release 2015.10.18

[Letv] Fix extraction

docs/supportedsites.md

@@ -81,6 +81,7 @@
  - **CBSSports**
  - **CeskaTelevize**
  - **channel9**: Channel 9
+ - **Chaturbate**
  - **Chilloutzone**
  - **chirbit**
  - **chirbit:profile**
@@ -158,6 +159,7 @@
  - **facebook**
  - **faz.net**
  - **fc2**
+ - **Fczenit**
  - **fernsehkritik.tv**
  - **Firstpost**
  - **FiveTV**
@@ -280,7 +282,6 @@
  - **Malemotion**
  - **MDR**
  - **media.ccc.de**
- - **MegaVideoz**
  - **metacafe**
  - **Metacritic**
  - **Mgoon**
@@ -587,7 +588,8 @@
  - **twitch:stream**
  - **twitch:video**
  - **twitch:vod**
- - **TwitterCard**
+ - **twitter**
+ - **twitter:card**
  - **Ubu**
  - **udemy**
  - **udemy:course**

test/test_InfoExtractor.py

@@ -35,10 +35,18 @@ class TestInfoExtractor(unittest.TestCase):
             <meta name="og:title" content='Foo'/>
             <meta content="Some video's description " name="og:description"/>
             <meta property='og:image' content='http://domain.com/pic.jpg?key1=val1&amp;key2=val2'/>
+            <meta content='application/x-shockwave-flash' property='og:video:type'>
+            <meta content='Foo' property=og:foobar>
+            <meta name="og:test1" content='foo > < bar'/>
+            <meta name="og:test2" content="foo >//< bar"/>
             '''
         self.assertEqual(ie._og_search_title(html), 'Foo')
         self.assertEqual(ie._og_search_description(html), 'Some video\'s description ')
         self.assertEqual(ie._og_search_thumbnail(html), 'http://domain.com/pic.jpg?key1=val1&key2=val2')
+        self.assertEqual(ie._og_search_video_url(html, default=None), None)
+        self.assertEqual(ie._og_search_property('foobar', html), 'Foo')
+        self.assertEqual(ie._og_search_property('test1', html), 'foo > < bar')
+        self.assertEqual(ie._og_search_property('test2', html), 'foo >//< bar')
 
     def test_html_search_meta(self):
         ie = self.ie

test/test_youtube_lists.py

@@ -57,5 +57,14 @@ class TestYoutubeLists(unittest.TestCase):
         entries = result['entries']
         self.assertEqual(len(entries), 100)
 
+    def test_youtube_flat_playlist_titles(self):
+        dl = FakeYDL()
+        dl.params['extract_flat'] = True
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
+        self.assertIsPlaylist(result)
+        for entry in result['entries']:
+            self.assertTrue(entry.get('title'))
+
 if __name__ == '__main__':
     unittest.main()

youtube_dl/YoutubeDL.py

@@ -37,6 +37,7 @@ from .compat import (
     compat_tokenize_tokenize,
     compat_urllib_error,
     compat_urllib_request,
+    compat_urllib_request_DataHandler,
 )
 from .utils import (
     ContentTooShortError,
@@ -1967,8 +1968,9 @@ class YoutubeDL(object):
         debuglevel = 1 if self.params.get('debug_printtraffic') else 0
         https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
         ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
+        data_handler = compat_urllib_request_DataHandler()
         opener = compat_urllib_request.build_opener(
-            proxy_handler, https_handler, cookie_processor, ydlh)
+            proxy_handler, https_handler, cookie_processor, ydlh, data_handler)
 
         # Delete the default user-agent header, which would otherwise apply in
         # cases where our custom HTTP handler doesn't come into play

youtube_dl/compat.py

@@ -1,7 +1,10 @@
 from __future__ import unicode_literals
 
+import binascii
 import collections
+import email
 import getpass
+import io
 import optparse
 import os
 import re
@@ -38,6 +41,11 @@ try:
 except ImportError:  # Python 2
     import urlparse as compat_urlparse
 
+try:
+    import urllib.response as compat_urllib_response
+except ImportError:  # Python 2
+    import urllib as compat_urllib_response
+
 try:
     import http.cookiejar as compat_cookiejar
 except ImportError:  # Python 2
@@ -155,6 +163,40 @@ except ImportError:  # Python 2
         string = string.replace('+', ' ')
         return compat_urllib_parse_unquote(string, encoding, errors)
 
+try:
+    from urllib.request import DataHandler as compat_urllib_request_DataHandler
+except ImportError:  # Python < 3.4
+    # Ported from CPython 98774:1733b3bd46db, Lib/urllib/request.py
+    class compat_urllib_request_DataHandler(compat_urllib_request.BaseHandler):
+        def data_open(self, req):
+            # data URLs as specified in RFC 2397.
+            #
+            # ignores POSTed data
+            #
+            # syntax:
+            # dataurl   := "data:" [ mediatype ] [ ";base64" ] "," data
+            # mediatype := [ type "/" subtype ] *( ";" parameter )
+            # data      := *urlchar
+            # parameter := attribute "=" value
+            url = req.get_full_url()
+
+            scheme, data = url.split(":", 1)
+            mediatype, data = data.split(",", 1)
+
+            # even base64 encoded data URLs might be quoted so unquote in any case:
+            data = compat_urllib_parse_unquote_to_bytes(data)
+            if mediatype.endswith(";base64"):
+                data = binascii.a2b_base64(data)
+                mediatype = mediatype[:-7]
+
+            if not mediatype:
+                mediatype = "text/plain;charset=US-ASCII"
+
+            headers = email.message_from_string(
+                "Content-type: %s\nContent-length: %d\n" % (mediatype, len(data)))
+
+            return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url)
+
 try:
     compat_basestring = basestring  # Python 2
 except NameError:
@@ -489,6 +531,8 @@ __all__ = [
     'compat_urllib_parse_unquote_to_bytes',
     'compat_urllib_parse_urlparse',
     'compat_urllib_request',
+    'compat_urllib_request_DataHandler',
+    'compat_urllib_response',
     'compat_urlparse',
     'compat_urlretrieve',
     'compat_xml_parse_error',

youtube_dl/downloader/common.py

@@ -325,7 +325,7 @@ class FileDownloader(object):
         )
 
         # Check file already present
-        if filename != '-' and nooverwrites_and_exists or continuedl_and_exists:
+        if filename != '-' and (nooverwrites_and_exists or continuedl_and_exists):
             self.report_file_already_downloaded(filename)
             self._hook_progress({
                 'filename': filename,

youtube_dl/downloader/hls.py

@@ -30,7 +30,7 @@ class HlsFD(FileDownloader):
 
         args = [ffpp.executable, '-y']
 
-        if info_dict['http_headers']:
+        if info_dict['http_headers'] and re.match(r'^https?://', url):
             # Trailing \r\n after each HTTP header is important to prevent warning from ffmpeg/avconv:
             # [http @ 00000000003d2fa0] No trailing CRLF found in HTTP header.
             args += [

youtube_dl/downloader/rtmp.py

@@ -105,7 +105,7 @@ class RtmpFD(FileDownloader):
         protocol = info_dict.get('rtmp_protocol', None)
         real_time = info_dict.get('rtmp_real_time', False)
         no_resume = info_dict.get('no_resume', False)
-        continue_dl = info_dict.get('continuedl', True)
+        continue_dl = self.params.get('continuedl', True)
 
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)

youtube_dl/extractor/__init__.py

@@ -76,6 +76,7 @@ from .cbssports import CBSSportsIE
 from .ccc import CCCIE
 from .ceskatelevize import CeskaTelevizeIE
 from .channel9 import Channel9IE
+from .chaturbate import ChaturbateIE
 from .chilloutzone import ChilloutzoneIE
 from .chirbit import (
     ChirbitIE,
@@ -166,6 +167,7 @@ from .extremetube import ExtremeTubeIE
 from .facebook import FacebookIE
 from .faz import FazIE
 from .fc2 import FC2IE
+from .fczenit import FczenitIE
 from .firstpost import FirstpostIE
 from .firsttv import FirstTVIE
 from .fivemin import FiveMinIE
@@ -317,7 +319,6 @@ from .macgamestore import MacGameStoreIE
 from .mailru import MailRuIE
 from .malemotion import MalemotionIE
 from .mdr import MDRIE
-from .megavideoz import MegaVideozIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mgoon import MgoonIE
@@ -689,7 +690,7 @@ from .twitch import (
     TwitchBookmarksIE,
     TwitchStreamIE,
 )
-from .twitter import TwitterCardIE
+from .twitter import TwitterCardIE, TwitterIE
 from .ubu import UbuIE
 from .udemy import (
     UdemyIE,

youtube_dl/extractor/adultswim.py

@@ -41,7 +41,8 @@ class AdultSwimIE(InfoExtractor):
             'id': 'rQxZvXQ4ROaSOqq-or2Mow',
             'title': 'Rick and Morty - Pilot',
             'description': "Rick moves in with his daughter's family and establishes himself as a bad influence on his grandson, Morty. "
-        }
+        },
+        'skip': 'This video is only available for registered users',
     }, {
         'url': 'http://www.adultswim.com/videos/playlists/american-parenting/putting-francine-out-of-business/',
         'playlist': [
@@ -134,7 +135,13 @@ class AdultSwimIE(InfoExtractor):
             show = bootstrapped_data['show']
             show_title = show['title']
             stream = video_info.get('stream')
-            clips = [stream] if stream else video_info['clips']
+            clips = [stream] if stream else video_info.get('clips')
+            if not clips:
+                raise ExtractorError(
+                    'This video is only available via cable service provider subscription that'
+                    ' is not currently supported. You may want to use --cookies.'
+                    if video_info.get('auth') is True else 'Unable to find stream or clips',
+                    expected=True)
             segment_ids = [clip['videoPlaybackID'] for clip in clips]
 
         episode_id = video_info['id']

youtube_dl/extractor/bandcamp.py

@@ -10,6 +10,8 @@ from ..compat import (
 )
 from ..utils import (
     ExtractorError,
+    float_or_none,
+    int_or_none,
 )
 
 
@@ -52,11 +54,11 @@ class BandcampIE(InfoExtractor):
                     ext, abr_str = format_id.split('-', 1)
                     formats.append({
                         'format_id': format_id,
-                        'url': format_url,
+                        'url': self._proto_relative_url(format_url, 'http:'),
                         'ext': ext,
                         'vcodec': 'none',
                         'acodec': ext,
-                        'abr': int(abr_str),
+                        'abr': int_or_none(abr_str),
                     })
 
                 self._sort_formats(formats)
@@ -65,7 +67,7 @@ class BandcampIE(InfoExtractor):
                     'id': compat_str(data['id']),
                     'title': data['title'],
                     'formats': formats,
-                    'duration': float(data['duration']),
+                    'duration': float_or_none(data.get('duration')),
                 }
             else:
                 raise ExtractorError('No free songs found')

youtube_dl/extractor/bbc.py

@@ -11,6 +11,8 @@ from ..utils import (
     int_or_none,
     parse_duration,
     parse_iso8601,
+    remove_end,
+    unescapeHTML,
 )
 from ..compat import compat_HTTPError
 
