release 2015.07.28

That's how YouTube reports them in their DASH manifest.

AUTHORS

@@ -127,3 +127,11 @@ Julian Richen
 Ping O.
 Mister Hat
 Peter Ding
+jackyzy823
+George Brighton
+Remita Amine
+Aurélio A. Heckert
+Bernhard Minks
+sceext
+Zach Bruggeman
+Tjark Saul

README.md

@@ -54,6 +54,7 @@ which means you can modify it, redistribute it or use it however you like.
     --dump-user-agent                Display the current browser identification
     --list-extractors                List all supported extractors
     --extractor-descriptions         Output descriptions of all supported extractors
+    --force-generic-extractor        Force extraction to use the generic extractor
     --default-search PREFIX          Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for youtube-dl "large apple".
                                      Use the value "auto" to let youtube-dl guess ("auto_warning" to emit a warning when guessing). "error" just throws an error. The
                                      default value "fixup_error" repairs broken URLs, but emits an error if this is not possible instead of searching.
@@ -74,7 +75,7 @@ which means you can modify it, redistribute it or use it however you like.
 ## Video Selection:
     --playlist-start NUMBER          Playlist video to start at (default is 1)
     --playlist-end NUMBER            Playlist video to end at (default is last)
-    --playlist-items ITEM_SPEC       Playlist video items to download. Specify indices of the videos in the playlist seperated by commas like: "--playlist-items 1,2,5,8"
+    --playlist-items ITEM_SPEC       Playlist video items to download. Specify indices of the videos in the playlist separated by commas like: "--playlist-items 1,2,5,8"
                                      if you want to download videos indexed 1, 2, 5, 8 in the playlist. You can specify range: "--playlist-items 1-3,7,10-13", it will
                                      download the videos at index 1, 2, 3, 7, 10, 11, 12 and 13.
     --match-title REGEX              Download only matching titles (regex or caseless sub-string)
@@ -107,7 +108,7 @@ which means you can modify it, redistribute it or use it however you like.
     --playlist-reverse               Download playlist videos in reverse order
     --xattr-set-filesize             Set file xattribute ytdl.filesize with expected filesize (experimental)
     --hls-prefer-native              Use the native HLS downloader instead of ffmpeg (experimental)
-    --external-downloader COMMAND    Use the specified external downloader. Currently supports aria2c,curl,wget
+    --external-downloader COMMAND    Use the specified external downloader. Currently supports aria2c,curl,httpie,wget
     --external-downloader-args ARGS  Give these arguments to the external downloader
 
 ## Filesystem Options:
@@ -189,8 +190,8 @@ which means you can modify it, redistribute it or use it however you like.
     --all-formats                    Download all available video formats
     --prefer-free-formats            Prefer free video formats unless a specific one is requested
     -F, --list-formats               List all available formats
-    --youtube-skip-dash-manifest     Do not download the DASH manifest on YouTube videos
-    --merge-output-format FORMAT     If a merge is required (e.g. bestvideo+bestaudio), output to given container format. One of mkv, mp4, ogg, webm, flv.Ignored if no
+    --youtube-skip-dash-manifest     Do not download the DASH manifests and related data on YouTube videos
+    --merge-output-format FORMAT     If a merge is required (e.g. bestvideo+bestaudio), output to given container format. One of mkv, mp4, ogg, webm, flv. Ignored if no
                                      merge is required
 
 ## Subtitle Options:
@@ -213,7 +214,8 @@ which means you can modify it, redistribute it or use it however you like.
     --audio-format FORMAT            Specify audio format: "best", "aac", "vorbis", "mp3", "m4a", "opus", or "wav"; "best" by default
     --audio-quality QUALITY          Specify ffmpeg/avconv audio quality, insert a value between 0 (better) and 9 (worse) for VBR or a specific bitrate like 128K (default
                                      5)
-    --recode-video FORMAT            Encode the video to another format if necessary (currently supported: mp4|flv|ogg|webm|mkv)
+    --recode-video FORMAT            Encode the video to another format if necessary (currently supported: mp4|flv|ogg|webm|mkv|avi)
+    --postprocessor-args ARGS        Give these arguments to the postprocessor
     -k, --keep-video                 Keep the video file on disk after the post-processing; the video is erased by default
     --no-post-overwrites             Do not overwrite post-processed files; the post-processed files are overwritten by default
     --embed-subs                     Embed subtitles in the video (only for mkv and mp4 videos)
@@ -236,6 +238,26 @@ which means you can modify it, redistribute it or use it however you like.
 
 You can configure youtube-dl by placing default arguments (such as `--extract-audio --no-mtime` to always extract the audio and not copy the mtime) into `/etc/youtube-dl.conf` and/or `~/.config/youtube-dl/config`. On Windows, the configuration file locations are `%APPDATA%\youtube-dl\config.txt` and `C:\Users\<user name>\youtube-dl.conf`.
 
+### Authentication with `.netrc` file ###
+
+You may also want to configure automatic credentials storage for extractors that support authentication (by providing login and password with `--username` and `--password`) in order not to pass credentials as command line arguments on every youtube-dl execution and prevent tracking plain text passwords in shell command history. You can achieve this using [`.netrc` file](http://stackoverflow.com/tags/.netrc/info) on per extractor basis. For that you will need to create `.netrc` file in your `$HOME` and restrict permissions to read/write by you only:
+```
+touch $HOME/.netrc
+chmod a-rwx,u+rw $HOME/.netrc
+```
+After that you can add credentials for extractor in the following format, where *extractor* is the name of extractor in lowercase:
+```
+machine <extractor> login <login> password <password>
+```
+For example:
+```
+machine youtube login myaccount@gmail.com password my_youtube_password
+machine twitch login my_twitch_account_name password my_twitch_password
+```
+To activate authentication with `.netrc` file you should pass `--netrc` to youtube-dl or to place it in [configuration file](#configuration).
+
+On Windows you may also need to setup `%HOME%` environment variable manually.
+
 # OUTPUT TEMPLATE
 
 The `-o` option allows users to indicate a template for the output file names. The basic usage is not to set any template arguments when downloading a single file, like in `youtube-dl -o funny_video.flv "http://some/video"`. However, it may contain special sequences that will be replaced when downloading each video. The special sequences have the format `%(NAME)s`. To clarify, that is a percent symbol followed by a name in parenthesis, followed by a lowercase S. Allowed names are:
@@ -379,7 +401,7 @@ In February 2015, the new YouTube player contained a character sequence in a str
 
 ### HTTP Error 429: Too Many Requests or 402: Payment Required
 
-These two error codes indicate that the service is blocking your IP address because of overuse. Contact the service and ask them to unblock your IP address, or - if you have acquired a whitelisted IP address already - use the [`--proxy` or `--network-address` options](#network-options) to select another IP address.
+These two error codes indicate that the service is blocking your IP address because of overuse. Contact the service and ask them to unblock your IP address, or - if you have acquired a whitelisted IP address already - use the [`--proxy` or `--source-address` options](#network-options) to select another IP address.
 
