
                 'channelId', channel_page, 'channel id', default=None)
             if not channel_playlist_id:
                 channel_playlist_id = self._search_regex(
-                    r'data-channel-external-id="([^"]+)"',
+                    r'data-(?:channel-external-|yt)id="([^"]+)"',
                     channel_page, 'channel id', default=None)
         if channel_playlist_id and channel_playlist_id.startswith('UC'):
             playlist_id = 'UU' + channel_playlist_id[2:]

youtube_dl/jsinterp.py

@@ -232,10 +232,10 @@ class JSInterpreter(object):
     def extract_function(self, funcname):
         func_m = re.search(
             r'''(?x)
-                (?:function\s+%s|[{;]%s\s*=\s*function)\s*
+                (?:function\s+%s|[{;]%s\s*=\s*function|var\s+%s\s*=\s*function)\s*
                 \((?P<args>[^)]*)\)\s*
                 \{(?P<code>[^}]+)\}''' % (
-                re.escape(funcname), re.escape(funcname)),
+                re.escape(funcname), re.escape(funcname), re.escape(funcname)),
             self.code)
         if func_m is None:
             raise ExtractorError('Could not find JS function %r' % funcname)

youtube_dl/postprocessor/ffmpeg.py

@@ -272,7 +272,7 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
             return [], information
 
         try:
-            self._downloader.to_screen('[' + self.basename + '] Destination: ' + new_path)
+            self._downloader.to_screen('[ffmpeg] Destination: ' + new_path)
             self.run_ffmpeg(path, new_path, acodec, more_opts)
         except AudioConversionError as e:
             raise PostProcessingError(

youtube_dl/utils.py

@@ -36,6 +36,7 @@ import zlib
 from .compat import (
     compat_basestring,
     compat_chr,
+    compat_etree_fromstring,
     compat_html_entities,
     compat_http_client,
     compat_kwargs,
@@ -178,10 +179,19 @@ def xpath_with_ns(path, ns_map):
 
 
 def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
-    if sys.version_info < (2, 7):  # Crazy 2.6
-        xpath = xpath.encode('ascii')
+    def _find_xpath(xpath):
+        if sys.version_info < (2, 7):  # Crazy 2.6
+            xpath = xpath.encode('ascii')
+        return node.find(xpath)
+
+    if isinstance(xpath, (str, compat_str)):
+        n = _find_xpath(xpath)
+    else:
+        for xp in xpath:
+            n = _find_xpath(xp)
+            if n is not None:
+                break
 
-    n = node.find(xpath)
     if n is None:
         if default is not NO_DEFAULT:
             return default
@@ -356,7 +366,7 @@ def sanitize_path(s):
     if drive_or_unc:
         norm_path.pop(0)
     sanitized_path = [
-        path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|\.$)', '#', path_part)
+        path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|[\s.]$)', '#', path_part)
         for path_part in norm_path]
     if drive_or_unc:
         sanitized_path.insert(0, drive_or_unc + os.path.sep)
@@ -814,9 +824,11 @@ def parse_iso8601(date_str, delimiter='T', timezone=None):
     if date_str is None:
         return None
 
+    date_str = re.sub(r'\.[0-9]+', '', date_str)
+
     if timezone is None:
         m = re.search(
-            r'(\.[0-9]+)?(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
+            r'(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
             date_str)
         if not m:
             timezone = datetime.timedelta()
@@ -829,9 +841,12 @@ def parse_iso8601(date_str, delimiter='T', timezone=None):
                 timezone = datetime.timedelta(
                     hours=sign * int(m.group('hours')),
                     minutes=sign * int(m.group('minutes')))
-    date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter)
-    dt = datetime.datetime.strptime(date_str, date_format) - timezone
-    return calendar.timegm(dt.timetuple())
+    try:
+        date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter)
+        dt = datetime.datetime.strptime(date_str, date_format) - timezone
+        return calendar.timegm(dt.timetuple())
+    except ValueError:
+        pass
 
 
 def unified_strdate(date_str, day_first=True):
@@ -896,7 +911,8 @@ def unified_strdate(date_str, day_first=True):
         timetuple = email.utils.parsedate_tz(date_str)
         if timetuple:
             upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
-    return upload_date
+    if upload_date is not None:
+        return compat_str(upload_date)
 
 
 def determine_ext(url, default_ext='unknown_video'):
@@ -1651,29 +1667,6 @@ def encode_dict(d, encoding='utf-8'):
     return dict((k.encode(encoding), v.encode(encoding)) for k, v in d.items())
 
 
-try:
-    etree_iter = xml.etree.ElementTree.Element.iter
-except AttributeError:  # Python <=2.6
-    etree_iter = lambda n: n.findall('.//*')
-
-
-def parse_xml(s):
-    class TreeBuilder(xml.etree.ElementTree.TreeBuilder):
-        def doctype(self, name, pubid, system):
-            pass  # Ignore doctypes
-
-    parser = xml.etree.ElementTree.XMLParser(target=TreeBuilder())
-    kwargs = {'parser': parser} if sys.version_info >= (2, 7) else {}
-    tree = xml.etree.ElementTree.XML(s.encode('utf-8'), **kwargs)
-    # Fix up XML parser in Python 2.x
-    if sys.version_info < (3, 0):
-        for n in etree_iter(tree):
-            if n.text is not None:
-                if not isinstance(n.text, compat_str):
-                    n.text = n.text.decode('utf-8')
-    return tree
-
-
 US_RATINGS = {
     'G': 0,
     'PG': 10,
@@ -1974,7 +1967,7 @@ def dfxp2srt(dfxp_data):
 
         return out
 
-    dfxp = xml.etree.ElementTree.fromstring(dfxp_data.encode('utf-8'))
+    dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8'))
     out = []
     paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall(_x('.//ttaf1:p')) or dfxp.findall('.//p')
 

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.10.23'
+__version__ = '2015.11.13'