@@ -28,6 +30,14 @@ class BBCCoUkIE(InfoExtractor):
         'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
     ]
 
+    _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
+    _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
+
+    _NAMESPACES = (
+        _MEDIASELECTION_NS,
+        _EMP_PLAYLIST_NS,
+    )
+
     _TESTS = [
         {
             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
@@ -193,6 +203,7 @@ class BBCCoUkIE(InfoExtractor):
 
     def _extract_connection(self, connection, programme_id):
         formats = []
+        kind = connection.get('kind')
         protocol = connection.get('protocol')
         supplier = connection.get('supplier')
         if protocol == 'http':
@@ -218,7 +229,7 @@ class BBCCoUkIE(InfoExtractor):
             else:
                 formats.append({
                     'url': href,
-                    'format_id': supplier,
+                    'format_id': supplier or kind or protocol,
                 })
         elif protocol == 'rtmp':
             application = connection.get('application', 'ondemand')
@@ -238,16 +249,24 @@ class BBCCoUkIE(InfoExtractor):
         return formats
 
     def _extract_items(self, playlist):
-        return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
+        return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
+
+    def _findall_ns(self, element, xpath):
+        elements = []
+        for ns in self._NAMESPACES:
+            elements.extend(element.findall(xpath % ns))
+        return elements
 
     def _extract_medias(self, media_selection):
-        error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
+        error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
+        if error is None:
+            media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
         if error is not None:
             raise BBCCoUkIE.MediaSelectionError(error.get('id'))
-        return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
+        return self._findall_ns(media_selection, './{%s}media')
 
     def _extract_connections(self, media):
-        return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
+        return self._findall_ns(media, './{%s}connection')
 
     def _extract_video(self, media, programme_id):
         formats = []
@@ -261,13 +280,14 @@ class BBCCoUkIE(InfoExtractor):
             conn_formats = self._extract_connection(connection, programme_id)
             for format in conn_formats:
                 format.update({
-                    'format_id': '%s_%s' % (service, format['format_id']),
                     'width': width,
                     'height': height,
                     'vbr': vbr,
                     'vcodec': vcodec,
                     'filesize': file_size,
                 })
+                if service:
+                    format['format_id'] = '%s_%s' % (service, format['format_id'])
             formats.extend(conn_formats)
         return formats
 
@@ -382,7 +402,7 @@ class BBCCoUkIE(InfoExtractor):
             url, playlist_id, 'Downloading legacy playlist XML')
 
     def _extract_from_legacy_playlist(self, playlist, playlist_id):
-        no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
+        no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
         if no_items is not None:
             reason = no_items.get('reason')
             if reason == 'preAvailability':
@@ -399,8 +419,9 @@ class BBCCoUkIE(InfoExtractor):
             kind = item.get('kind')
             if kind != 'programme' and kind != 'radioProgramme':
                 continue
-            title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
-            description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
+            title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
+            description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
+            description = description_el.text if description_el is not None else None
 
             def get_programme_id(item):
                 def get_from_attributes(item):
@@ -409,16 +430,18 @@ class BBCCoUkIE(InfoExtractor):
                         if value and re.match(r'^[pb][\da-z]{7}$', value):
                             return value
                 get_from_attributes(item)
-                mediator = item.find('./{http://bbc.co.uk/2008/emp/playlist}mediator')
+                mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
                 if mediator is not None:
                     return get_from_attributes(mediator)
 
             programme_id = get_programme_id(item)
             duration = int_or_none(item.get('duration'))
-            # TODO: programme_id can be None and media items can be incorporated right inside
-            # playlist's item (e.g. http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
-            # as f4m and m3u8
-            formats, subtitles = self._download_media_selector(programme_id)
+
+            if programme_id:
+                formats, subtitles = self._download_media_selector(programme_id)
+            else:
+                formats, subtitles = self._process_media_selector(item, playlist_id)
+                programme_id = playlist_id
 
         return programme_id, title, description, duration, formats, subtitles
 
@@ -470,6 +493,9 @@ class BBCIE(BBCCoUkIE):
     _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
 
     _MEDIASELECTOR_URLS = [
+        # Provides HQ HLS streams but fails with geolocation in some cases when it's
+        # even not geo restricted at all
+        'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
         # Provides more formats, namely direct mp4 links, but fails on some videos with
         # notukerror for non UK (?) users (e.g.
         # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
@@ -479,8 +505,7 @@ class BBCIE(BBCCoUkIE):
     ]
 
     _TESTS = [{
-        # article with multiple videos embedded with data-media-meta containing
-        # playlist.sxml, externalId and no direct video links
+        # article with multiple videos embedded with data-playable containing vpids
         'url': 'http://www.bbc.com/news/world-europe-32668511',
         'info_dict': {
             'id': 'world-europe-32668511',
@@ -489,7 +514,7 @@ class BBCIE(BBCCoUkIE):
         },
         'playlist_count': 2,
     }, {
-        # article with multiple videos embedded with data-media-meta (more videos)
+        # article with multiple videos embedded with data-playable (more videos)
         'url': 'http://www.bbc.com/news/business-28299555',
         'info_dict': {
             'id': 'business-28299555',
@@ -500,6 +525,7 @@ class BBCIE(BBCCoUkIE):
         'skip': 'Save time',
     }, {
         # article with multiple videos embedded with `new SMP()`
+        # broken
         'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
         'info_dict': {
             'id': '3662a707-0af9-3149-963f-47bea720b460',
@@ -507,12 +533,13 @@ class BBCIE(BBCCoUkIE):
         },
         'playlist_count': 18,
     }, {
-        # single video embedded with mediaAssetPage.init()
+        # single video embedded with data-playable containing vpid
         'url': 'http://www.bbc.com/news/world-europe-32041533',
         'info_dict': {
             'id': 'p02mprgb',
             'ext': 'mp4',
             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
+            'description': 'md5:2868290467291b37feda7863f7a83f54',
             'duration': 47,
             'timestamp': 1427219242,
             'upload_date': '20150324',
@@ -522,15 +549,14 @@ class BBCIE(BBCCoUkIE):
             'skip_download': True,
         }
     }, {
-        # article with single video embedded with data-media-meta containing
-        # direct video links (for now these are extracted) and playlist.xml (with
-        # media items as f4m and m3u8 - currently unsupported)
+        # article with single video embedded with data-playable containing XML playlist
+        # with direct video links as progressiveDownloadUrl (for now these are extracted)
+        # and playlist with f4m and m3u8 as streamingUrl
         'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
         'info_dict': {
             'id': '150615_telabyad_kentin_cogu',
             'ext': 'mp4',
             'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
-            'duration': 47,
             'timestamp': 1434397334,
             'upload_date': '20150615',
         },
@@ -538,13 +564,12 @@ class BBCIE(BBCCoUkIE):
             'skip_download': True,
         }
     }, {
-        # single video embedded with mediaAssetPage.init() (regional section)
+        # single video embedded with data-playable containing XML playlists (regional section)
         'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
         'info_dict': {
             'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
             'ext': 'mp4',
             'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
-            'duration': 87,
             'timestamp': 1434713142,
             'upload_date': '20150619',
         },
@@ -586,27 +611,34 @@ class BBCIE(BBCCoUkIE):
             'ext': 'mp4',
             'title': 'Hyundai Santa Fe Sport: Rock star',
             'description': 'md5:b042a26142c4154a6e472933cf20793d',
-            'timestamp': 1368473503,
-            'upload_date': '20130513',
+            'timestamp': 1415867444,
+            'upload_date': '20141113',
         },
         'params': {
             # rtmp download
             'skip_download': True,
         }
     }, {
-        # single video with playlist.sxml URL
+        # single video with playlist.sxml URL in playlist param
         'url': 'http://www.bbc.com/sport/0/football/33653409',
         'info_dict': {
             'id': 'p02xycnp',
             'ext': 'mp4',
             'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
-            'description': 'md5:398fca0e2e701c609d726e034fa1fc89',
             'duration': 140,
         },
         'params': {
             # rtmp download
             'skip_download': True,
         }
+    }, {
+        # article with multiple videos embedded with playlist.sxml in playlist param
+        'url': 'http://www.bbc.com/sport/0/football/34475836',
+        'info_dict': {
+            'id': '34475836',
+            'title': 'What Liverpool can expect from Klopp',
+        },
+        'playlist_count': 3,
     }, {
         # single video with playlist URL from weather section
         'url': 'http://www.bbc.com/weather/features/33601775',
@@ -648,40 +680,107 @@ class BBCIE(BBCCoUkIE):
 
         return [], []
 
+    def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
+        programme_id, title, description, duration, formats, subtitles = \
+            self._process_legacy_playlist_url(url, playlist_id)
+        self._sort_formats(formats)
+        return {
+            'id': programme_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'timestamp': timestamp,
+            'formats': formats,
+            'subtitles': subtitles,
+        }
+
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
 
         webpage = self._download_webpage(url, playlist_id)
 
-        timestamp = parse_iso8601(self._search_regex(
-            [r'"datePublished":\s*"([^"]+)',
-             r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
-             r'itemprop="datePublished"[^>]+datetime="([^"]+)"'],
-            webpage, 'date', default=None))
+        timestamp = None
+        playlist_title = None
+        playlist_description = None
 