 ### SyntaxError: Non-ASCII character ###
 

docs/supportedsites.md

@@ -17,6 +17,7 @@
  - **AcademicEarth:Course**
  - **AddAnime**
  - **AdobeTV**
+ - **AdobeTVVideo**
  - **AdultSwim**
  - **Aftenposten**
  - **Aftonbladet**
@@ -27,7 +28,8 @@
  - **anitube.se**
  - **AnySex**
  - **Aparat**
- - **AppleDaily**
+ - **AppleConnect**
+ - **AppleDaily**: 臺灣蘋果日報
  - **AppleTrailers**
  - **archive.org**: archive.org videos
  - **ARD**
@@ -44,11 +46,12 @@
  - **audiomack**
  - **audiomack:album**
  - **Azubu**
- - **BaiduVideo**
+ - **BaiduVideo**: 百度视频
  - **bambuser**
  - **bambuser:channel**
  - **Bandcamp**
  - **Bandcamp:album**
+ - **bbc**: BBC
  - **bbc.co.uk**: BBC iPlayer
  - **BeatportPro**
  - **Beeg**
@@ -105,11 +108,12 @@
  - **Crunchyroll**
  - **crunchyroll:playlist**
  - **CSpan**: C-SPAN
- - **CtsNews**
+ - **CtsNews**: 華視新聞
  - **culturebox.francetvinfo.fr**
  - **dailymotion**
  - **dailymotion:playlist**
  - **dailymotion:user**
+ - **DailymotionCloud**
  - **daum.net**
  - **DBTV**
  - **DctpTv**
@@ -119,7 +123,7 @@
  - **Discovery**
  - **divxstage**: DivxStage
  - **Dotsub**
- - **DouyuTV**
+ - **DouyuTV**: 斗鱼
  - **dramafever**
  - **dramafever:series**
  - **DRBonanza**
@@ -220,7 +224,8 @@
  - **instagram:user**: Instagram user profile
  - **InternetVideoArchive**
  - **IPrima**
- - **iqiyi**
+ - **iqiyi**: 爱奇艺
+ - **Ir90Tv**
  - **ivi**: ivi.ru
  - **ivi:compilation**: ivi.ru compilations
  - **Izlesene**
@@ -241,9 +246,16 @@
  - **kontrtube**: KontrTube.ru - Труба зовёт
  - **KrasView**: Красвью
  - **Ku6**
+ - **kuwo:album**: 酷我音乐 - 专辑
+ - **kuwo:category**: 酷我音乐 - 分类
+ - **kuwo:chart**: 酷我音乐 - 排行榜
+ - **kuwo:mv**: 酷我音乐 - MV
+ - **kuwo:singer**: 酷我音乐 - 歌手
+ - **kuwo:song**: 酷我音乐
  - **la7.tv**
  - **Laola1Tv**
- - **Letv**
+ - **Lecture2Go**
+ - **Letv**: 乐视网
  - **LetvPlaylist**
  - **LetvTv**
  - **Libsyn**
@@ -281,6 +293,7 @@
  - **Motherless**
  - **Motorsport**: motorsport.com
  - **MovieClips**
+ - **MovieFap**
  - **Moviezine**
  - **movshare**: MovShare
  - **MPORA**
@@ -294,6 +307,7 @@
  - **MySpace**
  - **MySpace:album**
  - **MySpass**
+ - **Myvi**
  - **myvideo**
  - **MyVidster**
  - **N-JOY**
@@ -309,11 +323,18 @@
  - **NDTV**
  - **NerdCubedFeed**
  - **Nerdist**
+ - **netease:album**: 网易云音乐 - 专辑
+ - **netease:djradio**: 网易云音乐 - 电台
+ - **netease:mv**: 网易云音乐 - MV
+ - **netease:playlist**: 网易云音乐 - 歌单
+ - **netease:program**: 网易云音乐 - 电台节目
+ - **netease:singer**: 网易云音乐 - 歌手
+ - **netease:song**: 网易云音乐
  - **Netzkino**
  - **Newgrounds**
  - **Newstube**
- - **NextMedia**
- - **NextMediaActionNews**
+ - **NextMedia**: 蘋果日報
+ - **NextMediaActionNews**: 蘋果日報 - 動新聞
  - **nfb**: National Film Board of Canada
  - **nfl.com**
  - **nhl.com**
@@ -329,13 +350,14 @@
  - **Nowness**
  - **NowTV**
  - **nowvideo**: NowVideo
- - **npo.nl**
+ - **npo**: npo.nl and ntr.nl
+ - **npo**: npo.nl and ntr.nl
  - **npo.nl:live**
  - **npo.nl:radio**
  - **npo.nl:radio:fragment**
  - **NRK**
  - **NRKPlaylist**
- - **NRKTV**
+ - **NRKTV**: NRK TV and NRK Radio
  - **ntv.ru**
  - **Nuvid**
  - **NYTimes**
@@ -344,6 +366,7 @@
  - **Odnoklassniki**
  - **OktoberfestTV**
  - **on.aol.com**
+ - **OnionStudios**
  - **Ooyala**
  - **OoyalaExternal**
  - **OpenFilm**
@@ -357,6 +380,7 @@
  - **PhilharmonieDeParis**: Philharmonie de Paris
  - **Phoenix**
  - **Photobucket**
+ - **Pinkbike**
  - **Pladform**
  - **PlanetaPlay**
  - **play.fm**
@@ -377,10 +401,11 @@
  - **prosiebensat1**: ProSiebenSat.1 Digital
  - **Puls4**
  - **Pyvideo**
- - **qqmusic**
- - **qqmusic:album**
- - **qqmusic:singer**
- - **qqmusic:toplist**
+ - **qqmusic**: QQ音乐
+ - **qqmusic:album**: QQ音乐 - 专辑
+ - **qqmusic:playlist**: QQ音乐 - 歌单
+ - **qqmusic:singer**: QQ音乐 - 歌手
+ - **qqmusic:toplist**: QQ音乐 - 排行榜
  - **QuickVid**
  - **R7**
  - **radio.de**
@@ -389,6 +414,7 @@
  - **RadioJavan**
  - **Rai**
  - **RBMARadio**
+ - **RDS**: RDS.ca
  - **RedTube**
  - **Restudy**
  - **ReverbNation**
@@ -436,6 +462,8 @@
  - **smotri:broadcast**: Smotri.com broadcasts
  - **smotri:community**: Smotri.com community videos
  - **smotri:user**: Smotri.com user videos
+ - **SnagFilms**
+ - **SnagFilmsEmbed**
  - **Snotr**
  - **Sohu**
  - **soompi**
@@ -462,6 +490,7 @@
  - **SportBox**
  - **SportBoxEmbed**
  - **SportDeutschland**
+ - **Sportschau**
  - **Srf**
  - **SRMediathek**: Saarländischer Rundfunk
  - **SSA**
@@ -487,7 +516,6 @@
  - **TechTalks**
  - **techtv.mit.edu**
  - **ted**
- - **tegenlicht.vpro.nl**
  - **TeleBruxelles**
  - **telecinco.es**
  - **TeleMB**
@@ -498,6 +526,7 @@
  - **TheOnion**
  - **ThePlatform**
  - **TheSixtyOne**
+ - **ThisAmericanLife**
  - **ThisAV**
  - **THVideo**
  - **THVideoPlaylist**
@@ -538,10 +567,11 @@
  - **twitch:stream**
  - **twitch:video**
  - **twitch:vod**
+ - **TwitterCard**
  - **Ubu**
  - **udemy**
  - **udemy:course**
- - **UDNEmbed**
+ - **UDNEmbed**: 聯合影音
  - **Ultimedia**
  - **Unistra**
  - **Urort**: NRK P3 Urørt
@@ -586,8 +616,8 @@
  - **Vimple**: Vimple - one-click video hosting
  - **Vine**
  - **vine:user**
- - **vk.com**
- - **vk.com:user-videos**: vk.com:All of a user's videos
+ - **vk**: VK
+ - **vk:uservideos**: VK - User's Videos
  - **Vodlocker**
  - **VoiceRepublic**
  - **Vporn**
@@ -603,32 +633,36 @@
  - **wdr:mobile**
  - **WDRMaus**: Sendung mit der Maus
  - **WebOfStories**
+ - **WebOfStoriesPlaylist**
  - **Weibo**
  - **Wimp**
  - **Wistia**
+ - **WNL**
  - **WorldStarHipHop**
  - **wrzuta.pl**
  - **WSJ**: Wall Street Journal
  - **XBef**
  - **XboxClips**
  - **XHamster**
+ - **XHamsterEmbed**
  - **XMinus**
  - **XNXX**
  - **Xstream**
  - **XTube**
  - **XTubeUser**: XTube user profile
- - **Xuite**
+ - **Xuite**: 隨意窩Xuite影音
  - **XVideos**
  - **XXXYMovies**
  - **Yahoo**: Yahoo screen and movies
- - **Yam**
+ - **Yam**: 蕃薯藤yam天空部落
  - **yandexmusic:album**: Яндекс.Музыка - Альбом
  - **yandexmusic:playlist**: Яндекс.Музыка - Плейлист
  - **yandexmusic:track**: Яндекс.Музыка - Трек
  - **YesJapan**
+ - **yinyuetai:video**: 音悦Tai
  - **Ynet**
  - **YouJizz**
- - **Youku**
+ - **youku**: 优酷
  - **YouPorn**
  - **YourUpload**
  - **youtube**: YouTube.com

test/test_compat.py

@@ -14,6 +14,8 @@ from youtube_dl.utils import get_filesystem_encoding
 from youtube_dl.compat import (
     compat_getenv,
     compat_expanduser,
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_unquote_plus,
 )
 
 
@@ -42,5 +44,28 @@ class TestCompat(unittest.TestCase):
             dir(youtube_dl.compat))) - set(['unicode_literals'])
         self.assertEqual(all_names, sorted(present_names))
 
+    def test_compat_urllib_parse_unquote(self):
+        self.assertEqual(compat_urllib_parse_unquote('abc%20def'), 'abc def')
+        self.assertEqual(compat_urllib_parse_unquote('%7e/abc+def'), '~/abc+def')
+        self.assertEqual(compat_urllib_parse_unquote(''), '')
+        self.assertEqual(compat_urllib_parse_unquote('%'), '%')
+        self.assertEqual(compat_urllib_parse_unquote('%%'), '%%')
+        self.assertEqual(compat_urllib_parse_unquote('%%%'), '%%%')
+        self.assertEqual(compat_urllib_parse_unquote('%2F'), '/')
+        self.assertEqual(compat_urllib_parse_unquote('%2f'), '/')
+        self.assertEqual(compat_urllib_parse_unquote('%E6%B4%A5%E6%B3%A2'), '津波')
+        self.assertEqual(
+            compat_urllib_parse_unquote('''<meta property="og:description" content="%E2%96%81%E2%96%82%E2%96%83%E2%96%84%25%E2%96%85%E2%96%86%E2%96%87%E2%96%88" />
+%<a href="https://ar.wikipedia.org/wiki/%D8%AA%D8%B3%D9%88%D9%86%D8%A7%D9%85%D9%8A">%a'''),
+            '''<meta property="og:description" content="▁▂▃▄%▅▆▇█" />
+%<a href="https://ar.wikipedia.org/wiki/تسونامي">%a''')
+        self.assertEqual(
+            compat_urllib_parse_unquote('''%28%5E%E2%97%A3_%E2%97%A2%5E%29%E3%81%A3%EF%B8%BB%E3%83%87%E2%95%90%E4%B8%80    %E2%87%80    %E2%87%80    %E2%87%80    %E2%87%80    %E2%87%80    %E2%86%B6%I%Break%25Things%'''),
+            '''(^◣_◢^)っ︻デ═一    ⇀    ⇀    ⇀    ⇀    ⇀    ↶%I%Break%Things%''')
+
+    def test_compat_urllib_parse_unquote_plus(self):
+        self.assertEqual(compat_urllib_parse_unquote_plus('abc%20def'), 'abc def')
+        self.assertEqual(compat_urllib_parse_unquote_plus('%7e/abc+def'), '~/abc def')
+
 if __name__ == '__main__':
     unittest.main()

test/test_utils.py

@@ -324,6 +324,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('02:03:04'), 7384)
         self.assertEqual(parse_duration('01:02:03:04'), 93784)
         self.assertEqual(parse_duration('1 hour 3 minutes'), 3780)
+        self.assertEqual(parse_duration('87 Min.'), 5220)
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(

youtube_dl/YoutubeDL.py

@@ -139,6 +139,7 @@ class YoutubeDL(object):
     outtmpl:           Template for output names.
     restrictfilenames: Do not allow "&" and spaces in file names
     ignoreerrors:      Do not stop on download errors.
+    force_generic_extractor: Force downloader to use the generic extractor
     nooverwrites:      Prevent overwriting files.
     playliststart:     Playlist item to start at.
     playlistend:       Playlist item to end at.
@@ -261,6 +262,8 @@ class YoutubeDL(object):
     The following options are used by the post processors:
     prefer_ffmpeg:     If True, use ffmpeg instead of avconv if both are available,
                        otherwise prefer avconv.
+    postprocessor_args: A list of additional command-line arguments for the
+                        postprocessor.
     """
 
     params = None
@@ -626,13 +629,16 @@ class YoutubeDL(object):
             info_dict.setdefault(key, value)
 
     def extract_info(self, url, download=True, ie_key=None, extra_info={},
-                     process=True):
+                     process=True, force_generic_extractor=False):
         '''
         Returns a list with a dictionary for each video we find.
         If 'download', also downloads the videos.
         extra_info is a dict containing the extra values to add to each result
         '''
 
+        if not ie_key and force_generic_extractor:
+            ie_key = 'Generic'
+
         if ie_key:
             ies = [self.get_info_extractor(ie_key)]
         else:
@@ -1004,7 +1010,7 @@ class YoutubeDL(object):
                 t.get('preference'), t.get('width'), t.get('height'),
                 t.get('id'), t.get('url')))
             for i, t in enumerate(thumbnails):
-                if 'width' in t and 'height' in t:
+                if t.get('width') and t.get('height'):
                     t['resolution'] = '%dx%d' % (t['width'], t['height'])
                 if t.get('id') is None:
                     t['id'] = '%d' % i
@@ -1033,12 +1039,6 @@ class YoutubeDL(object):
             info_dict['id'], info_dict.get('subtitles'),
             info_dict.get('automatic_captions'))
 