-        # single video with playlist.sxml URL (e.g. http://www.bbc.com/sport/0/football/3365340ng)
-        playlist = self._search_regex(
-            r'<param[^>]+name="playlist"[^>]+value="([^"]+)"',
-            webpage, 'playlist', default=None)
-        if playlist:
-            programme_id, title, description, duration, formats, subtitles = \
-                self._process_legacy_playlist_url(playlist, playlist_id)
-            self._sort_formats(formats)
-            return {
-                'id': programme_id,
-                'title': title,
-                'description': description,
-                'duration': duration,
-                'timestamp': timestamp,
-                'formats': formats,
-                'subtitles': subtitles,
-            }
+        ld = self._parse_json(
+            self._search_regex(
+                r'(?s)<script type="application/ld\+json">(.+?)</script>',
+                webpage, 'ld json', default='{}'),
+            playlist_id, fatal=False)
+        if ld:
+            timestamp = parse_iso8601(ld.get('datePublished'))
+            playlist_title = ld.get('headline')
+            playlist_description = ld.get('articleBody')
+
+        if not timestamp:
+            timestamp = parse_iso8601(self._search_regex(
+                [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
+                 r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
+                 r'"datePublished":\s*"([^"]+)'],
+                webpage, 'date', default=None))
+
+        entries = []
+
+        # article with multiple videos embedded with playlist.sxml (e.g.
+        # http://www.bbc.com/sport/0/football/34475836)
+        playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
+        if playlists:
+            entries = [
+                self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
+                for playlist_url in playlists]
+
+        # news article with multiple videos embedded with data-playable
+        data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
+        if data_playables:
+            for _, data_playable_json in data_playables:
+                data_playable = self._parse_json(
+                    unescapeHTML(data_playable_json), playlist_id, fatal=False)
+                if not data_playable:
+                    continue
+                settings = data_playable.get('settings', {})
+                if settings:
+                    # data-playable with video vpid in settings.playlistObject.items (e.g.
+                    # http://www.bbc.com/news/world-us-canada-34473351)
+                    playlist_object = settings.get('playlistObject', {})
+                    if playlist_object:
+                        items = playlist_object.get('items')
+                        if items and isinstance(items, list):
+                            title = playlist_object['title']
+                            description = playlist_object.get('summary')
+                            duration = int_or_none(items[0].get('duration'))
+                            programme_id = items[0].get('vpid')
+                            formats, subtitles = self._download_media_selector(programme_id)
+                            self._sort_formats(formats)
+                            entries.append({
+                                'id': programme_id,
+                                'title': title,
+                                'description': description,
+                                'timestamp': timestamp,
+                                'duration': duration,
+                                'formats': formats,
+                                'subtitles': subtitles,
+                            })
+                    else:
+                        # data-playable without vpid but with a playlist.sxml URLs
+                        # in otherSettings.playlist (e.g.
+                        # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
+                        playlist = data_playable.get('otherSettings', {}).get('playlist', {})
+                        if playlist:
+                            entries.append(self._extract_from_playlist_sxml(
+                                playlist.get('progressiveDownloadUrl'), playlist_id, timestamp))
+
+        if entries:
+            playlist_title = playlist_title or remove_end(self._og_search_title(webpage), ' - BBC News')
+            playlist_description = playlist_description or self._og_search_description(webpage, default=None)
+            return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 
         # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
         programme_id = self._search_regex(
             [r'data-video-player-vpid="([\da-z]{8})"',
              r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'],
             webpage, 'vpid', default=None)
+
         if programme_id:
             formats, subtitles = self._download_media_selector(programme_id)
             self._sort_formats(formats)

youtube_dl/extractor/beeg.py

@@ -1,65 +1,67 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+)
 
 
 class BeegIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)'
     _TEST = {
         'url': 'http://beeg.com/5416503',
-        'md5': '1bff67111adb785c51d1b42959ec10e5',
+        'md5': '46c384def73b33dbc581262e5ee67cef',
         'info_dict': {
             'id': '5416503',
             'ext': 'mp4',
             'title': 'Sultry Striptease',
-            'description': 'md5:6db3c6177972822aaba18652ff59c773',
-            'categories': list,  # NSFW
-            'thumbnail': 're:https?://.*\.jpg$',
+            'description': 'md5:d22219c09da287c14bed3d6c37ce4bc2',
+            'timestamp': 1391813355,
+            'upload_date': '20140207',
+            'duration': 383,
+            'tags': list,
             'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
-
-        quality_arr = self._search_regex(
-            r'(?s)var\s+qualityArr\s*=\s*{\s*(.+?)\s*}', webpage, 'quality formats')
-
-        formats = [{
-            'url': fmt[1],
-            'format_id': fmt[0],
-            'height': int(fmt[0][:-1]),
-        } for fmt in re.findall(r"'([^']+)'\s*:\s*'([^']+)'", quality_arr)]
+        video = self._download_json(
+            'http://beeg.com/api/v1/video/%s' % video_id, video_id)
 
+        formats = []
+        for format_id, video_url in video.items():
+            height = self._search_regex(
+                r'^(\d+)[pP]$', format_id, 'height', default=None)
+            if not height:
+                continue
+            formats.append({
+                'url': self._proto_relative_url(video_url.replace('{DATA_MARKERS}', ''), 'http:'),
+                'format_id': format_id,
+                'height': int(height),
+            })
         self._sort_formats(formats)
 
-        title = self._html_search_regex(
-            r'<title>([^<]+)\s*-\s*beeg\.?</title>', webpage, 'title')
+        title = video['title']
+        video_id = video.get('id') or video_id
+        display_id = video.get('code')
+        description = video.get('desc')
 
-        description = self._html_search_regex(
-            r'<meta name="description" content="([^"]*)"',
-            webpage, 'description', fatal=False)
-        thumbnail = self._html_search_regex(
-            r'\'previewer.url\'\s*:\s*"([^"]*)"',
-            webpage, 'thumbnail', fatal=False)
+        timestamp = parse_iso8601(video.get('date'), ' ')
+        duration = int_or_none(video.get('duration'))
 
-        categories_str = self._html_search_regex(
-            r'<meta name="keywords" content="([^"]+)"', webpage, 'categories', fatal=False)
-        categories = (
-            None if categories_str is None
-            else categories_str.split(','))
+        tags = [tag.strip() for tag in video['tags'].split(',')] if video.get('tags') else None
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'title': title,
             'description': description,
-            'thumbnail': thumbnail,
-            'categories': categories,
+            'timestamp': timestamp,
+            'duration': duration,
+            'tags': tags,
             'formats': formats,
             'age_limit': 18,
         }

youtube_dl/extractor/bild.py

@@ -4,7 +4,7 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
-    fix_xml_ampersands,
+    unescapeHTML,
 )
 
 
@@ -17,26 +17,24 @@ class BildIE(InfoExtractor):
         'info_dict': {
             'id': '38184146',
             'ext': 'mp4',
-            'title': 'BILD hat sie getestet',
+            'title': 'Das können die  neuen iPads',
+            'description': 'md5:a4058c4fa2a804ab59c00d7244bbf62f',
             'thumbnail': 're:^https?://.*\.jpg$',
             'duration': 196,
-            'description': 'Mit dem iPad Air 2 und dem iPad Mini 3 hat Apple zwei neue Tablet-Modelle präsentiert. BILD-Reporter Sven Stein durfte die Geräte bereits testen. ',
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        xml_url = url.split(".bild.html")[0] + ",view=xml.bild.xml"
-        doc = self._download_xml(xml_url, video_id, transform_source=fix_xml_ampersands)
-
-        duration = int_or_none(doc.attrib.get('duration'), scale=1000)
+        video_data = self._download_json(
+            url.split('.bild.html')[0] + ',view=json.bild.html', video_id)
 
         return {
             'id': video_id,
-            'title': doc.attrib['ueberschrift'],
-            'description': doc.attrib.get('text'),
-            'url': doc.attrib['src'],
-            'thumbnail': doc.attrib.get('img'),
-            'duration': duration,
+            'title': unescapeHTML(video_data['title']).strip(),
+            'description': unescapeHTML(video_data.get('description')),
+            'url': video_data['clipList'][0]['srces'][0]['src'],
+            'thumbnail': video_data.get('poster'),
+            'duration': int_or_none(video_data.get('durationSec')),
         }

youtube_dl/extractor/canalc2.py

@@ -4,38 +4,53 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import parse_duration
 
 
 class Canalc2IE(InfoExtractor):
     IE_NAME = 'canalc2.tv'
-    _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?.*?idVideo=(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?canalc2\.tv/video/(?P<id>\d+)'
 
     _TEST = {
-        'url': 'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui',
+        'url': 'http://www.canalc2.tv/video/12163',
         'md5': '060158428b650f896c542dfbb3d6487f',
         'info_dict': {
             'id': '12163',
-            'ext': 'mp4',
-            'title': 'Terrasses du Numérique'
+            'ext': 'flv',
+            'title': 'Terrasses du Numérique',
+            'duration': 122,
+        },
+        'params': {
+            'skip_download': True,  # Requires rtmpdump
         }
     }
 
     def _real_extract(self, url):
-        video_id = re.match(self._VALID_URL, url).group('id')
-        # We need to set the voir field for getting the file name
-        url = 'http://www.canalc2.tv/video.asp?idVideo=%s&voir=oui' % video_id
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
-        file_name = self._search_regex(
-            r"so\.addVariable\('file','(.*?)'\);",
-            webpage, 'file name')
-        video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file_name
+        video_url = self._search_regex(
+            r'jwplayer\((["\'])Player\1\)\.setup\({[^}]*file\s*:\s*(["\'])(?P<file>.+?)\2',
+            webpage, 'video_url', group='file')
+        formats = [{'url': video_url}]
+        if video_url.startswith('rtmp://'):
+            rtmp = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>.+/))(?P<play_path>mp4:.+)$', video_url)
+            formats[0].update({
+                'url': rtmp.group('url'),
+                'ext': 'flv',
+                'app': rtmp.group('app'),
+                'play_path': rtmp.group('play_path'),
+                'page_url': url,
+            })
 
         title = self._html_search_regex(
-            r'class="evenement8">(.*?)</a>', webpage, 'title')
+            r'(?s)class="[^"]*col_description[^"]*">.*?<h3>(.*?)</h3>', webpage, 'title')
+        duration = parse_duration(self._search_regex(
+            r'id=["\']video_duree["\'][^>]*>([^<]+)',
+            webpage, 'duration', fatal=False))
 
         return {
             'id': video_id,
-            'ext': 'mp4',
-            'url': video_url,
             'title': title,
+            'duration': duration,
+            'formats': formats,
         }

youtube_dl/extractor/channel9.py

@@ -3,7 +3,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    parse_filesize,
+    qualities,
+)
 
 
 class Channel9IE(InfoExtractor):
@@ -28,7 +32,7 @@ class Channel9IE(InfoExtractor):
                 'title': 'Developer Kick-Off Session: Stuff We Love',
                 'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
                 'duration': 4576,
-                'thumbnail': 'http://video.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
+                'thumbnail': 're:http://.*\.jpg',
                 'session_code': 'KOS002',
                 'session_day': 'Day 1',
                 'session_room': 'Arena 1A',
@@ -44,31 +48,29 @@ class Channel9IE(InfoExtractor):
                 'title': 'Self-service BI with Power BI - nuclear testing',
                 'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
                 'duration': 1540,
-                'thumbnail': 'http://video.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
+                'thumbnail': 're:http://.*\.jpg',
                 'authors': ['Mike Wilmot'],
             },
+        },
+        {
+            # low quality mp4 is best
+            'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+            'info_dict': {
+                'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+                'ext': 'mp4',
+                'title': 'Ranges for the Standard Library',
+                'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
+                'duration': 5646,
+                'thumbnail': 're:http://.*\.jpg',
+            },
+            'params': {
+                'skip_download': True,
+            },
         }
     ]
 
     _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
 
-    # Sorted by quality
-    _known_formats = ['MP3', 'MP4', 'Mid Quality WMV', 'Mid Quality MP4', 'High Quality WMV', 'High Quality MP4']
-
-    def _restore_bytes(self, formatted_size):
-        if not formatted_size:
-            return 0
-        m = re.match(r'^(?P<size>\d+(?:\.\d+)?)\s+(?P<units>[a-zA-Z]+)', formatted_size)
-        if not m:
-            return 0
-        units = m.group('units')
-        try:
-            exponent = ['B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB', 'ZB', 'YB'].index(units.upper())
-        except ValueError:
-            return 0
-        size = float(m.group('size'))
-        return int(size * (1024 ** exponent))
-
     def _formats_from_html(self, html):
         FORMAT_REGEX = r'''
             (?x)
@@ -78,16 +80,20 @@ class Channel9IE(InfoExtractor):
             <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s*
             </div>)?                                                # File size part may be missing
         '''
-        # Extract known formats
+        quality = qualities((
+            'MP3', 'MP4',
+            'Low Quality WMV', 'Low Quality MP4',
+            'Mid Quality WMV', 'Mid Quality MP4',
+            'High Quality WMV', 'High Quality MP4'))
         formats = [{
             'url': x.group('url'),
             'format_id': x.group('quality'),
             'format_note': x.group('note'),
             'format': '%s (%s)' % (x.group('quality'), x.group('note')),
-            'filesize': self._restore_bytes(x.group('filesize')),  # File size is approximate
-            'preference': self._known_formats.index(x.group('quality')),
+            'filesize_approx': parse_filesize(x.group('filesize')),
+            'quality': quality(x.group('quality')),
             'vcodec': 'none' if x.group('note') == 'Audio only' else None,
-        } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats]
+        } for x in list(re.finditer(FORMAT_REGEX, html))]
 
         self._sort_formats(formats)
 
@@ -158,7 +164,7 @@ class Channel9IE(InfoExtractor):
 
     def _extract_session_day(self, html):
         m = re.search(r'<li class="day">\s*<a href="/Events/[^"]+">(?P<day>[^<]+)</a>\s*</li>', html)
-        return m.group('day') if m is not None else None
+        return m.group('day').strip() if m is not None else None
 
     def _extract_session_room(self, html):
         m = re.search(r'<li class="room">\s*(?P<room>.+?)\s*</li>', html)
@@ -224,12 +230,12 @@ class Channel9IE(InfoExtractor):
         if contents is None:
             return contents
 
-        authors = self._extract_authors(html)
+        if len(contents) > 1:
+            raise ExtractorError('Got more than one entry')
+        result = contents[0]
+        result['authors'] = self._extract_authors(html)
 
-        for content in contents:
-            content['authors'] = authors
-
-        return contents
+        return result
 
     def _extract_session(self, html, content_path):
         contents = self._extract_content(html, content_path)

youtube_dl/extractor/chaturbate.py

@@ -0,0 +1,50 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class ChaturbateIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.com/(?P<id>[^/?#]+)'
+    _TESTS = [{
+        'url': 'https://www.chaturbate.com/siswet19/',
+        'info_dict': {
+            'id': 'siswet19',
+            'ext': 'mp4',
+            'title': 're:^siswet19 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'age_limit': 18,
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        'url': 'https://en.chaturbate.com/siswet19/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        m3u8_url = self._search_regex(
+            r'src=(["\'])(?P<url>http.+?\.m3u8.*?)\1', webpage,
+            'playlist', default=None, group='url')
+
+        if not m3u8_url:
+            error = self._search_regex(
+                r'<span[^>]+class=(["\'])desc_span\1[^>]*>(?P<error>[^<]+)</span>',
+                webpage, 'error', group='error')
+            raise ExtractorError(error, expected=True)
+
+        formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
+
+        return {
+            'id': video_id,
+            'title': self._live_title(video_id),
+            'thumbnail': 'https://cdn-s.highwebmedia.com/uHK3McUtGCG3SMFcd4ZJsRv8/roomimage/%s.jpg' % video_id,
+            'age_limit': self._rta_search(webpage),
+            'is_live': True,
+            'formats': formats,
+        }

youtube_dl/extractor/common.py

@@ -172,6 +172,7 @@ class InfoExtractor(object):
     view_count:     How many users have watched the video on the platform.
     like_count:     Number of positive ratings of the video
     dislike_count:  Number of negative ratings of the video
+    repost_count:   Number of reposts of the video
     average_rating: Average rating give by users, the scale used depends on the webpage
     comment_count:  Number of comments on the video
     comments:       A list of comments, each with one or more of the following
@@ -645,8 +646,9 @@ class InfoExtractor(object):
     # Helper functions for extracting OpenGraph info
     @staticmethod
     def _og_regexes(prop):
-        content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')'
-        property_re = r'(?:name|property)=[\'"]?og:%s[\'"]?' % re.escape(prop)
+        content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))'
+        property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)'
+                       % {'prop': re.escape(prop)})
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
             template % (property_re, content_re),

youtube_dl/extractor/criterion.py

@@ -27,9 +27,7 @@ class CriterionIE(InfoExtractor):
         final_url = self._search_regex(
             r'so.addVariable\("videoURL", "(.+?)"\)\;', webpage, 'video url')
         title = self._og_search_title(webpage)
-        description = self._html_search_regex(
-            r'<meta name="description" content="(.+?)" />',
-            webpage, 'video description')
+        description = self._html_search_meta('description', webpage)
         thumbnail = self._search_regex(
             r'so.addVariable\("thumbnailURL", "(.+?)"\)\;',
             webpage, 'thumbnail url')

youtube_dl/extractor/crunchyroll.py

@@ -32,6 +32,26 @@ from ..aes import (
 
 
 class CrunchyrollBaseIE(InfoExtractor):
+    _NETRC_MACHINE = 'crunchyroll'
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+        self.report_login()
+        login_url = 'https://www.crunchyroll.com/?a=formhandler'
+        data = urlencode_postdata({
+            'formname': 'RpcApiUser_Login',
+            'name': username,
+            'password': password,
+        })
+        login_request = compat_urllib_request.Request(login_url, data)
+        login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        self._download_webpage(login_request, None, False, 'Wrong login info')
+
+    def _real_initialize(self):
+        self._login()
+
     def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None):
         request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request)
                    else compat_urllib_request.Request(url_or_request))
@@ -46,10 +66,22 @@ class CrunchyrollBaseIE(InfoExtractor):
         return super(CrunchyrollBaseIE, self)._download_webpage(
             request, video_id, note, errnote, fatal, tries, timeout, encoding)
 
+    @staticmethod
+    def _add_skip_wall(url):
+        parsed_url = compat_urlparse.urlparse(url)
+        qs = compat_urlparse.parse_qs(parsed_url.query)
+        # Always force skip_wall to bypass maturity wall, namely 18+ confirmation message:
+        # > This content may be inappropriate for some people.
+        # > Are you sure you want to continue?
+        # since it's not disabled by default in crunchyroll account's settings.
+        # See https://github.com/rg3/youtube-dl/issues/7202.
+        qs['skip_wall'] = ['1']
+        return compat_urlparse.urlunparse(
+            parsed_url._replace(query=compat_urllib_parse.urlencode(qs, True)))
+
 
 class CrunchyrollIE(CrunchyrollBaseIE):
     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
-    _NETRC_MACHINE = 'crunchyroll'
     _TESTS = [{
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
         'info_dict': {
@@ -81,10 +113,13 @@ class CrunchyrollIE(CrunchyrollBaseIE):
             # rtmp
             'skip_download': True,
         },
-
     }, {
         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
         'only_matching': True,
+    }, {
+        # geo-restricted (US), 18+ maturity wall, non-premium available
+        'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617',
+        'only_matching': True,
     }]
 
     _FORMAT_IDS = {
@@ -94,24 +129,6 @@ class CrunchyrollIE(CrunchyrollBaseIE):
         '1080': ('80', '108'),
     }
 
-    def _login(self):
-        (username, password) = self._get_login_info()
-        if username is None:
-            return
-        self.report_login()
-        login_url = 'https://www.crunchyroll.com/?a=formhandler'
-        data = urlencode_postdata({
-            'formname': 'RpcApiUser_Login',
-            'name': username,
-            'password': password,
-        })
-        login_request = compat_urllib_request.Request(login_url, data)
-        login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        self._download_webpage(login_request, None, False, 'Wrong login info')
-
-    def _real_initialize(self):
-        self._login()
-
     def _decrypt_subtitles(self, data, iv, id):
         data = bytes_to_intlist(base64.b64decode(data.encode('utf-8')))
         iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8')))
@@ -254,7 +271,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         else:
             webpage_url = 'http://www.' + mobj.group('url')
 
-        webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
+        webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage')
         note_m = self._html_search_regex(
             r'<div class="showmedia-trailer-notice">(.+?)</div>',
             webpage, 'trailer-notice', default='')
@@ -352,7 +369,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
 class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
     IE_NAME = "crunchyroll:playlist"
-    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?$'
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?(?:\?|$)'
 
     _TESTS = [{
         'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
@@ -361,12 +378,25 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
             'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
         },
         'playlist_count': 13,
+    }, {
+        # geo-restricted (US), 18+ maturity wall, non-premium available
+        'url': 'http://www.crunchyroll.com/cosplay-complex-ova',
+        'info_dict': {
+            'id': 'cosplay-complex-ova',
+            'title': 'Cosplay Complex OVA'
+        },
+        'playlist_count': 3,
+        'skip': 'Georestricted',
+    }, {
+        # geo-restricted (US), 18+ maturity wall, non-premium will be available since 2015.11.14
+        'url': 'http://www.crunchyroll.com/ladies-versus-butlers?skip_wall=1',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         show_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, show_id)
+        webpage = self._download_webpage(self._add_skip_wall(url), show_id)
         title = self._html_search_regex(
             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
             webpage, 'title')

youtube_dl/extractor/dailymotion.py

@@ -96,6 +96,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                 'uploader': 'HotWaves1012',
                 'age_limit': 18,
             }
+        },
+        # geo-restricted, player v5
+        {
+            'url': 'http://www.dailymotion.com/video/xhza0o',
+            'only_matching': True,
         }
     ]
 
@@ -124,6 +129,9 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         if player_v5:
             player = self._parse_json(player_v5, video_id)
             metadata = player['metadata']
+
+            self._check_error(metadata)
+
             formats = []
             for quality, media_list in metadata['qualities'].items():
                 for media in media_list:
@@ -201,9 +209,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                 'video info', flags=re.MULTILINE),
             video_id)
 
-        if info.get('error') is not None:
-            msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
-            raise ExtractorError(msg, expected=True)
+        self._check_error(info)
 
         formats = []
         for (key, format_id) in self._FORMATS:
@@ -246,6 +252,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
             'duration': info['duration']
         }
 