-        # This extractors handle format selection themselves
-        if info_dict['extractor'] in ['Youku']:
-            if download:
-                self.process_info(info_dict)
-            return info_dict
-
         # We now pick which formats have to be downloaded
         if info_dict.get('formats') is None:
             # There's only one format available
@@ -1104,7 +1104,8 @@ class YoutubeDL(object):
         if req_format is None:
             req_format_list = []
             if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
-                    info_dict['extractor'] in ['youtube', 'ted']):
+                    info_dict['extractor'] in ['youtube', 'ted'] and
+                    not info_dict.get('is_live')):
                 merger = FFmpegMergerPP(self)
                 if merger.available and merger.can_merge():
                     req_format_list.append('bestvideo+bestaudio')
@@ -1499,7 +1500,8 @@ class YoutubeDL(object):
         for url in url_list:
             try:
                 # It also downloads the videos
-                res = self.extract_info(url)
+                res = self.extract_info(
+                    url, force_generic_extractor=self.params.get('force_generic_extractor', False))
             except UnavailableVideoError:
                 self.report_error('unable to download video')
             except MaxDownloadsReached:

youtube_dl/__init__.py

@@ -169,7 +169,7 @@ def _real_main(argv=None):
         if not opts.audioquality.isdigit():
             parser.error('invalid audio quality specified')
     if opts.recodevideo is not None:
-        if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv']:
+        if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv', 'avi']:
             parser.error('invalid video recode format specified')
     if opts.convertsubtitles is not None:
         if opts.convertsubtitles not in ['srt', 'vtt', 'ass']:
@@ -263,6 +263,9 @@ def _real_main(argv=None):
     external_downloader_args = None
     if opts.external_downloader_args:
         external_downloader_args = shlex.split(opts.external_downloader_args)
+    postprocessor_args = None
+    if opts.postprocessor_args:
+        postprocessor_args = shlex.split(opts.postprocessor_args)
     match_filter = (
         None if opts.match_filter is None
         else match_filter_func(opts.match_filter))
@@ -293,6 +296,7 @@ def _real_main(argv=None):
         'autonumber_size': opts.autonumber_size,
         'restrictfilenames': opts.restrictfilenames,
         'ignoreerrors': opts.ignoreerrors,
+        'force_generic_extractor': opts.force_generic_extractor,
         'ratelimit': opts.ratelimit,
         'nooverwrites': opts.nooverwrites,
         'retries': opts_retries,
@@ -366,6 +370,7 @@ def _real_main(argv=None):
         'ffmpeg_location': opts.ffmpeg_location,
         'hls_prefer_native': opts.hls_prefer_native,
         'external_downloader_args': external_downloader_args,
+        'postprocessor_args': postprocessor_args,
         'cn_verification_proxy': opts.cn_verification_proxy,
     }
 

youtube_dl/compat.py

@@ -9,6 +9,7 @@ import shutil
 import socket
 import subprocess
 import sys
+import itertools
 
 
 try:
@@ -74,42 +75,74 @@ except ImportError:
     import BaseHTTPServer as compat_http_server
 
 try:
+    from urllib.parse import unquote_to_bytes as compat_urllib_parse_unquote_to_bytes
     from urllib.parse import unquote as compat_urllib_parse_unquote
-except ImportError:
-    def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'):
-        if string == '':
+    from urllib.parse import unquote_plus as compat_urllib_parse_unquote_plus
+except ImportError:  # Python 2
+    _asciire = (compat_urllib_parse._asciire if hasattr(compat_urllib_parse, '_asciire')
+                else re.compile('([\x00-\x7f]+)'))
+
+    # HACK: The following are the correct unquote_to_bytes, unquote and unquote_plus
+    # implementations from cpython 3.4.3's stdlib. Python 2's version
+    # is apparently broken (see https://github.com/rg3/youtube-dl/pull/6244)
+
+    def compat_urllib_parse_unquote_to_bytes(string):
+        """unquote_to_bytes('abc%20def') -> b'abc def'."""
+        # Note: strings are encoded as UTF-8. This is only an issue if it contains
+        # unescaped non-ASCII characters, which URIs should not.
+        if not string:
+            # Is it a string-like object?
+            string.split
+            return b''
+        if isinstance(string, unicode):
+            string = string.encode('utf-8')
+        bits = string.split(b'%')
+        if len(bits) == 1:
             return string
-        res = string.split('%')
-        if len(res) == 1:
+        res = [bits[0]]
+        append = res.append
+        for item in bits[1:]:
+            try:
+                append(compat_urllib_parse._hextochr[item[:2]])
+                append(item[2:])
+            except KeyError:
+                append(b'%')
+                append(item)
+        return b''.join(res)
+
+    def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'):
+        """Replace %xx escapes by their single-character equivalent. The optional
+        encoding and errors parameters specify how to decode percent-encoded
+        sequences into Unicode characters, as accepted by the bytes.decode()
+        method.
+        By default, percent-encoded sequences are decoded with UTF-8, and invalid
+        sequences are replaced by a placeholder character.
+
+        unquote('abc%20def') -> 'abc def'.
+        """
+        if '%' not in string:
+            string.split
             return string
         if encoding is None:
             encoding = 'utf-8'
         if errors is None:
             errors = 'replace'
-        # pct_sequence: contiguous sequence of percent-encoded bytes, decoded
-        pct_sequence = b''
-        string = res[0]
-        for item in res[1:]:
-            try:
-                if not item:
-                    raise ValueError
-                pct_sequence += item[:2].decode('hex')
-                rest = item[2:]
-                if not rest:
-                    # This segment was just a single percent-encoded character.
-                    # May be part of a sequence of code units, so delay decoding.
-                    # (Stored in pct_sequence).
-                    continue
-            except ValueError:
-                rest = '%' + item
-            # Encountered non-percent-encoded characters. Flush the current
-            # pct_sequence.
-            string += pct_sequence.decode(encoding, errors) + rest
-            pct_sequence = b''
-        if pct_sequence:
-            # Flush the final pct_sequence
-            string += pct_sequence.decode(encoding, errors)
-        return string
+        bits = _asciire.split(string)
+        res = [bits[0]]
+        append = res.append
+        for i in range(1, len(bits), 2):
+            append(compat_urllib_parse_unquote_to_bytes(bits[i]).decode(encoding, errors))
+            append(bits[i + 1])
+        return ''.join(res)
+
+    def compat_urllib_parse_unquote_plus(string, encoding='utf-8', errors='replace'):
+        """Like unquote(), but also replace plus signs by spaces, as required for
+        unquoting HTML form values.
+
+        unquote_plus('%7e/abc+def') -> '~/abc def'
+        """
+        string = string.replace('+', ' ')
+        return compat_urllib_parse_unquote(string, encoding, errors)
 
 try:
     compat_str = unicode  # Python 2
@@ -388,6 +421,15 @@ else:
             pass
         return _terminal_size(columns, lines)
 
+try:
+    itertools.count(start=0, step=1)
+    compat_itertools_count = itertools.count
+except TypeError:  # Python 2.6
+    def compat_itertools_count(start=0, step=1):
+        n = start
+        while True:
+            yield n
+            n += step
 
 __all__ = [
     'compat_HTTPError',
@@ -401,6 +443,7 @@ __all__ = [
     'compat_html_entities',
     'compat_http_client',
     'compat_http_server',
+    'compat_itertools_count',
     'compat_kwargs',
     'compat_ord',
     'compat_parse_qs',
@@ -411,6 +454,8 @@ __all__ = [
     'compat_urllib_error',
     'compat_urllib_parse',
     'compat_urllib_parse_unquote',
+    'compat_urllib_parse_unquote_plus',
+    'compat_urllib_parse_unquote_to_bytes',
     'compat_urllib_parse_urlparse',
     'compat_urllib_request',
     'compat_urlparse',

youtube_dl/downloader/__init__.py

@@ -8,6 +8,7 @@ from .hls import NativeHlsFD
 from .http import HttpFD
 from .rtsp import RtspFD
 from .rtmp import RtmpFD
+from .dash import DashSegmentsFD
 
 from ..utils import (
     determine_protocol,
@@ -20,6 +21,7 @@ PROTOCOL_MAP = {
     'mms': RtspFD,
     'rtsp': RtspFD,
     'f4m': F4mFD,
+    'http_dash_segments': DashSegmentsFD,
 }
 
 

youtube_dl/downloader/dash.py

@@ -0,0 +1,66 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import FileDownloader
+from ..compat import compat_urllib_request
+
+
+class DashSegmentsFD(FileDownloader):
+    """
+    Download segments in a DASH manifest
+    """
+    def real_download(self, filename, info_dict):
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+        base_url = info_dict['url']
+        segment_urls = info_dict['segment_urls']
+
+        is_test = self.params.get('test', False)
+        remaining_bytes = self._TEST_FILE_SIZE if is_test else None
+        byte_counter = 0
+
+        def append_url_to_file(outf, target_url, target_name, remaining_bytes=None):
+            self.to_screen('[DashSegments] %s: Downloading %s' % (info_dict['id'], target_name))
+            req = compat_urllib_request.Request(target_url)
+            if remaining_bytes is not None:
+                req.add_header('Range', 'bytes=0-%d' % (remaining_bytes - 1))
+
+            data = self.ydl.urlopen(req).read()
+
+            if remaining_bytes is not None:
+                data = data[:remaining_bytes]
+
+            outf.write(data)
+            return len(data)
+
+        def combine_url(base_url, target_url):
+            if re.match(r'^https?://', target_url):
+                return target_url
+            return '%s%s%s' % (base_url, '' if base_url.endswith('/') else '/', target_url)
+
+        with open(tmpfilename, 'wb') as outf:
+            append_url_to_file(
+                outf, combine_url(base_url, info_dict['initialization_url']),
+                'initialization segment')
+            for i, segment_url in enumerate(segment_urls):
+                segment_len = append_url_to_file(
+                    outf, combine_url(base_url, segment_url),
+                    'segment %d / %d' % (i + 1, len(segment_urls)),
+                    remaining_bytes)
+                byte_counter += segment_len
+                if remaining_bytes is not None:
+                    remaining_bytes -= segment_len
+                    if remaining_bytes <= 0:
+                        break
+
+        self.try_rename(tmpfilename, filename)
+
+        self._hook_progress({
+            'downloaded_bytes': byte_counter,
+            'total_bytes': byte_counter,
+            'filename': filename,
+            'status': 'finished',
+        })
+
+        return True

youtube_dl/downloader/external.py

@@ -109,6 +109,14 @@ class Aria2cFD(ExternalFD):
         cmd += ['--', info_dict['url']]
         return cmd
 
+
+class HttpieFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = ['http', '--download', '--output', tmpfilename, info_dict['url']]
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['%s:%s' % (key, val)]
+        return cmd
+
 _BY_NAME = dict(
     (klass.get_basename(), klass)
     for name, klass in globals().items()
@@ -123,5 +131,6 @@ def list_external_downloaders():
 def get_external_downloader(external_downloader):
     """ Given the name of the executable, see whether we support the given
         downloader . """
-    bn = os.path.basename(external_downloader)
+    # Drop .exe extension on Windows
+    bn = os.path.splitext(os.path.basename(external_downloader))[0]
     return _BY_NAME[bn]

youtube_dl/extractor/__init__.py

@@ -4,7 +4,10 @@ from .abc import ABCIE
 from .abc7news import Abc7NewsIE
 from .academicearth import AcademicEarthCourseIE
 from .addanime import AddAnimeIE
-from .adobetv import AdobeTVIE
+from .adobetv import (
+    AdobeTVIE,
+    AdobeTVVideoIE,
+)
 from .adultswim import AdultSwimIE
 from .aftenposten import AftenpostenIE
 from .aftonbladet import AftonbladetIE
@@ -16,9 +19,14 @@ from .anysex import AnySexIE
 from .aol import AolIE
 from .allocine import AllocineIE
 from .aparat import AparatIE
+from .appleconnect import AppleConnectIE
 from .appletrailers import AppleTrailersIE
 from .archiveorg import ArchiveOrgIE
-from .ard import ARDIE, ARDMediathekIE
+from .ard import (
+    ARDIE,
+    ARDMediathekIE,
+    SportschauIE,
+)
 from .arte import (
     ArteTvIE,
     ArteTVPlus7IE,
@@ -35,7 +43,10 @@ from .azubu import AzubuIE
 from .baidu import BaiduVideoIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
-from .bbccouk import BBCCoUkIE
+from .bbc import (
+    BBCCoUkIE,
+    BBCIE,
+)
 from .beeg import BeegIE
 from .behindkink import BehindKinkIE
 from .beatportpro import BeatportProIE
@@ -103,6 +114,7 @@ from .dailymotion import (
     DailymotionIE,
     DailymotionPlaylistIE,
     DailymotionUserIE,
+    DailymotionCloudIE,
 )
 from .daum import DaumIE
 from .dbtv import DBTVIE
@@ -140,7 +152,6 @@ from .ellentv import (
 )
 from .elpais import ElPaisIE
 from .embedly import EmbedlyIE
-from .empflix import EMPFlixIE
 from .engadget import EngadgetIE
 from .eporner import EpornerIE
 from .eroprofile import EroProfileIE
@@ -235,6 +246,7 @@ from .instagram import InstagramIE, InstagramUserIE
 from .internetvideoarchive import InternetVideoArchiveIE
 from .iprima import IPrimaIE
 from .iqiyi import IqiyiIE
+from .ir90tv import Ir90TvIE
 from .ivi import (
     IviIE,
     IviCompilationIE
@@ -257,8 +269,17 @@ from .keek import KeekIE
 from .kontrtube import KontrTubeIE
 from .krasview import KrasViewIE
 from .ku6 import Ku6IE
+from .kuwo import (
+    KuwoIE,
+    KuwoAlbumIE,
+    KuwoChartIE,
+    KuwoSingerIE,
+    KuwoCategoryIE,
+    KuwoMvIE,
+)
 from .la7 import LA7IE
 from .laola1tv import Laola1TvIE
+from .lecture2go import Lecture2GoIE
 from .letv import (
     LetvIE,
     LetvTvIE,
@@ -320,6 +341,7 @@ from .musicvault import MusicVaultIE
 from .muzu import MuzuTVIE
 from .myspace import MySpaceIE, MySpaceAlbumIE
 from .myspass import MySpassIE
+from .myvi import MyviIE
 from .myvideo import MyVideoIE
 from .myvidster import MyVidsterIE
 from .nationalgeographic import NationalGeographicIE
@@ -339,6 +361,15 @@ from .ndtv import NDTVIE
 from .netzkino import NetzkinoIE
 from .nerdcubed import NerdCubedFeedIE
 from .nerdist import NerdistIE
+from .neteasemusic import (
+    NetEaseMusicIE,
+    NetEaseMusicAlbumIE,
+    NetEaseMusicSingerIE,
+    NetEaseMusicListIE,
+    NetEaseMusicMvIE,
+    NetEaseMusicProgramIE,
+    NetEaseMusicDjRadioIE,
+)
 from .newgrounds import NewgroundsIE
 from .newstube import NewstubeIE
 from .nextmedia import (
@@ -368,7 +399,8 @@ from .npo import (
     NPOLiveIE,
     NPORadioIE,
     NPORadioFragmentIE,
-    TegenlichtVproIE,
+    VPROIE,
+    WNLIE
 )
 from .nrk import (
     NRKIE,
@@ -384,6 +416,7 @@ from .nytimes import (
 from .nuvid import NuvidIE
 from .odnoklassniki import OdnoklassnikiIE
 from .oktoberfesttv import OktoberfestTVIE
+from .onionstudios import OnionStudiosIE
 from .ooyala import (
     OoyalaIE,
     OoyalaExternalIE,
@@ -401,6 +434,7 @@ from .pbs import PBSIE
 from .philharmoniedeparis import PhilharmonieDeParisIE
 from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
+from .pinkbike import PinkbikeIE
 from .planetaplay import PlanetaPlayIE
 from .pladform import PladformIE
 from .played import PlayedIE
@@ -427,6 +461,7 @@ from .qqmusic import (
     QQMusicSingerIE,
     QQMusicAlbumIE,
     QQMusicToplistIE,
+    QQMusicPlaylistIE,
 )
 from .quickvid import QuickVidIE
 from .r7 import R7IE
@@ -436,6 +471,7 @@ from .radiobremen import RadioBremenIE
 from .radiofrance import RadioFranceIE
 from .rai import RaiIE
 from .rbmaradio import RBMARadioIE
+from .rds import RDSIE
 from .redtube import RedTubeIE
 from .restudy import RestudyIE
 from .reverbnation import ReverbNationIE
@@ -487,6 +523,10 @@ from .smotri import (
     SmotriUserIE,
     SmotriBroadcastIE,
 )
+from .snagfilms import (
+    SnagFilmsIE,
+    SnagFilmsEmbedIE,
+)
 from .snotr import SnotrIE
 from .sohu import SohuIE
 from .soompi import (
@@ -560,6 +600,7 @@ from .tf1 import TF1IE
 from .theonion import TheOnionIE
 from .theplatform import ThePlatformIE
 from .thesixtyone import TheSixtyOneIE
+from .thisamericanlife import ThisAmericanLifeIE
 from .thisav import ThisAVIE
 from .tinypic import TinyPicIE
 from .tlc import TlcIE, TlcDeIE
@@ -567,7 +608,11 @@ from .tmz import (
     TMZIE,
     TMZArticleIE,
 )
-from .tnaflix import TNAFlixIE
+from .tnaflix import (
+    TNAFlixIE,
+    EMPFlixIE,
+    MovieFapIE,
+)
 from .thvideo import (
     THVideoIE,
     THVideoPlaylistIE
@@ -611,6 +656,7 @@ from .twitch import (
     TwitchBookmarksIE,
     TwitchStreamIE,
 )
+from .twitter import TwitterCardIE
 from .ubu import UbuIE
 from .udemy import (
     UdemyIE,
@@ -687,7 +733,10 @@ from .wdr import (
     WDRMobileIE,
     WDRMausIE,
 )
-from .webofstories import WebOfStoriesIE
+from .webofstories import (
+    WebOfStoriesIE,
+    WebOfStoriesPlaylistIE,
+)
 from .weibo import WeiboIE
 from .wimp import WimpIE
 from .wistia import WistiaIE
@@ -696,7 +745,10 @@ from .wrzuta import WrzutaIE
 from .wsj import WSJIE
 from .xbef import XBefIE
 from .xboxclips import XboxClipsIE
-from .xhamster import XHamsterIE
+from .xhamster import (
+    XHamsterIE,
+    XHamsterEmbedIE,
+)
 from .xminus import XMinusIE
 from .xnxx import XNXXIE
 from .xstream import XstreamIE
@@ -715,6 +767,7 @@ from .yandexmusic import (
     YandexMusicPlaylistIE,
 )
 from .yesjapan import YesJapanIE
+from .yinyuetai import YinYueTaiIE
 from .ynet import YnetIE
 from .youjizz import YouJizzIE
 from .youku import YoukuIE

youtube_dl/extractor/adobetv.py

@@ -5,6 +5,8 @@ from ..utils import (
     parse_duration,
     unified_strdate,
     str_to_int,
+    float_or_none,
+    ISO639Utils,
 )
 
 
@@ -69,3 +71,61 @@ class AdobeTVIE(InfoExtractor):
             'view_count': view_count,
             'formats': formats,
         }
+
+
+class AdobeTVVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://video\.tv\.adobe\.com/v/(?P<id>\d+)'
+
+    _TEST = {
+        # From https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners
+        'url': 'https://video.tv.adobe.com/v/2456/',
+        'md5': '43662b577c018ad707a63766462b1e87',
+        'info_dict': {
+            'id': '2456',
+            'ext': 'mp4',
+            'title': 'New experience with Acrobat DC',
+            'description': 'New experience with Acrobat DC',
+            'duration': 248.667,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        player_params = self._parse_json(self._search_regex(
+            r'var\s+bridge\s*=\s*([^;]+);', webpage, 'player parameters'),
+            video_id)
+
+        formats = [{
+            'url': source['src'],
+            'width': source.get('width'),
+            'height': source.get('height'),
+            'tbr': source.get('bitrate'),
+        } for source in player_params['sources']]
+
+        # For both metadata and downloaded files the duration varies among
+        # formats. I just pick the max one
+        duration = max(filter(None, [
+            float_or_none(source.get('duration'), scale=1000)
+            for source in player_params['sources']]))
+
+        subtitles = {}
+        for translation in player_params.get('translations', []):
+            lang_id = translation.get('language_w3c') or ISO639Utils.long2short(translation['language_medium'])
+            if lang_id not in subtitles:
+                subtitles[lang_id] = []
+            subtitles[lang_id].append({
+                'url': translation['vttPath'],
+                'ext': 'vtt',
+            })
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': player_params['title'],
+            'description': self._og_search_description(webpage),
+            'duration': duration,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/appleconnect.py

@@ -0,0 +1,50 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    str_to_int,
+    ExtractorError
+)
+
+
+class AppleConnectIE(InfoExtractor):
+    _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)'
+    _TEST = {
+        'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
+        'md5': '10d0f2799111df4cb1c924520ca78f98',
+        'info_dict': {
+            'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
+            'ext': 'm4v',
+            'title': 'Energy',
+            'uploader': 'Drake',
+            'thumbnail': 'http://is5.mzstatic.com/image/thumb/Video5/v4/78/61/c5/7861c5fa-ad6d-294b-1464-cf7605b911d6/source/1920x1080sr.jpg',
+            'upload_date': '20150710',
+            'timestamp': 1436545535,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        try:
+            video_json = self._html_search_regex(
+                r'class="auc-video-data">(\{.*?\})', webpage, 'json')
+        except ExtractorError:
+            raise ExtractorError('This post doesn\'t contain a video', expected=True)
+
+        video_data = self._parse_json(video_json, video_id)
+        timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp'))
+        like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count'))
+
+        return {
+            'id': video_id,
+            'url': video_data['sslSrc'],
+            'title': video_data['title'],
+            'description': video_data['description'],
+            'uploader': video_data['artistName'],
+            'thumbnail': video_data['artworkUrl'],
+            'timestamp': timestamp,
+            'like_count': like_count,
+        }

youtube_dl/extractor/ard.py

@@ -8,6 +8,7 @@ from .generic import GenericIE
 from ..utils import (
     determine_ext,
     ExtractorError,
+    get_element_by_attribute,
     qualities,
     int_or_none,
     parse_duration,
@@ -22,19 +23,125 @@ class ARDMediathekIE(InfoExtractor):
     _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
 