+    def _check_error(self, info):
+        if info.get('error') is not None:
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
+
     def _get_subtitles(self, video_id, webpage):
         try:
             sub_list = self._download_webpage(

youtube_dl/extractor/eagleplatform.py

@@ -87,7 +87,7 @@ class EaglePlatformIE(InfoExtractor):
         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
         formats = self._extract_m3u8_formats(
             m3u8_url, video_id,
-            'mp4', entry_protocol='m3u8_native')
+            'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
 
         mp4_url = self._get_video_url(
             # Secure mp4 URL is constructed according to Player.prototype.mp4 from

youtube_dl/extractor/expotv.py

@@ -33,20 +33,27 @@ class ExpoTVIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         player_key = self._search_regex(
             r'<param name="playerKey" value="([^"]+)"', webpage, 'player key')
-        config_url = 'http://client.expotv.com/video/config/%s/%s' % (
-            video_id, player_key)
         config = self._download_json(
-            config_url, video_id,
-            note='Downloading video configuration')
+            'http://client.expotv.com/video/config/%s/%s' % (video_id, player_key),
+            video_id, 'Downloading video configuration')
 
-        formats = [{
-            'url': fcfg['file'],
-            'height': int_or_none(fcfg.get('height')),
-            'format_note': fcfg.get('label'),
-            'ext': self._search_regex(
-                r'filename=.*\.([a-z0-9_A-Z]+)&', fcfg['file'],
-                'file extension', default=None),
-        } for fcfg in config['sources']]
+        formats = []
+        for fcfg in config['sources']:
+            media_url = fcfg.get('file')
+            if not media_url:
+                continue
+            if fcfg.get('type') == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    media_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls'))
+            else:
+                formats.append({
+                    'url': media_url,
+                    'height': int_or_none(fcfg.get('height')),
+                    'format_id': fcfg.get('label'),
+                    'ext': self._search_regex(
+                        r'filename=.*\.([a-z0-9_A-Z]+)&', media_url,
+                        'file extension', default=None) or fcfg.get('type'),
+                })
         self._sort_formats(formats)
 
         title = self._og_search_title(webpage)

youtube_dl/extractor/fczenit.py

@@ -0,0 +1,41 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class FczenitIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?fc-zenit\.ru/video/gl(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://fc-zenit.ru/video/gl6785/',
+        'md5': '458bacc24549173fe5a5aa29174a5606',
+        'info_dict': {
+            'id': '6785',
+            'ext': 'mp4',
+            'title': '«Зенит-ТВ»: как Олег Шатов играл против «Урала»',
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        video_title = self._html_search_regex(r'<div class=\"photoalbum__title\">([^<]+)', webpage, 'title')
+
+        bitrates_raw = self._html_search_regex(r'bitrates:.*\n(.*)\]', webpage, 'video URL')
+        bitrates = re.findall(r'url:.?\'(.+?)\'.*?bitrate:.?([0-9]{3}?)', bitrates_raw)
+
+        formats = [{
+            'url': furl,
+            'tbr': tbr,
+        } for furl, tbr in bitrates]
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'formats': formats,
+        }

youtube_dl/extractor/fivemin.py

@@ -2,11 +2,15 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_str,
     compat_urllib_parse,
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
+    parse_duration,
+    replace_extension,
 )
 
 
@@ -28,6 +32,7 @@ class FiveMinIE(InfoExtractor):
                 'id': '518013791',
                 'ext': 'mp4',
                 'title': 'iPad Mini with Retina Display Review',
+                'duration': 177,
             },
         },
         {
@@ -38,9 +43,52 @@ class FiveMinIE(InfoExtractor):
                 'id': '518086247',
                 'ext': 'mp4',
                 'title': 'How to Make a Next-Level Fruit Salad',
+                'duration': 184,
             },
         },
     ]
+    _ERRORS = {
+        'ErrorVideoNotExist': 'We\'re sorry, but the video you are trying to watch does not exist.',
+        'ErrorVideoNoLongerAvailable': 'We\'re sorry, but the video you are trying to watch is no longer available.',
+        'ErrorVideoRejected': 'We\'re sorry, but the video you are trying to watch has been removed.',
+        'ErrorVideoUserNotGeo': 'We\'re sorry, but the video you are trying to watch cannot be viewed from your current location.',
+        'ErrorVideoLibraryRestriction': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.',
+        'ErrorExposurePermission': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.',
+    }
+    _QUALITIES = {
+        1: {
+            'width': 640,
+            'height': 360,
+        },
+        2: {
+            'width': 854,
+            'height': 480,
+        },
+        4: {
+            'width': 1280,
+            'height': 720,
+        },
+        8: {
+            'width': 1920,
+            'height': 1080,
+        },
+        16: {
+            'width': 640,
+            'height': 360,
+        },
+        32: {
+            'width': 854,
+            'height': 480,
+        },
+        64: {
+            'width': 1280,
+            'height': 720,
+        },
+        128: {
+            'width': 640,
+            'height': 360,
+        },
+    }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -59,26 +107,36 @@ class FiveMinIE(InfoExtractor):
             'https://syn.5min.com/handlers/SenseHandler.ashx?' + query,
             video_id)
         if not response['success']:
-            err_msg = response['errorMessage']
-            if err_msg == 'ErrorVideoUserNotGeo':
-                msg = 'Video not available from your location'
-            else:
-                msg = 'Aol said: %s' % err_msg
-            raise ExtractorError(msg, expected=True, video_id=video_id)
+            raise ExtractorError(
+                '%s said: %s' % (
+                    self.IE_NAME,
+                    self._ERRORS.get(response['errorMessage'], response['errorMessage'])),
+                expected=True)
         info = response['binding'][0]
 
-        second_id = compat_str(int(video_id[:-2]) + 1)
         formats = []
-        for quality, height in [(1, 320), (2, 480), (4, 720), (8, 1080)]:
-            if any(r['ID'] == quality for r in info['Renditions']):
+        parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs(
+            compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0])
+        for rendition in info['Renditions']:
+            if rendition['RenditionType'] == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(rendition['Url'], video_id, m3u8_id='hls'))
+            elif rendition['RenditionType'] == 'aac':
+                continue
+            else:
+                rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType'])))
+                quality = self._QUALITIES.get(rendition['ID'], {})
                 formats.append({
-                    'format_id': compat_str(quality),
-                    'url': 'http://avideos.5min.com/%s/%s/%s_%s.mp4' % (second_id[-3:], second_id, video_id, quality),
-                    'height': height,
+                    'format_id': '%s-%d' % (rendition['RenditionType'], rendition['ID']),
+                    'url': rendition_url,
+                    'width': quality.get('width'),
+                    'height': quality.get('height'),
                 })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': info['Title'],
+            'thumbnail': info.get('ThumbURL'),
+            'duration': parse_duration(info.get('Duration')),
             'formats': formats,
         }

youtube_dl/extractor/imdb.py

@@ -4,8 +4,8 @@ import re
 import json
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urlparse,
+from ..utils import (
+    qualities,
 )
 
 
@@ -30,24 +30,33 @@ class ImdbIE(InfoExtractor):
         descr = self._html_search_regex(
             r'(?s)<span itemprop="description">(.*?)</span>',
             webpage, 'description', fatal=False)
-        available_formats = re.findall(
-            r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage,
-            flags=re.MULTILINE)
+        player_url = 'http://www.imdb.com/video/imdb/vi%s/imdb/single' % video_id
+        player_page = self._download_webpage(
+            player_url, video_id, 'Downloading player page')
+        # the player page contains the info for the default format, we have to
+        # fetch other pages for the rest of the formats
+        extra_formats = re.findall(r'href="(?P<url>%s.*?)".*?>(?P<name>.*?)<' % re.escape(player_url), player_page)
+        format_pages = [
+            self._download_webpage(
+                f_url, video_id, 'Downloading info for %s format' % f_name)
+            for f_url, f_name in extra_formats]
+        format_pages.append(player_page)
+
+        quality = qualities(['SD', '480p', '720p'])
         formats = []
-        for f_id, f_path in available_formats:
-            f_path = f_path.strip()
-            format_page = self._download_webpage(
-                compat_urlparse.urljoin(url, f_path),
-                'Downloading info for %s format' % f_id)
+        for format_page in format_pages:
             json_data = self._search_regex(
                 r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>',
                 format_page, 'json data', flags=re.DOTALL)
             info = json.loads(json_data)
             format_info = info['videoPlayerObject']['video']
+            f_id = format_info['ffname']
             formats.append({
                 'format_id': f_id,
                 'url': format_info['videoInfoList'][0]['videoUrl'],
+                'quality': quality(f_id),
             })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,

youtube_dl/extractor/iqiyi.py

@@ -191,7 +191,7 @@ class IqiyiIE(InfoExtractor):
             'vid': video_id,
             'vinfo': 1,
             'tm': tm,
-            'enc': self.md5_text((enc_key + tail)[1:64:2] + tail),
+            'enc': self.md5_text(enc_key + tail),
             'qyid': _uuid,
             'tn': random.random(),
             'um': 0,
@@ -205,7 +205,9 @@ class IqiyiIE(InfoExtractor):
 
     def get_enc_key(self, swf_url, video_id):
         # TODO: automatic key extraction
-        enc_key = 'eac64f22daf001da6ba9aa8da4d501508bbe90a4d4091fea3b0582a85b38c2cc'  # last update at 2015-09-23-23 for Zombie::bite
+        # last update at 2015-10-10 for Zombie::bite
+        # '7239670519b6ac209a0bee4ef0446a6b24894b8ac2751506e42116212a0d0272e505'[2:66][1::2]
+        enc_key = '97596c0abee04ab49ba25564161ad225'
         return enc_key
 
     def _real_extract(self, url):

youtube_dl/extractor/jeuxvideo.py

@@ -28,7 +28,7 @@ class JeuxVideoIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         title = mobj.group(1)
         webpage = self._download_webpage(url, title)
-        title = self._html_search_meta('name', webpage)
+        title = self._html_search_meta('name', webpage) or self._og_search_title(webpage)
         config_url = self._html_search_regex(
             r'data-src="(/contenu/medias/video.php.*?)"',
             webpage, 'config URL')

youtube_dl/extractor/letv.py

@@ -9,13 +9,14 @@ from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
-    compat_urlparse,
+    compat_ord,
 )
 from ..utils import (
     determine_ext,
     ExtractorError,
     parse_iso8601,
     int_or_none,
+    encode_data_uri,
 )
 
 
@@ -25,15 +26,16 @@ class LetvIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.letv.com/ptv/vplay/22005890.html',
-        'md5': 'cab23bd68d5a8db9be31c9a222c1e8df',
+        'md5': 'edadcfe5406976f42f9f266057ee5e40',
         'info_dict': {
             'id': '22005890',
             'ext': 'mp4',
             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
-            'timestamp': 1424747397,
-            'upload_date': '20150224',
             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
-        }
+        },
+        'params': {
+            'hls_prefer_native': True,
+        },
     }, {
         'url': 'http://www.letv.com/ptv/vplay/1415246.html',
         'info_dict': {
@@ -42,16 +44,22 @@ class LetvIE(InfoExtractor):
             'title': '美人天下01',
             'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
         },
+        'params': {
+            'hls_prefer_native': True,
+        },
     }, {
         'note': 'This video is available only in Mainland China, thus a proxy is needed',
         'url': 'http://www.letv.com/ptv/vplay/1118082.html',
-        'md5': 'f80936fbe20fb2f58648e81386ff7927',
+        'md5': '2424c74948a62e5f31988438979c5ad1',
         'info_dict': {
             'id': '1118082',
             'ext': 'mp4',
             'title': '与龙共舞 完整版',
             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
         },
+        'params': {
+            'hls_prefer_native': True,
+        },
         'skip': 'Only available in China',
     }]
 