     _TESTS = [{
+        'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
+        'info_dict': {
+            'id': '29582122',
+            'ext': 'mp4',
+            'title': 'Ich liebe das Leben trotzdem',
+            'description': 'md5:45e4c225c72b27993314b31a84a5261c',
+            'duration': 4557,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
+        'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
+        'info_dict': {
+            'id': '29522730',
+            'ext': 'mp4',
+            'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)',
+            'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
+            'duration': 5252,
+        },
+    }, {
+        # audio
+        'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
+        'md5': '219d94d8980b4f538c7fcb0865eb7f2c',
+        'info_dict': {
+            'id': '28488308',
+            'ext': 'mp3',
+            'title': 'Tod eines Fußballers',
+            'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
+            'duration': 3240,
+        },
+    }, {
         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
         'only_matching': True,
-    }, {
-        'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916',
-        'info_dict': {
-            'id': '22490580',
-            'ext': 'mp4',
-            'title': 'Das Wunder von Wolbeck (Video tgl. ab 20 Uhr)',
-            'description': 'Auf einem restaurierten Hof bei Wolbeck wird der Heilpraktiker Raffael Lembeck eines morgens von seiner Frau Stella tot aufgefunden. Das Opfer war offensichtlich in seiner Praxis zu Fall gekommen und ist dann verblutet, erklärt Prof. Boerne am Tatort.',
-        },
-        'skip': 'Blocked outside of Germany',
     }]
 
+    def _extract_media_info(self, media_info_url, webpage, video_id):
+        media_info = self._download_json(
+            media_info_url, video_id, 'Downloading media JSON')
+
+        formats = self._extract_formats(media_info, video_id)
+
+        if not formats:
+            if '"fsk"' in webpage:
+                raise ExtractorError(
+                    'This video is only available after 20:00', expected=True)
+            elif media_info.get('_geoblocked'):
+                raise ExtractorError('This video is not available due to geo restriction', expected=True)
+
+        self._sort_formats(formats)
+
+        duration = int_or_none(media_info.get('_duration'))
+        thumbnail = media_info.get('_previewImage')
+
+        subtitles = {}
+        subtitle_url = media_info.get('_subtitleUrl')
+        if subtitle_url:
+            subtitles['de'] = [{
+                'ext': 'srt',
+                'url': subtitle_url,
+            }]
+
+        return {
+            'id': video_id,
+            'duration': duration,
+            'thumbnail': thumbnail,
+            'formats': formats,
+            'subtitles': subtitles,
+        }
+
+    def _extract_formats(self, media_info, video_id):
+        type_ = media_info.get('_type')
+        media_array = media_info.get('_mediaArray', [])
+        formats = []
+        for num, media in enumerate(media_array):
+            for stream in media.get('_mediaStreamArray', []):
+                stream_urls = stream.get('_stream')
+                if not stream_urls:
+                    continue
+                if not isinstance(stream_urls, list):
+                    stream_urls = [stream_urls]
+                quality = stream.get('_quality')
+                server = stream.get('_server')
+                for stream_url in stream_urls:
+                    ext = determine_ext(stream_url)
+                    if ext == 'f4m':
+                        formats.extend(self._extract_f4m_formats(
+                            stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124',
+                            video_id, preference=-1, f4m_id='hds'))
+                    elif ext == 'm3u8':
+                        formats.extend(self._extract_m3u8_formats(
+                            stream_url, video_id, 'mp4', preference=1, m3u8_id='hls'))
+                    else:
+                        if server and server.startswith('rtmp'):
+                            f = {
+                                'url': server,
+                                'play_path': stream_url,
+                                'format_id': 'a%s-rtmp-%s' % (num, quality),
+                            }
+                        elif stream_url.startswith('http'):
+                            f = {
+                                'url': stream_url,
+                                'format_id': 'a%s-%s-%s' % (num, ext, quality)
+                            }
+                        else:
+                            continue
+                        m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
+                        if m:
+                            f.update({
+                                'width': int(m.group('width')),
+                                'height': int(m.group('height')),
+                            })
+                        if type_ == 'audio':
+                            f['vcodec'] = 'none'
+                        formats.append(f)
+        return formats
+
     def _real_extract(self, url):
         # determine video id from url
         m = re.match(self._VALID_URL, url)
@@ -92,46 +199,22 @@ class ARDMediathekIE(InfoExtractor):
                     'format_id': fid,
                     'url': furl,
                 })
+            self._sort_formats(formats)
+            info = {
+                'formats': formats,
+            }
         else:  # request JSON file
-            media_info = self._download_json(
-                'http://www.ardmediathek.de/play/media/%s' % video_id, video_id)
-            # The second element of the _mediaArray contains the standard http urls
-            streams = media_info['_mediaArray'][1]['_mediaStreamArray']
-            if not streams:
-                if '"fsk"' in webpage:
-                    raise ExtractorError('This video is only available after 20:00')
+            info = self._extract_media_info(
+                'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id)
 
-            formats = []
-            for s in streams:
-                if type(s['_stream']) == list:
-                    for index, url in enumerate(s['_stream'][::-1]):
-                        quality = s['_quality'] + index
-                        formats.append({
-                            'quality': quality,
-                            'url': url,
-                            'format_id': '%s-%s' % (determine_ext(url), quality)
-                        })
-                    continue
-
-                format = {
-                    'quality': s['_quality'],
-                    'url': s['_stream'],
-                }
-
-                format['format_id'] = '%s-%s' % (
-                    determine_ext(format['url']), format['quality'])
-
-                formats.append(format)
-
-        self._sort_formats(formats)
-
-        return {
+        info.update({
             'id': video_id,
             'title': title,
             'description': description,
-            'formats': formats,
             'thumbnail': thumbnail,
-        }
+        })
+
+        return info
 
 
 class ARDIE(InfoExtractor):
@@ -189,3 +272,41 @@ class ARDIE(InfoExtractor):
             'upload_date': upload_date,
             'thumbnail': thumbnail,
         }
+
+
+class SportschauIE(ARDMediathekIE):
+    IE_NAME = 'Sportschau'
+    _VALID_URL = r'(?P<baseurl>https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P<id>[^/#?]+))\.html'
+    _TESTS = [{
+        'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html',
+        'info_dict': {
+            'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100',
+            'ext': 'mp4',
+            'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        base_url = mobj.group('baseurl')
+
+        webpage = self._download_webpage(url, video_id)
+        title = get_element_by_attribute('class', 'headline', webpage)
+        description = self._html_search_meta('description', webpage, 'description')
+
+        info = self._extract_media_info(
+            base_url + '-mc_defaultQuality-h.json', webpage, video_id)
+
+        info.update({
+            'title': title,
+            'description': description,
+        })
+
+        return info

youtube_dl/extractor/baidu.py

@@ -8,6 +8,7 @@ from ..compat import compat_urlparse
 
 
 class BaiduVideoIE(InfoExtractor):
+    IE_DESC = '百度视频'
     _VALID_URL = r'http://v\.baidu\.com/(?P<type>[a-z]+)/(?P<id>\d+)\.htm'
     _TESTS = [{
         'url': 'http://v.baidu.com/comic/1069.htm?frp=bdbrand&q=%E4%B8%AD%E5%8D%8E%E5%B0%8F%E5%BD%93%E5%AE%B6',

youtube_dl/extractor/bbc.py

@@ -0,0 +1,722 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+)
+from ..compat import compat_HTTPError
+
+
+class BBCCoUkIE(InfoExtractor):
+    IE_NAME = 'bbc.co.uk'
+    IE_DESC = 'BBC iPlayer'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
+
+    _MEDIASELECTOR_URL = 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'
+
+    _TESTS = [
+        {
+            'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
+            'info_dict': {
+                'id': 'b039d07m',
+                'ext': 'flv',
+                'title': 'Kaleidoscope, Leonard Cohen',
+                'description': 'The Canadian poet and songwriter reflects on his musical career.',
+                'duration': 1740,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Man in Black: Series 3: The Printed Name',
+                'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
+                'duration': 1800,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Episode is no longer available on BBC iPlayer Radio',
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Voice UK: Series 3: Blind Auditions 5',
+                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
+                'duration': 5100,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
+            'info_dict': {
+                'id': 'b03k3pb7',
+                'ext': 'flv',
+                'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
+                'description': '2. Invasion',
+                'duration': 3600,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+        }, {
+            'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
+            'info_dict': {
+                'id': 'b04v209v',
+                'ext': 'flv',
+                'title': 'Pete Tong, The Essential New Tune Special',
+                'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
+                'duration': 10800,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        }, {
+            'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
+            'note': 'Audio',
+            'info_dict': {
+                'id': 'p02frcch',
+                'ext': 'flv',
+                'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
+                'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
+                'duration': 3507,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        }, {
+            'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
+            'note': 'Video',
+            'info_dict': {
+                'id': 'p025c103',
+                'ext': 'flv',
+                'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
+                'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
+                'duration': 226,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
+            'info_dict': {
+                'id': 'p02n76xf',
+                'ext': 'flv',
+                'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
+                'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
+                'duration': 3540,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'geolocation',
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
+            'info_dict': {
+                'id': 'b05zmgw1',
+                'ext': 'flv',
+                'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.',
+                'title': 'Royal Academy Summer Exhibition',
+                'duration': 3540,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'geolocation',
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
+            'only_matching': True,
+        }, {
+            'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
+            'only_matching': True,
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
+            'only_matching': True,
+        }
+    ]
+
+    def _extract_asx_playlist(self, connection, programme_id):
+        asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
+        return [ref.get('href') for ref in asx.findall('./Entry/ref')]
+
+    def _extract_connection(self, connection, programme_id):
+        formats = []
+        protocol = connection.get('protocol')
+        supplier = connection.get('supplier')
+        if protocol == 'http':
+            href = connection.get('href')
+            transfer_format = connection.get('transferFormat')
+            # ASX playlist
+            if supplier == 'asx':
+                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
+                    formats.append({
+                        'url': ref,
+                        'format_id': 'ref%s_%s' % (i, supplier),
+                    })
+            # Skip DASH until supported
+            elif transfer_format == 'dash':
+                pass
+            # Direct link
+            else:
+                formats.append({
+                    'url': href,
+                    'format_id': supplier,
+                })
+        elif protocol == 'rtmp':
+            application = connection.get('application', 'ondemand')
+            auth_string = connection.get('authString')
+            identifier = connection.get('identifier')
+            server = connection.get('server')
+            formats.append({
+                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
+                'play_path': identifier,
+                'app': '%s?%s' % (application, auth_string),
+                'page_url': 'http://www.bbc.co.uk',
+                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
+                'rtmp_live': False,
+                'ext': 'flv',
+                'format_id': supplier,
+            })
+        return formats
+
+    def _extract_items(self, playlist):
+        return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
+
+    def _extract_medias(self, media_selection):
+        error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
+        if error is not None:
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
+        return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
+
+    def _extract_connections(self, media):
+        return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
+
+    def _extract_video(self, media, programme_id):
+        formats = []
+        vbr = int_or_none(media.get('bitrate'))
+        vcodec = media.get('encoding')
+        service = media.get('service')
+        width = int_or_none(media.get('width'))
+        height = int_or_none(media.get('height'))
+        file_size = int_or_none(media.get('media_file_size'))
+        for connection in self._extract_connections(media):
+            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,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _extract_audio(self, media, programme_id):
+        formats = []
+        abr = int_or_none(media.get('bitrate'))
+        acodec = media.get('encoding')
+        service = media.get('service')
+        for connection in self._extract_connections(media):
+            conn_formats = self._extract_connection(connection, programme_id)
+            for format in conn_formats:
+                format.update({
+                    'format_id': '%s_%s' % (service, format['format_id']),
+                    'abr': abr,
+                    'acodec': acodec,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _get_subtitles(self, media, programme_id):
+        subtitles = {}
+        for connection in self._extract_connections(media):
+            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
+            lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
+            subtitles[lang] = [
+                {
+                    'url': connection.get('href'),
+                    'ext': 'ttml',
+                },
+            ]
+        return subtitles
+
+    def _download_media_selector(self, programme_id):
+        return self._download_media_selector_url(
+            self._MEDIASELECTOR_URL % programme_id, programme_id)
+
+    def _download_media_selector_url(self, url, programme_id=None):
+        try:
+            media_selection = self._download_xml(
+                url, programme_id, 'Downloading media selection XML')
+        except ExtractorError as ee:
+            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
+                media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
+            else:
+                raise
+        return self._process_media_selector(media_selection, programme_id)
+
+    def _process_media_selector(self, media_selection, programme_id):
+        formats = []
+        subtitles = None
+
+        for media in self._extract_medias(media_selection):
+            kind = media.get('kind')
+            if kind == 'audio':
+                formats.extend(self._extract_audio(media, programme_id))
+            elif kind == 'video':
+                formats.extend(self._extract_video(media, programme_id))
+            elif kind == 'captions':
+                subtitles = self.extract_subtitles(media, programme_id)
+
+        return formats, subtitles
+
+    def _download_playlist(self, playlist_id):
+        try:
+            playlist = self._download_json(
+                'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
+                playlist_id, 'Downloading playlist JSON')
+
+            version = playlist.get('defaultAvailableVersion')
+            if version:
+                smp_config = version['smpConfig']
+                title = smp_config['title']
+                description = smp_config['summary']
+                for item in smp_config['items']:
+                    kind = item['kind']
+                    if kind != 'programme' and kind != 'radioProgramme':
+                        continue
+                    programme_id = item.get('vpid')
+                    duration = int_or_none(item.get('duration'))
+                    formats, subtitles = self._download_media_selector(programme_id)
+                return programme_id, title, description, duration, formats, subtitles
+        except ExtractorError as ee:
+            if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
+                raise
+
+        # fallback to legacy playlist
+        return self._process_legacy_playlist(playlist_id)
+
+    def _process_legacy_playlist_url(self, url, display_id):
+        playlist = self._download_legacy_playlist_url(url, display_id)
+        return self._extract_from_legacy_playlist(playlist, display_id)
+
+    def _process_legacy_playlist(self, playlist_id):
+        return self._process_legacy_playlist_url(
+            'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
+
+    def _download_legacy_playlist_url(self, url, playlist_id=None):
+        return self._download_xml(
+            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')
+        if no_items is not None:
+            reason = no_items.get('reason')
+            if reason == 'preAvailability':
+                msg = 'Episode %s is not yet available' % playlist_id
+            elif reason == 'postAvailability':
+                msg = 'Episode %s is no longer available' % playlist_id
+            elif reason == 'noMedia':
+                msg = 'Episode %s is not currently available' % playlist_id
+            else:
+                msg = 'Episode %s is not available: %s' % (playlist_id, reason)
+            raise ExtractorError(msg, expected=True)
+
+        for item in self._extract_items(playlist):
+            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
+
+            def get_programme_id(item):
+                def get_from_attributes(item):
+                    for p in('identifier', 'group'):
+                        value = item.get(p)
+                        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')
+                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)
+
+        return programme_id, title, description, duration, formats, subtitles
+
+    def _real_extract(self, url):
+        group_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, group_id, 'Downloading video page')
+
+        programme_id = None
+
+        tviplayer = self._search_regex(
+            r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
+            webpage, 'player', default=None)
+
+        if tviplayer:
+            player = self._parse_json(tviplayer, group_id).get('player', {})
+            duration = int_or_none(player.get('duration'))
+            programme_id = player.get('vpid')
+
+        if not programme_id:
+            programme_id = self._search_regex(
+                r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
+
+        if programme_id:
+            formats, subtitles = self._download_media_selector(programme_id)
+            title = self._og_search_title(webpage)
+            description = self._search_regex(
+                r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
+                webpage, 'description', fatal=False)
+        else:
+            programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
+
+        self._sort_formats(formats)
+
+        return {
+            'id': programme_id,
+            'title': title,
+            'description': description,
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
+            'duration': duration,
+            'formats': formats,
+            'subtitles': subtitles,
+        }
+
+
+class BBCIE(BBCCoUkIE):
+    IE_NAME = 'bbc'
+    IE_DESC = 'BBC'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
+
+    # fails with notukerror for some videos
+    # _MEDIASELECTOR_URL = 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s'
+    _MEDIASELECTOR_URL = 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s'
+
+    _TESTS = [{
+        # article with multiple videos embedded with data-media-meta containing
+        # playlist.sxml, externalId and no direct video links
+        'url': 'http://www.bbc.com/news/world-europe-32668511',
+        'info_dict': {
+            'id': 'world-europe-32668511',
+            'title': 'Russia stages massive WW2 parade despite Western boycott',
+            'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
+        },
+        'playlist_count': 2,
+    }, {
+        # article with multiple videos embedded with data-media-meta (more videos)
+        'url': 'http://www.bbc.com/news/business-28299555',
+        'info_dict': {
+            'id': 'business-28299555',
+            'title': 'Farnborough Airshow: Video highlights',
+            'description': 'BBC reports and video highlights at the Farnborough Airshow.',
+        },
+        'playlist_count': 9,
+        'skip': 'Save time',
+    }, {
+        # article with multiple videos embedded with `new SMP()`
+        'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
+        'info_dict': {
+            'id': '3662a707-0af9-3149-963f-47bea720b460',
+            'title': 'BBC Blogs - Adam Curtis - BUGGER',
+        },
+        'playlist_count': 18,
+    }, {
+        # single video embedded with mediaAssetPage.init()
+        'url': 'http://www.bbc.com/news/world-europe-32041533',
+        'info_dict': {
+            'id': 'p02mprgb',
+            'ext': 'flv',
+            'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
+            'duration': 47,
+            'timestamp': 1427219242,
+            'upload_date': '20150324',
+        },
+        'params': {
+            # rtmp download
+            '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)
+        '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',
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        # single video embedded with mediaAssetPage.init() (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',
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        # single video story with digitalData
+        'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
+        'info_dict': {
+            'id': 'p02q6gc4',
+            'ext': 'flv',
+            'title': 'Sri Lanka’s spicy secret',
+            'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
+            'timestamp': 1437674293,
+            'upload_date': '20150723',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }, {
+        # single video story without digitalData
+        'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
+        'info_dict': {
+            'id': 'p018zqqg',
+            'ext': 'flv',
+            'title': 'Hyundai Santa Fe Sport: Rock star',
+            'description': 'md5:b042a26142c4154a6e472933cf20793d',
+            'timestamp': 1368473503,
+            'upload_date': '20130513',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }, {
+        # single video with playlist.sxml URL
+        'url': 'http://www.bbc.com/sport/0/football/33653409',
+        'info_dict': {
+            'id': 'p02xycnp',
+            'ext': 'flv',
+            'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
+            'description': 'md5:398fca0e2e701c609d726e034fa1fc89',
+            'duration': 140,
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }, {
+        # single video with playlist URL from weather section
+        'url': 'http://www.bbc.com/weather/features/33601775',
+        'only_matching': True,
+    }, {
+        # custom redirection to www.bbc.com
+        'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if BBCCoUkIE.suitable(url) else super(BBCIE, cls).suitable(url)
+
+    def _extract_from_media_meta(self, media_meta, video_id):
+        # Direct links to media in media metadata (e.g.
+        # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
+        # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
+        source_files = media_meta.get('sourceFiles')
+        if source_files:
+            return [{
+                'url': f['url'],
+                'format_id': format_id,
+                'ext': f.get('encoding'),
+                'tbr': float_or_none(f.get('bitrate'), 1000),
+                'filesize': int_or_none(f.get('filesize')),
+            } for format_id, f in source_files.items() if f.get('url')], []
+
+        programme_id = media_meta.get('externalId')
+        if programme_id:
+            return self._download_media_selector(programme_id)
+
+        # Process playlist.sxml as legacy playlist
+        href = media_meta.get('href')
+        if href:
+            playlist = self._download_legacy_playlist_url(href)
+            _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
+            return formats, subtitles
+
+        return [], []
+
+    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))
+
+        # 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,
+            }
+
+        # 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)
+            # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
+            digital_data = self._parse_json(
+                self._search_regex(
+                    r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
+                programme_id, fatal=False)
+            page_info = digital_data.get('page', {}).get('pageInfo', {})
+            title = page_info.get('pageName') or self._og_search_title(webpage)
+            description = page_info.get('description') or self._og_search_description(webpage)
+            timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
+            return {
+                'id': programme_id,
+                'title': title,
+                'description': description,
+                'timestamp': timestamp,
+                'formats': formats,
+                'subtitles': subtitles,
+            }
+
+        playlist_title = self._html_search_regex(
+            r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title')
+        playlist_description = self._og_search_description(webpage, default=None)
+
+        def extract_all(pattern):
+            return list(filter(None, map(
+                lambda s: self._parse_json(s, playlist_id, fatal=False),
+                re.findall(pattern, webpage))))
+
+        # Multiple video article (e.g.
+        # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
+        EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+[\da-z]{8}(?:\b[^"]+)?'
+        entries = []
+        for match in extract_all(r'new\s+SMP\(({.+?})\)'):
+            embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
+            if embed_url and re.match(EMBED_URL, embed_url):
+                entries.append(embed_url)
+        entries.extend(re.findall(
+            r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
+        if entries:
+            return self.playlist_result(
+                [self.url_result(entry, 'BBCCoUk') for entry in entries],
+                playlist_id, playlist_title, playlist_description)
+
+        # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
+        medias = extract_all(r"data-media-meta='({[^']+})'")
+
+        if not medias:
+            # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
+            media_asset_page = self._parse_json(
+                self._search_regex(
+                    r'mediaAssetPage\.init\(\s*({.+?}), "/', webpage, 'media asset'),
+                playlist_id)
+            medias = []
+            for video in media_asset_page.get('videos', {}).values():
+                medias.extend(video.values())
+
+        entries = []
+        for num, media_meta in enumerate(medias, start=1):
+            formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
+            if not formats:
+                continue
+            self._sort_formats(formats)
+
+            video_id = media_meta.get('externalId')
+            if not video_id:
+                video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
+
+            title = media_meta.get('caption')
+            if not title:
+                title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
+
+            duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
+
+            images = []
+            for image in media_meta.get('images', {}).values():
+                images.extend(image.values())
+            if 'image' in media_meta:
+                images.append(media_meta['image'])
+
+            thumbnails = [{
+                'url': image.get('href'),
+                'width': int_or_none(image.get('width')),
+                'height': int_or_none(image.get('height')),
+            } for image in images]
+
+            entries.append({
+                'id': video_id,
+                'title': title,
+                'thumbnails': thumbnails,
+                'duration': duration,
+                'timestamp': timestamp,
+                'formats': formats,
+                'subtitles': subtitles,
+            })
+
+        return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)