@@ -74,6 +82,27 @@ class LetvIE(InfoExtractor):
         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
         return _loc3_
 
+    # see M3U8Encryption class in KLetvPlayer.swf
+    @staticmethod
+    def decrypt_m3u8(encrypted_data):
+        if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
+            return encrypted_data
+        encrypted_data = encrypted_data[5:]
+
+        _loc4_ = bytearray()
+        while encrypted_data:
+            b = compat_ord(encrypted_data[0])
+            _loc4_.extend([b // 16, b & 0x0f])
+            encrypted_data = encrypted_data[1:]
+        idx = len(_loc4_) - 11
+        _loc4_ = _loc4_[idx:] + _loc4_[:idx]
+        _loc7_ = bytearray()
+        while _loc4_:
+            _loc7_.append(_loc4_[0] * 16 + _loc4_[1])
+            _loc4_ = _loc4_[2:]
+
+        return bytes(_loc7_)
+
     def _real_extract(self, url):
         media_id = self._match_id(url)
         page = self._download_webpage(url, media_id)
@@ -115,23 +144,28 @@ class LetvIE(InfoExtractor):
         for format_id in formats:
             if format_id in dispatch:
                 media_url = playurl['domain'][0] + dispatch[format_id][0]
-
-                # Mimic what flvxz.com do
-                url_parts = list(compat_urlparse.urlparse(media_url))
-                qs = dict(compat_urlparse.parse_qs(url_parts[4]))
-                qs.update({
-                    'platid': '14',
-                    'splatid': '1401',
-                    'tss': 'no',
-                    'retry': 1
+                media_url += '&' + compat_urllib_parse.urlencode({
+                    'm3v': 1,
+                    'format': 1,
+                    'expect': 3,
+                    'rateid': format_id,
                 })
-                url_parts[4] = compat_urllib_parse.urlencode(qs)
-                media_url = compat_urlparse.urlunparse(url_parts)
+
+                nodes_data = self._download_json(
+                    media_url, media_id,
+                    'Download JSON metadata for format %s' % format_id)
+
+                req = self._request_webpage(
+                    nodes_data['nodelist'][0]['location'], media_id,
+                    note='Downloading m3u8 information for format %s' % format_id)
+
+                m3u8_data = self.decrypt_m3u8(req.read())
 
                 url_info_dict = {
-                    'url': media_url,
+                    'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
                     'ext': determine_ext(dispatch[format_id][1]),
                     'format_id': format_id,
+                    'protocol': 'm3u8',
                 }
 
                 if format_id[-1:] == 'p':

youtube_dl/extractor/lynda.py

@@ -140,13 +140,14 @@ class LyndaIE(LyndaBaseIE):
 
         prioritized_streams = video_json.get('PrioritizedStreams')
         if prioritized_streams:
-            formats.extend([
-                {
-                    'url': video_url,
-                    'width': int_or_none(format_id),
-                    'format_id': format_id,
-                } for format_id, video_url in prioritized_streams['0'].items()
-            ])
+            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()
+                ])
 
         self._check_formats(formats, video_id)
         self._sort_formats(formats)

youtube_dl/extractor/megavideoz.py

@@ -1,56 +0,0 @@
-# encoding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    xpath_text,
-)
-
-
-class MegaVideozIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?megavideoz\.eu/video/(?P<id>[^/]+)(?:/(?P<display_id>[^/]+))?'
-    _TEST = {
-        'url': 'http://megavideoz.eu/video/WM6UB919XMXH/SMPTE-Universal-Film-Leader',
-        'info_dict': {
-            'id': '48723',
-            'display_id': 'SMPTE-Universal-Film-Leader',
-            'ext': 'mp4',
-            'title': 'SMPTE Universal Film Leader',
-            'thumbnail': 're:https?://.*?\.jpg',
-            'duration': 10.93,
-        }
-    }
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id') or video_id
-
-        webpage = self._download_webpage(url, display_id)
-
-        if any(p in webpage for p in ('>Video Not Found<', '>404 Error<')):
-            raise ExtractorError('Video %s does not exist' % video_id, expected=True)
-
-        config = self._download_xml(
-            self._search_regex(
-                r"var\s+cnf\s*=\s*'([^']+)'", webpage, 'cnf url'),
-            display_id)
-
-        video_url = xpath_text(config, './file', 'video url', fatal=True)
-        title = xpath_text(config, './title', 'title', fatal=True)
-        thumbnail = xpath_text(config, './image', 'thumbnail')
-        duration = float_or_none(xpath_text(config, './duration', 'duration'))
-        video_id = xpath_text(config, './mediaid', 'video id') or video_id
-
-        return {
-            'id': video_id,
-            'display_id': display_id,
-            'url': video_url,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration
-        }

youtube_dl/extractor/rtbf.py

@@ -36,7 +36,7 @@ class RTBFIE(InfoExtractor):
 
         data = self._parse_json(
             unescapeHTML(self._search_regex(
-                r'data-video="([^"]+)"', webpage, 'data video')),
+                r'data-media="([^"]+)"', webpage, 'data video')),
             video_id)
 
         if data.get('provider').lower() == 'youtube':

youtube_dl/extractor/rte.py

@@ -9,16 +9,16 @@ from ..utils import (
 
 
 class RteIE(InfoExtractor):
-    _VALID_URL = r'http?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/(?P<id>[0-9]+)/'
+    _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P<id>[0-9]+)'
     _TEST = {
-        'url': 'http://www.rte.ie/player/de/show/10363114/',
+        'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/',
         'info_dict': {
-            'id': '10363114',
+            'id': '10478715',
             'ext': 'mp4',
-            'title': 'One News',
+            'title': 'Watch iWitness  online',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'description': 'The One O\'Clock News followed by Weather.',
-            'duration': 436.844,
+            'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.',
+            'duration': 60.046,
         },
         'params': {
             'skip_download': 'f4m fails with --test atm'

youtube_dl/extractor/twitch.py

@@ -15,6 +15,7 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
+    encode_dict,
     ExtractorError,
     int_or_none,
     parse_duration,
@@ -27,8 +28,7 @@ class TwitchBaseIE(InfoExtractor):
 
     _API_BASE = 'https://api.twitch.tv'
     _USHER_BASE = 'http://usher.twitch.tv'
-    _LOGIN_URL = 'https://secure.twitch.tv/login'
-    _LOGIN_POST_URL = 'https://passport.twitch.tv/authentications/new'
+    _LOGIN_URL = 'http://www.twitch.tv/login'
     _NETRC_MACHINE = 'twitch'
 
     def _handle_error(self, response):
@@ -61,26 +61,28 @@ class TwitchBaseIE(InfoExtractor):
         if username is None:
             return
 
-        login_page = self._download_webpage(
+        login_page, handle = self._download_webpage_handle(
             self._LOGIN_URL, None, 'Downloading login page')
 
         login_form = self._hidden_inputs(login_page)
 
         login_form.update({
-            'login': username.encode('utf-8'),
-            'password': password.encode('utf-8'),
+            'username': username,
+            'password': password,
         })
 
+        redirect_url = handle.geturl()
+
         post_url = self._search_regex(
             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
-            'post url', default=self._LOGIN_POST_URL, group='url')
+            'post url', default=redirect_url, group='url')
 
         if not post_url.startswith('http'):
-            post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
+            post_url = compat_urlparse.urljoin(redirect_url, post_url)
 
         request = compat_urllib_request.Request(
-            post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
-        request.add_header('Referer', self._LOGIN_URL)
+            post_url, compat_urllib_parse.urlencode(encode_dict(login_form)).encode('utf-8'))
+        request.add_header('Referer', redirect_url)
         response = self._download_webpage(
             request, None, 'Logging in as %s' % username)
 
@@ -238,14 +240,24 @@ class TwitchVodIE(TwitchItemBaseIE):
 
     def _real_extract(self, url):
         item_id = self._match_id(url)
+
         info = self._download_info(self._ITEM_SHORTCUT, item_id)
         access_token = self._download_json(
             '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id,
             'Downloading %s access token' % self._ITEM_TYPE)
+
         formats = self._extract_m3u8_formats(
-            '%s/vod/%s?nauth=%s&nauthsig=%s&allow_source=true'
-            % (self._USHER_BASE, item_id, access_token['token'], access_token['sig']),
+            '%s/vod/%s?%s' % (
+                self._USHER_BASE, item_id,
+                compat_urllib_parse.urlencode({
+                    'allow_source': 'true',
+                    'allow_spectre': 'true',
+                    'player': 'twitchweb',
+                    'nauth': access_token['token'],
+                    'nauthsig': access_token['sig'],
+                })),
             item_id, 'mp4')
+
         self._prefer_source(formats)
         info['formats'] = formats
 

youtube_dl/extractor/twitter.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -6,23 +7,51 @@ from .common import InfoExtractor
 from ..compat import compat_urllib_request
 from ..utils import (
     float_or_none,
-    unescapeHTML,
+    xpath_text,
+    remove_end,
 )
 
 
 class TwitterCardIE(InfoExtractor):
+    IE_NAME = 'twitter:card'
     _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'
-    _TEST = {
-        'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
-        'md5': 'a74f50b310c83170319ba16de6955192',
-        'info_dict': {
-            'id': '560070183650213889',
-            'ext': 'mp4',
-            'title': 'TwitterCard',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'duration': 30.033,
+    _TESTS = [
+        {
+            'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
+            'md5': '7d2f6b4d2eb841a7ccc893d479bfceb4',
+            'info_dict': {
+                'id': '560070183650213889',
+                'ext': 'mp4',
+                'title': 'TwitterCard',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'duration': 30.033,
+            }
         },
-    }
+        {
+            'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
+            'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
+            'info_dict': {
+                'id': '623160978427936768',
+                'ext': 'mp4',
+                'title': 'TwitterCard',
+                'thumbnail': 're:^https?://.*\.jpg',
+                'duration': 80.155,
+            },
+        },
+        {
+            'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
+            'md5': 'b6f35e8b08a0bec6c8af77a2f4b3a814',
+            'info_dict': {
+                'id': 'dq4Oj5quskI',
+                'ext': 'mp4',
+                'title': 'Ubuntu 11.10 Overview',
+                'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10-things-to-do-after-installing-ubuntu-11-10/',
+                'upload_date': '20111013',
+                'uploader': 'OMG! Ubuntu!',
+                'uploader_id': 'omgubuntu',
+            },
+        }
+    ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -40,10 +69,24 @@ class TwitterCardIE(InfoExtractor):
             request.add_header('User-Agent', user_agent)
             webpage = self._download_webpage(request, video_id)
 
-            config = self._parse_json(
-                unescapeHTML(self._search_regex(
-                    r'data-player-config="([^"]+)"', webpage, 'data player config')),
+            youtube_url = self._html_search_regex(
+                r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
+                webpage, 'youtube iframe', default=None)
+            if youtube_url:
+                return self.url_result(youtube_url, 'Youtube')
+
+            config = self._parse_json(self._html_search_regex(
+                r'data-player-config="([^"]+)"', webpage, 'data player config'),
                 video_id)
+            if 'playlist' not in config:
+                if 'vmapUrl' in config:
+                    vmap_data = self._download_xml(config['vmapUrl'], video_id)
+                    video_url = xpath_text(vmap_data, './/MediaFile').strip()
+                    formats.append({
+                        'url': video_url,
+                    })
+                    break   # same video regardless of UA
+                continue
 
             video_url = config['playlist'][0]['source']
 
@@ -70,3 +113,54 @@ class TwitterCardIE(InfoExtractor):
             'duration': duration,
             'formats': formats,
         }
+
+
+class TwitterIE(InfoExtractor):
+    IE_NAME = 'twitter'
+    _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 = {
+        'url': 'https://twitter.com/freethenipple/status/643211948184596480',
+        'md5': '31cd83a116fc41f99ae3d909d4caf6a0',
+        'info_dict': {
+            'id': '643211948184596480',
+            'ext': 'mp4',
+            'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 12.922,
+            'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
+            'uploader': 'FREE THE NIPPLE',
+            'uploader_id': 'freethenipple',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        user_id = mobj.group('user_id')
+        twid = mobj.group('id')
+
+        webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid)
+
+        username = remove_end(self._og_search_title(webpage), ' on Twitter')
+
+        title = self._og_search_description(webpage).strip('').replace('\n', ' ')
+
+        # strip  'https -_t.co_BJYgOjSeGA' junk from filenames
+        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',
+            '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,
+        }

youtube_dl/extractor/vidme.py

@@ -93,6 +93,10 @@ class VidmeIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        # nsfw, user-disabled
+        'url': 'https://vid.me/dzGJ',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -114,6 +118,12 @@ class VidmeIE(InfoExtractor):
 
         video = response['video']
 
+        if video.get('state') == 'user-disabled':
+            raise ExtractorError(
+                'Vidme said: This video has been suspended either due to a copyright claim, '
+                'or for violating the terms of use.',
+                expected=True)
+
         formats = [{
             'format_id': f.get('type'),
             'url': f['uri'],

youtube_dl/extractor/viewster.py

@@ -131,10 +131,11 @@ class ViewsterIE(InfoExtractor):
                 formats.extend(self._extract_f4m_formats(
                     video_url, video_id, f4m_id='hds'))
             elif ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
+                m3u8_formats = self._extract_m3u8_formats(
                     video_url, video_id, 'mp4', m3u8_id='hls',
-                    fatal=False  # m3u8 sometimes fail
-                ))
+                    fatal=False)  # m3u8 sometimes fail
+                if m3u8_formats:
+                    formats.extend(m3u8_formats)
             else:
                 format_id = media.get('Bitrate')
                 f = {

youtube_dl/extractor/vimeo.py

@@ -39,7 +39,7 @@ class VimeoBaseInfoExtractor(InfoExtractor):
             return
         self.report_login()
         webpage = self._download_webpage(self._LOGIN_URL, None, False)
-        token = self._extract_xsrft(webpage)
+        token, vuid = self._extract_xsrft_and_vuid(webpage)
         data = urlencode_postdata({
             'action': 'login',
             'email': username,
@@ -49,13 +49,18 @@ class VimeoBaseInfoExtractor(InfoExtractor):
         })
         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._download_webpage(login_request, None, False, 'Wrong login info')
 
-    def _extract_xsrft(self, webpage):
-        return self._search_regex(
+    def _extract_xsrft_and_vuid(self, webpage):
+        xsrft = self._search_regex(
             r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
             webpage, 'login token', group='xsrft')
+        vuid = self._search_regex(
+            r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
+            webpage, 'vuid', group='vuid')
+        return xsrft, vuid
 
 
 class VimeoIE(VimeoBaseInfoExtractor):
@@ -80,12 +85,12 @@ class VimeoIE(VimeoBaseInfoExtractor):
             'info_dict': {
                 'id': '56015672',
                 'ext': 'mp4',
-                "upload_date": "20121220",
-                "description": "This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
-                "uploader_id": "user7108434",
-                "uploader": "Filippo Valsorda",
-                "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
-                "duration": 10,
+                'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
+                'description': 'md5:2d3305bad981a06ff79f027f19865021',
+                'upload_date': '20121220',
+                'uploader_id': 'user7108434',
+                'uploader': 'Filippo Valsorda',
+                'duration': 10,
             },
         },
         {
@@ -98,7 +103,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
                 'uploader_id': 'openstreetmapus',
                 'uploader': 'OpenStreetMap US',
                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
-                'description': 'md5:380943ec71b89736ff4bf27183233d09',
+                'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
                 'duration': 1595,
             },
         },
@@ -152,7 +157,6 @@ class VimeoIE(VimeoBaseInfoExtractor):
         },
         {
             'url': 'http://vimeo.com/76979871',
-            'md5': '3363dd6ffebe3784d56f4132317fd446',
             'note': 'Video with subtitles',
             'info_dict': {
                 'id': '76979871',
@@ -198,7 +202,7 @@ 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', expected=True)
-        token = self._extract_xsrft(webpage)
+        token, vuid = self._extract_xsrft_and_vuid(webpage)
         data = urlencode_postdata({
             'password': password,
             'token': token,
@@ -208,6 +212,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
             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)
         return self._download_webpage(
             password_request, video_id,
@@ -281,7 +286,17 @@ class VimeoIE(VimeoBaseInfoExtractor):
         try:
             try:
                 config_url = self._html_search_regex(
-                    r' data-config-url="(.+?)"', webpage, 'config URL')
+                    r' data-config-url="(.+?)"', webpage,
+                    'config URL', default=None)
+                if not config_url:
+                    # Sometimes new react-based page is served instead of old one that require
+                    # different config URL extraction approach (see
+                    # https://github.com/rg3/youtube-dl/pull/7209)
+                    vimeo_clip_page_config = self._search_regex(
+                        r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
+                        'vimeo clip page config')
+                    config_url = self._parse_json(
+                        vimeo_clip_page_config, video_id)['player']['config_url']
                 config_json = self._download_webpage(config_url, video_id)
                 config = json.loads(config_json)
             except RegexNotFoundError:
@@ -391,14 +406,20 @@ class VimeoIE(VimeoBaseInfoExtractor):
                     'ext': codec_extension,
                     'url': video_url,
                     'format_id': format_id,
-                    'width': file_info.get('width'),
-                    'height': file_info.get('height'),
+                    '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')
+        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]
-        if len(formats) == 0:
-            raise ExtractorError('No known codec found')
+        self._sort_formats(formats)
 
         subtitles = {}
         text_tracks = config['request'].get('text_tracks')
@@ -459,7 +480,7 @@ class VimeoChannelIE(VimeoBaseInfoExtractor):
         if password is None:
             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
         fields = self._hidden_inputs(login_form)
-        token = self._extract_xsrft(webpage)
+        token, vuid = self._extract_xsrft_and_vuid(webpage)
         fields['token'] = token
         fields['password'] = password
         post = urlencode_postdata(fields)
@@ -468,6 +489,7 @@ class VimeoChannelIE(VimeoBaseInfoExtractor):
         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)
 
         return self._download_webpage(

youtube_dl/extractor/vine.py

@@ -1,10 +1,14 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
 import itertools
 
 from .common import InfoExtractor
-from ..utils import unified_strdate
+from ..utils import (
+    int_or_none,
+    unified_strdate,
+)
 
 
 class VineIE(InfoExtractor):
@@ -17,10 +21,12 @@ class VineIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Chicken.',
             'alt_title': 'Vine by Jack Dorsey',
-            'description': 'Chicken.',
             'upload_date': '20130519',
             'uploader': 'Jack Dorsey',
             'uploader_id': '76',
+            'like_count': int,
+            'comment_count': int,
+            'repost_count': int,
         },
     }, {
         'url': 'https://vine.co/v/MYxVapFvz2z',
@@ -29,11 +35,13 @@ class VineIE(InfoExtractor):
             'id': 'MYxVapFvz2z',
             'ext': 'mp4',
             'title': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14',
-            'alt_title': 'Vine by Luna',
-            'description': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14',
+            'alt_title': 'Vine by Mars Ruiz',
             'upload_date': '20140815',
-            'uploader': 'Luna',
+            'uploader': 'Mars Ruiz',
             'uploader_id': '1102363502380728320',
+            'like_count': int,
+            'comment_count': int,
+            'repost_count': int,
         },
     }, {
         'url': 'https://vine.co/v/bxVjBbZlPUH',
@@ -43,14 +51,33 @@ class VineIE(InfoExtractor):
             'ext': 'mp4',
             'title': '#mw3 #ac130 #killcam #angelofdeath',
             'alt_title': 'Vine by Z3k3',
-            'description': '#mw3 #ac130 #killcam #angelofdeath',
             'upload_date': '20130430',
             'uploader': 'Z3k3',
             'uploader_id': '936470460173008896',
+            'like_count': int,
+            'comment_count': int,
+            'repost_count': int,
         },
     }, {
         'url': 'https://vine.co/oembed/MYxVapFvz2z.json',
         'only_matching': True,
+    }, {
+        'url': 'https://vine.co/v/e192BnZnZ9V',
+        'info_dict': {
+            'id': 'e192BnZnZ9V',
+            'ext': 'mp4',
+            'title': 'ยิ้ม~ เขิน~ อาย~ น่าร้ากอ้ะ >//< @n_whitewo @orlameena #lovesicktheseries  #lovesickseason2',
+            'alt_title': 'Vine by Pimry_zaa',
+            'upload_date': '20150705',
+            'uploader': 'Pimry_zaa',
+            'uploader_id': '1135760698325307392',
+            'like_count': int,
+            'comment_count': int,
+            'repost_count': int,
+        },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
@@ -65,25 +92,26 @@ class VineIE(InfoExtractor):
 
         formats = [{
             'format_id': '%(format)s-%(rate)s' % f,
-            'vcodec': f['format'],
-            'quality': f['rate'],
+            'vcodec': f.get('format'),
+            'quality': f.get('rate'),
             'url': f['videoUrl'],
-        } for f in data['videoUrls']]
+        } for f in data['videoUrls'] if f.get('videoUrl')]
 
         self._sort_formats(formats)
 