youtube_dl/extractor/bbccouk.py

@@ -1,380 +0,0 @@
-from __future__ import unicode_literals
-
-import xml.etree.ElementTree
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    int_or_none,
-)
-from ..compat import compat_HTTPError
-
-
-class BBCCoUkIE(InfoExtractor):
-    IE_NAME = 'bbc.co.uk'
-    IE_DESC = 'BBC iPlayer'
-    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
-
-    _TESTS = [
-        {
-            'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
-            'info_dict': {
-                'id': 'b039d07m',
-                'ext': 'flv',
-                'title': 'Kaleidoscope, Leonard Cohen',
-                'description': 'The Canadian poet and songwriter reflects on his musical career.',
-                'duration': 1740,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            }
-        },
-        {
-            'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
-            'info_dict': {
-                'id': 'b00yng1d',
-                'ext': 'flv',
-                'title': 'The Man in Black: Series 3: The Printed Name',
-                'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
-                'duration': 1800,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
-            'skip': 'Episode is no longer available on BBC iPlayer Radio',
-        },
-        {
-            'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
-            'info_dict': {
-                'id': 'b00yng1d',
-                'ext': 'flv',
-                'title': 'The Voice UK: Series 3: Blind Auditions 5',
-                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
-                'duration': 5100,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
-            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
-        },
-        {
-            'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
-            'info_dict': {
-                'id': 'b03k3pb7',
-                'ext': 'flv',
-                'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
-                'description': '2. Invasion',
-                'duration': 3600,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
-            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
-        }, {
-            'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
-            'info_dict': {
-                'id': 'b04v209v',
-                'ext': 'flv',
-                'title': 'Pete Tong, The Essential New Tune Special',
-                'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
-                'duration': 10800,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            }
-        }, {
-            'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
-            'note': 'Audio',
-            'info_dict': {
-                'id': 'p02frcch',
-                'ext': 'flv',
-                'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
-                'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
-                'duration': 3507,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            }
-        }, {
-            'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
-            'note': 'Video',
-            'info_dict': {
-                'id': 'p025c103',
-                'ext': 'flv',
-                'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
-                'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
-                'duration': 226,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            }
-        }, {
-            'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
-            'info_dict': {
-                'id': 'p02n76xf',
-                'ext': 'flv',
-                'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
-                'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
-                'duration': 3540,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
-            'skip': 'geolocation',
-        }, {
-            'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
-            'only_matching': True,
-        }, {
-            'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
-            'only_matching': True,
-        }, {
-            'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
-            'only_matching': True,
-        }
-    ]
-
-    def _extract_asx_playlist(self, connection, programme_id):
-        asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
-        return [ref.get('href') for ref in asx.findall('./Entry/ref')]
-
-    def _extract_connection(self, connection, programme_id):
-        formats = []
-        protocol = connection.get('protocol')
-        supplier = connection.get('supplier')
-        if protocol == 'http':
-            href = connection.get('href')
-            # ASX playlist
-            if supplier == 'asx':
-                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
-                    formats.append({
-                        'url': ref,
-                        'format_id': 'ref%s_%s' % (i, supplier),
-                    })
-            # Direct link
-            else:
-                formats.append({
-                    'url': href,
-                    'format_id': supplier,
-                })
-        elif protocol == 'rtmp':
-            application = connection.get('application', 'ondemand')
-            auth_string = connection.get('authString')
-            identifier = connection.get('identifier')
-            server = connection.get('server')
-            formats.append({
-                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
-                'play_path': identifier,
-                'app': '%s?%s' % (application, auth_string),
-                'page_url': 'http://www.bbc.co.uk',
-                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
-                'rtmp_live': False,
-                'ext': 'flv',
-                'format_id': supplier,
-            })
-        return formats
-
-    def _extract_items(self, playlist):
-        return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
-
-    def _extract_medias(self, media_selection):
-        error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
-        if error is not None:
-            raise ExtractorError(
-                '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
-        return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
-
-    def _extract_connections(self, media):
-        return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
-
-    def _extract_video(self, media, programme_id):
-        formats = []
-        vbr = int(media.get('bitrate'))
-        vcodec = media.get('encoding')
-        service = media.get('service')
-        width = int(media.get('width'))
-        height = int(media.get('height'))
-        file_size = int(media.get('media_file_size'))
-        for connection in self._extract_connections(media):
-            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,
-                })
-            formats.extend(conn_formats)
-        return formats
-
-    def _extract_audio(self, media, programme_id):
-        formats = []
-        abr = int(media.get('bitrate'))
-        acodec = media.get('encoding')
-        service = media.get('service')
-        for connection in self._extract_connections(media):
-            conn_formats = self._extract_connection(connection, programme_id)
-            for format in conn_formats:
-                format.update({
-                    'format_id': '%s_%s' % (service, format['format_id']),
-                    'abr': abr,
-                    'acodec': acodec,
-                })
-            formats.extend(conn_formats)
-        return formats
-
-    def _get_subtitles(self, media, programme_id):
-        subtitles = {}
-        for connection in self._extract_connections(media):
-            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
-            lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
-            ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
-            srt = ''
-
-            def _extract_text(p):
-                if p.text is not None:
-                    stripped_text = p.text.strip()
-                    if stripped_text:
-                        return stripped_text
-                return ' '.join(span.text.strip() for span in p.findall('{http://www.w3.org/2006/10/ttaf1}span'))
-            for pos, p in enumerate(ps):
-                srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), _extract_text(p))
-            subtitles[lang] = [
-                {
-                    'url': connection.get('href'),
-                    'ext': 'ttml',
-                },
-                {
-                    'data': srt,
-                    'ext': 'srt',
-                },
-            ]
-        return subtitles
-
-    def _download_media_selector(self, programme_id):
-        try:
-            media_selection = self._download_xml(
-                'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
-                programme_id, 'Downloading media selection XML')
-        except ExtractorError as ee:
-            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
-                media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
-            else:
-                raise
-
-        formats = []
-        subtitles = None
-
-        for media in self._extract_medias(media_selection):
-            kind = media.get('kind')
-            if kind == 'audio':
-                formats.extend(self._extract_audio(media, programme_id))
-            elif kind == 'video':
-                formats.extend(self._extract_video(media, programme_id))
-            elif kind == 'captions':
-                subtitles = self.extract_subtitles(media, programme_id)
-
-        return formats, subtitles
-
-    def _download_playlist(self, playlist_id):
-        try:
-            playlist = self._download_json(
-                'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
-                playlist_id, 'Downloading playlist JSON')
-
-            version = playlist.get('defaultAvailableVersion')
-            if version:
-                smp_config = version['smpConfig']
-                title = smp_config['title']
-                description = smp_config['summary']
-                for item in smp_config['items']:
-                    kind = item['kind']
-                    if kind != 'programme' and kind != 'radioProgramme':
-                        continue
-                    programme_id = item.get('vpid')
-                    duration = int(item.get('duration'))
-                    formats, subtitles = self._download_media_selector(programme_id)
-                return programme_id, title, description, duration, formats, subtitles
-        except ExtractorError as ee:
-            if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
-                raise
-
-        # fallback to legacy playlist
-        playlist = self._download_xml(
-            'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
-            playlist_id, 'Downloading legacy playlist XML')
-
-        no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
-        if no_items is not None:
-            reason = no_items.get('reason')
-            if reason == 'preAvailability':
-                msg = 'Episode %s is not yet available' % playlist_id
-            elif reason == 'postAvailability':
-                msg = 'Episode %s is no longer available' % playlist_id
-            elif reason == 'noMedia':
-                msg = 'Episode %s is not currently available' % playlist_id
-            else:
-                msg = 'Episode %s is not available: %s' % (playlist_id, reason)
-            raise ExtractorError(msg, expected=True)
-
-        for item in self._extract_items(playlist):
-            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
-            programme_id = item.get('identifier')
-            duration = int(item.get('duration'))
-            formats, subtitles = self._download_media_selector(programme_id)
-
-        return programme_id, title, description, duration, formats, subtitles
-
-    def _real_extract(self, url):
-        group_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, group_id, 'Downloading video page')
-
-        programme_id = None
-
-        tviplayer = self._search_regex(
-            r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
-            webpage, 'player', default=None)
-
-        if tviplayer:
-            player = self._parse_json(tviplayer, group_id).get('player', {})
-            duration = int_or_none(player.get('duration'))
-            programme_id = player.get('vpid')
-
-        if not programme_id:
-            programme_id = self._search_regex(
-                r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
-
-        if programme_id:
-            formats, subtitles = self._download_media_selector(programme_id)
-            title = self._og_search_title(webpage)
-            description = self._search_regex(
-                r'<p class="medium-description">([^<]+)</p>',
-                webpage, 'description', fatal=False)
-        else:
-            programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
-
-        self._sort_formats(formats)
-
-        return {
-            'id': programme_id,
-            'title': title,
-            'description': description,
-            'thumbnail': self._og_search_thumbnail(webpage, default=None),
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,
-        }

youtube_dl/extractor/bet.py

@@ -1,7 +1,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     xpath_text,
     xpath_with_ns,
@@ -57,7 +57,7 @@ class BetIE(InfoExtractor):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
 