+        username = data.get('username')
+
         return {
             'id': video_id,
-            'title': self._og_search_title(webpage),
-            'alt_title': self._og_search_description(webpage, default=None),
-            'description': data['description'],
-            'thumbnail': data['thumbnailUrl'],
-            'upload_date': unified_strdate(data['created']),
-            'uploader': data['username'],
-            'uploader_id': data['userIdStr'],
-            'like_count': data['likes']['count'],
-            'comment_count': data['comments']['count'],
-            'repost_count': data['reposts']['count'],
+            'title': data.get('description') or self._og_search_title(webpage),
+            'alt_title': 'Vine by %s' % username if username else self._og_search_description(webpage, default=None),
+            'thumbnail': data.get('thumbnailUrl'),
+            'upload_date': unified_strdate(data.get('created')),
+            'uploader': username,
+            'uploader_id': data.get('userIdStr'),
+            'like_count': int_or_none(data.get('likes', {}).get('count')),
+            'comment_count': int_or_none(data.get('comments', {}).get('count')),
+            'repost_count': int_or_none(data.get('reposts', {}).get('count')),
             'formats': formats,
         }
 

youtube_dl/extractor/yandexmusic.py

@@ -46,6 +46,12 @@ class YandexMusicTrackIE(InfoExtractor):
                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
 
     def _get_track_info(self, track):
+        thumbnail = None
+        cover_uri = track.get('albums', [{}])[0].get('coverUri')
+        if cover_uri:
+            thumbnail = cover_uri.replace('%%', 'orig')
+            if not thumbnail.startswith('http'):
+                thumbnail = 'http://' + thumbnail
         return {
             'id': track['id'],
             'ext': 'mp3',
@@ -53,6 +59,7 @@ class YandexMusicTrackIE(InfoExtractor):
             'title': '%s - %s' % (track['artists'][0]['name'], track['title']),
             'filesize': int_or_none(track.get('fileSize')),
             'duration': float_or_none(track.get('durationMs'), 1000),
+            'thumbnail': thumbnail,
         }
 
     def _real_extract(self, url):

youtube_dl/extractor/youtube.py

@@ -178,6 +178,52 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             return
 
 
+class YoutubePlaylistBaseInfoExtractor(InfoExtractor):
+    # Extract the video ids from the playlist pages
+    def _entries(self, page, playlist_id):
+        more_widget_html = content_html = page
+        for page_num in itertools.count(1):
+            for video_id, video_title in self.extract_videos_from_page(content_html):
+                yield self.url_result(
+                    video_id, 'Youtube', video_id=video_id,
+                    video_title=video_title)
+
+            mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
+            if not mobj:
+                break
+
+            more = self._download_json(
+                'https://youtube.com/%s' % mobj.group('more'), playlist_id,
+                'Downloading page #%s' % page_num,
+                transform_source=uppercase_escape)
+            content_html = more['content_html']
+            if not content_html.strip():
+                # Some webpages show a "Load more" button but they don't
+                # have more videos
+                break
+            more_widget_html = more['load_more_widget_html']
+
+    def extract_videos_from_page(self, page):
+        ids_in_page = []
+        titles_in_page = []
+        for mobj in re.finditer(self._VIDEO_RE, page):
+            # The link with index 0 is not the first video of the playlist (not sure if still actual)
+            if 'index' in mobj.groupdict() and mobj.group('id') == '0':
+                continue
+            video_id = mobj.group('id')
+            video_title = unescapeHTML(mobj.group('title'))
+            if video_title:
+                video_title = video_title.strip()
+            try:
+                idx = ids_in_page.index(video_id)
+                if video_title and not titles_in_page[idx]:
+                    titles_in_page[idx] = video_title
+            except ValueError:
+                ids_in_page.append(video_id)
+                titles_in_page.append(video_title)
+        return zip(ids_in_page, titles_in_page)
+
+
 class YoutubeIE(YoutubeBaseInfoExtractor):
     IE_DESC = 'YouTube.com'
     _VALID_URL = r"""(?x)^
@@ -1419,7 +1465,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         }
 
 
-class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
+class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor):
     IE_DESC = 'YouTube.com playlists'
     _VALID_URL = r"""(?x)(?:
                         (?:https?://)?
@@ -1440,7 +1486,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
                         ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
                      )"""
     _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
-    _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
+    _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
     IE_NAME = 'youtube:playlist'
     _TESTS = [{
         'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
@@ -1557,37 +1603,11 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
             else:
                 self.report_warning('Youtube gives an alert message: ' + match)
 
-        # Extract the video ids from the playlist pages
-        def _entries():
-            more_widget_html = content_html = page
-            for page_num in itertools.count(1):
-                matches = re.finditer(self._VIDEO_RE, content_html)
-                # We remove the duplicates and the link with index 0
-                # (it's not the first video of the playlist)
-                new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
-                for vid_id in new_ids:
-                    yield self.url_result(vid_id, 'Youtube', video_id=vid_id)
-
-                mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
-                if not mobj:
-                    break
-
-                more = self._download_json(
-                    'https://youtube.com/%s' % mobj.group('more'), playlist_id,
-                    'Downloading page #%s' % page_num,
-                    transform_source=uppercase_escape)
-                content_html = more['content_html']
-                if not content_html.strip():
-                    # Some webpages show a "Load more" button but they don't
-                    # have more videos
-                    break
-                more_widget_html = more['load_more_widget_html']
-
         playlist_title = self._html_search_regex(
             r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
             page, 'title')
 
-        return self.playlist_result(_entries(), playlist_id, playlist_title)
+        return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
 
     def _real_extract(self, url):
         # Extract playlist id
@@ -1613,10 +1633,11 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         return self._extract_playlist(playlist_id)
 
 
-class YoutubeChannelIE(InfoExtractor):
+class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
     IE_DESC = 'YouTube.com channels'
     _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
     _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
+    _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
     IE_NAME = 'youtube:channel'
     _TESTS = [{
         'note': 'paginated channel',
@@ -1627,22 +1648,6 @@ class YoutubeChannelIE(InfoExtractor):
         }
     }]
 
-    @staticmethod
-    def extract_videos_from_page(page):
-        ids_in_page = []
-        titles_in_page = []
-        for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
-            video_id = mobj.group('id')
-            video_title = unescapeHTML(mobj.group('title'))
-            try:
-                idx = ids_in_page.index(video_id)
-                if video_title and not titles_in_page[idx]:
-                    titles_in_page[idx] = video_title
-            except ValueError:
-                ids_in_page.append(video_id)
-                titles_in_page.append(video_title)
-        return zip(ids_in_page, titles_in_page)
-
     def _real_extract(self, url):
         channel_id = self._match_id(url)
 
@@ -1685,29 +1690,7 @@ class YoutubeChannelIE(InfoExtractor):
                 for video_id, video_title in self.extract_videos_from_page(channel_page)]
             return self.playlist_result(entries, channel_id)
 
-        def _entries():
-            more_widget_html = content_html = channel_page
-            for pagenum in itertools.count(1):
-
-                for video_id, video_title in self.extract_videos_from_page(content_html):
-                    yield self.url_result(
-                        video_id, 'Youtube', video_id=video_id,
-                        video_title=video_title)
-
-                mobj = re.search(
-                    r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
-                    more_widget_html)
-                if not mobj:
-                    break
-
-                more = self._download_json(
-                    'https://youtube.com/%s' % mobj.group('more'), channel_id,
-                    'Downloading page #%s' % (pagenum + 1),
-                    transform_source=uppercase_escape)
-                content_html = more['content_html']
-                more_widget_html = more['load_more_widget_html']
-
-        return self.playlist_result(_entries(), channel_id)
+        return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
 
 
 class YoutubeUserIE(YoutubeChannelIE):

youtube_dl/extractor/zingmp3.py

@@ -9,9 +9,11 @@ from ..utils import ExtractorError
 
 class ZingMp3BaseInfoExtractor(InfoExtractor):
 
-    def _extract_item(self, item):
+    def _extract_item(self, item, fatal=True):
         error_message = item.find('./errormessage').text
         if error_message:
+            if not fatal:
+                return
             raise ExtractorError(
                 '%s returned error: %s' % (self.IE_NAME, error_message),
                 expected=True)
@@ -43,7 +45,9 @@ class ZingMp3BaseInfoExtractor(InfoExtractor):
             entries = []
 
             for i, item in enumerate(items, 1):
-                entry = self._extract_item(item)
+                entry = self._extract_item(item, fatal=False)
+                if not entry:
+                    continue
                 entry['id'] = '%s-%d' % (id, i)
                 entries.append(entry)
 
@@ -85,7 +89,7 @@ class ZingMp3SongIE(ZingMp3BaseInfoExtractor):
 
 
 class ZingMp3AlbumIE(ZingMp3BaseInfoExtractor):
-    _VALID_URL = r'https?://mp3\.zing\.vn/album/(?P<slug>[^/]+)/(?P<album_id>\w+)\.html'
+    _VALID_URL = r'https?://mp3\.zing\.vn/(?:album|playlist)/(?P<slug>[^/]+)/(?P<album_id>\w+)\.html'
     _TESTS = [{
         'url': 'http://mp3.zing.vn/album/Lau-Dai-Tinh-Ai-Bang-Kieu-Minh-Tuyet/ZWZBWDAF.html',
         'info_dict': {
@@ -94,6 +98,9 @@ class ZingMp3AlbumIE(ZingMp3BaseInfoExtractor):
             'title': 'Lâu Đài Tình Ái - Bằng Kiều ft. Minh Tuyết | Album 320 lossless',
         },
         'playlist_count': 10,
+    }, {
+        'url': 'http://mp3.zing.vn/playlist/Duong-Hong-Loan-apollobee/IWCAACCB.html',
+        'only_matching': True,
     }]
     IE_NAME = 'zingmp3:album'
     IE_DESC = 'mp3.zing.vn albums'

youtube_dl/utils.py

@@ -3,6 +3,7 @@
 
 from __future__ import unicode_literals
 
+import base64
 import calendar
 import codecs
 import contextlib
@@ -1371,7 +1372,12 @@ def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
             v = getattr(v, get_attr, None)
     if v == '':
         v = None
-    return default if v is None else (int(v) * invscale // scale)
+    if v is None:
+        return default
+    try:
+        return int(v) * invscale // scale
+    except ValueError:
+        return default
 
 
 def str_or_none(v, default=None):
@@ -1387,7 +1393,12 @@ def str_to_int(int_str):
 
 
 def float_or_none(v, scale=1, invscale=1, default=None):
-    return default if v is None else (float(v) * invscale / scale)
+    if v is None:
+        return default
+    try:
+        return float(v) * invscale / scale
+    except ValueError:
+        return default
 
 
 def parse_duration(s):
@@ -1785,6 +1796,10 @@ def urlhandle_detect_ext(url_handle):
     return mimetype2ext(getheader('Content-Type'))
 
 
+def encode_data_uri(data, mime_type):
+    return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii'))
+
+
 def age_restricted(content_limit, age_limit):
     """ Returns True iff the content should be blocked """
 

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.10.09'
+__version__ = '2015.10.18'