-        media_url = compat_urllib_parse.unquote(self._search_regex(
+        media_url = compat_urllib_parse_unquote(self._search_regex(
             [r'mediaURL\s*:\s*"([^"]+)"', r"var\s+mrssMediaUrl\s*=\s*'([^']+)'"],
             webpage, 'media URL'))
 

youtube_dl/extractor/bilibili.py

@@ -41,8 +41,15 @@ class BiliBiliIE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        if self._search_regex(r'(此视频不存在或被删除)', webpage, 'error message', default=None):
-            raise ExtractorError('The video does not exist or was deleted', expected=True)
+        if '(此视频不存在或被删除)' in webpage:
+            raise ExtractorError(
+                'The video does not exist or was deleted', expected=True)
+
+        if '>你没有权限浏览！ 由于版权相关问题 我们不对您所在的地区提供服务<' in webpage:
+            raise ExtractorError(
+                'The video is not available in your region due to copyright reasons',
+                expected=True)
+
         video_code = self._search_regex(
             r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code')
 

youtube_dl/extractor/bliptv.py

@@ -5,7 +5,6 @@ import re
 from .common import InfoExtractor
 
 from ..compat import (
-    compat_str,
     compat_urllib_request,
     compat_urlparse,
 )
@@ -14,6 +13,8 @@ from ..utils import (
     int_or_none,
     parse_iso8601,
     unescapeHTML,
+    xpath_text,
+    xpath_with_ns,
 )
 
 
@@ -23,10 +24,10 @@ class BlipTVIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
-            'md5': 'c6934ad0b6acf2bd920720ec888eb812',
+            'md5': '80baf1ec5c3d2019037c1c707d676b9f',
             'info_dict': {
                 'id': '5779306',
-                'ext': 'mov',
+                'ext': 'm4v',
                 'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
                 'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
                 'timestamp': 1323138843,
@@ -100,6 +101,20 @@ class BlipTVIE(InfoExtractor):
                 'vcodec': 'none',
             }
         },
+        {
+            # missing duration
+            'url': 'http://blip.tv/rss/flash/6700880',
+            'info_dict': {
+                'id': '6684191',
+                'ext': 'm4v',
+                'title': 'Cowboy Bebop: Gateway Shuffle Review',
+                'description': 'md5:3acc480c0f9ae157f5fe88547ecaf3f8',
+                'timestamp': 1386639757,
+                'upload_date': '20131210',
+                'uploader': 'sfdebris',
+                'uploader_id': '706520',
+            }
+        }
     ]
 
     @staticmethod
@@ -128,35 +143,34 @@ class BlipTVIE(InfoExtractor):
 
         rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
 
-        def blip(s):
-            return '{http://blip.tv/dtd/blip/1.0}%s' % s
-
-        def media(s):
-            return '{http://search.yahoo.com/mrss/}%s' % s
-
-        def itunes(s):
-            return '{http://www.itunes.com/dtds/podcast-1.0.dtd}%s' % s
+        def _x(p):
+            return xpath_with_ns(p, {
+                'blip': 'http://blip.tv/dtd/blip/1.0',
+                'media': 'http://search.yahoo.com/mrss/',
+                'itunes': 'http://www.itunes.com/dtds/podcast-1.0.dtd',
+            })
 
         item = rss.find('channel/item')
 
-        video_id = item.find(blip('item_id')).text
-        title = item.find('./title').text
-        description = clean_html(compat_str(item.find(blip('puredescription')).text))
-        timestamp = parse_iso8601(item.find(blip('datestamp')).text)
-        uploader = item.find(blip('user')).text
-        uploader_id = item.find(blip('userid')).text
-        duration = int(item.find(blip('runtime')).text)
-        media_thumbnail = item.find(media('thumbnail'))
-        thumbnail = media_thumbnail.get('url') if media_thumbnail is not None else item.find(itunes('image')).text
-        categories = [category.text for category in item.findall('category')]
+        video_id = xpath_text(item, _x('blip:item_id'), 'video id') or lookup_id
+        title = xpath_text(item, 'title', 'title', fatal=True)
+        description = clean_html(xpath_text(item, _x('blip:puredescription'), 'description'))
+        timestamp = parse_iso8601(xpath_text(item, _x('blip:datestamp'), 'timestamp'))
+        uploader = xpath_text(item, _x('blip:user'), 'uploader')
+        uploader_id = xpath_text(item, _x('blip:userid'), 'uploader id')
+        duration = int_or_none(xpath_text(item, _x('blip:runtime'), 'duration'))
+        media_thumbnail = item.find(_x('media:thumbnail'))
+        thumbnail = (media_thumbnail.get('url') if media_thumbnail is not None
+                     else xpath_text(item, 'image', 'thumbnail'))
+        categories = [category.text for category in item.findall('category') if category is not None]
 
         formats = []
         subtitles_urls = {}
 
-        media_group = item.find(media('group'))
-        for media_content in media_group.findall(media('content')):
+        media_group = item.find(_x('media:group'))
+        for media_content in media_group.findall(_x('media:content')):
             url = media_content.get('url')
-            role = media_content.get(blip('role'))
+            role = media_content.get(_x('blip:role'))
             msg = self._download_webpage(
                 url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url',
                 video_id, 'Resolving URL for %s' % role)
@@ -175,8 +189,8 @@ class BlipTVIE(InfoExtractor):
                     'url': real_url,
                     'format_id': role,
                     'format_note': media_type,
-                    'vcodec': media_content.get(blip('vcodec')) or 'none',
-                    'acodec': media_content.get(blip('acodec')),
+                    'vcodec': media_content.get(_x('blip:vcodec')) or 'none',
+                    'acodec': media_content.get(_x('blip:acodec')),
                     'filesize': media_content.get('filesize'),
                     'width': int_or_none(media_content.get('width')),
                     'height': int_or_none(media_content.get('height')),

youtube_dl/extractor/brightcove.py

@@ -13,6 +13,7 @@ from ..compat import (
     compat_urllib_parse_urlparse,
     compat_urllib_request,
     compat_urlparse,
+    compat_xml_parse_error,
 )
 from ..utils import (
     determine_ext,
@@ -119,7 +120,7 @@ class BrightcoveIE(InfoExtractor):
 
         try:
             object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
-        except xml.etree.ElementTree.ParseError:
+        except compat_xml_parse_error:
             return
 
         fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')

youtube_dl/extractor/canalplus.py

@@ -106,15 +106,11 @@ class CanalplusIE(InfoExtractor):
                 continue
             format_id = fmt.tag
             if format_id == 'HLS':
-                hls_formats = self._extract_m3u8_formats(format_url, video_id, 'flv')
-                for fmt in hls_formats:
-                    fmt['preference'] = preference(format_id)
-                formats.extend(hls_formats)
+                formats.extend(self._extract_m3u8_formats(
+                    format_url, video_id, 'mp4', preference=preference(format_id)))
             elif format_id == 'HDS':
-                hds_formats = self._extract_f4m_formats(format_url + '?hdcore=2.11.3', video_id)
-                for fmt in hds_formats:
-                    fmt['preference'] = preference(format_id)
-                formats.extend(hds_formats)
+                formats.extend(self._extract_f4m_formats(
+                    format_url + '?hdcore=2.11.3', video_id, preference=preference(format_id)))
             else:
                 formats.append({
                     'url': format_url,

youtube_dl/extractor/ceskatelevize.py

@@ -7,6 +7,7 @@ from .common import InfoExtractor
 from ..compat import (
     compat_urllib_request,
     compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urllib_parse_urlparse,
 )
 from ..utils import (
@@ -88,7 +89,7 @@ class CeskaTelevizeIE(InfoExtractor):
         if playlist_url == 'error_region':
             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 
-        req = compat_urllib_request.Request(compat_urllib_parse.unquote(playlist_url))
+        req = compat_urllib_request.Request(compat_urllib_parse_unquote(playlist_url))
         req.add_header('Referer', url)
 
         playlist = self._download_json(req, video_id)

youtube_dl/extractor/clipsyndicate.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     find_xpath_attr,
@@ -10,9 +8,9 @@ from ..utils import (
 
 
 class ClipsyndicateIE(InfoExtractor):
-    _VALID_URL = r'http://www\.clipsyndicate\.com/video/play(list/\d+)?/(?P<id>\d+)'
+    _VALID_URL = r'http://(?:chic|www)\.clipsyndicate\.com/video/play(list/\d+)?/(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.clipsyndicate.com/video/play/4629301/brick_briscoe',
         'md5': '4d7d549451bad625e0ff3d7bd56d776c',
         'info_dict': {
@@ -22,11 +20,13 @@ class ClipsyndicateIE(InfoExtractor):
             'duration': 612,
             'thumbnail': 're:^https?://.+\.jpg',
         },
-    }
+    }, {
+        'url': 'http://chic.clipsyndicate.com/video/play/5844117/shark_attack',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
         js_player = self._download_webpage(
             'http://eplayer.clipsyndicate.com/embed/player.js?va_id=%s' % video_id,
             video_id, 'Downlaoding player')

youtube_dl/extractor/comcarcoff.py

@@ -36,7 +36,7 @@ class ComCarCoffIE(InfoExtractor):
             webpage, 'full data json'))
 
         video_id = full_data['activeVideo']['video']
-        video_data = full_data['videos'][video_id]
+        video_data = full_data.get('videos', {}).get(video_id) or full_data['singleshots'][video_id]
         thumbnails = [{
             'url': video_data['images']['thumb'],
         }, {

youtube_dl/extractor/common.py

@@ -22,18 +22,20 @@ from ..compat import (
     compat_str,
 )
 from ..utils import (
+    NO_DEFAULT,
     age_restricted,
     bug_reports_message,
     clean_html,
     compiled_regex_type,
+    determine_ext,
     ExtractorError,
+    fix_xml_ampersands,
     float_or_none,
     int_or_none,
     RegexNotFoundError,
     sanitize_filename,
     unescapeHTML,
 )
-_NO_DEFAULT = object()
 
 
 class InfoExtractor(object):
@@ -63,7 +65,7 @@ class InfoExtractor(object):
 
                     Potential fields:
                     * url        Mandatory. The URL of the video file
-                    * ext        Will be calculated from url if missing
+                    * ext        Will be calculated from URL if missing
                     * format     A human-readable description of the format
                                  ("mp4 container with h264/opus").
                                  Calculated from the format_id, width, height.
@@ -153,7 +155,7 @@ class InfoExtractor(object):
                     lower to higher preference, each element is a dictionary
                     with the "ext" entry and one of:
                         * "data": The subtitles file contents
-                        * "url": A url pointing to the subtitles file
+                        * "url": A URL pointing to the subtitles file
     automatic_captions: Like 'subtitles', used by the YoutubeIE for
                     automatically generated captions
     duration:       Length of the video in seconds, as an integer.
@@ -174,13 +176,17 @@ class InfoExtractor(object):
                                      Set to "root" to indicate that this is a
                                      comment to the original video.
     age_limit:      Age restriction for the video, as an integer (years)
-    webpage_url:    The url to the video webpage, if given to youtube-dl it
+    webpage_url:    The URL to the video webpage, if given to youtube-dl it
                     should allow to get the same result again. (It will be set
                     by YoutubeDL if it's missing)
     categories:     A list of categories that the video falls in, for example
                     ["Sports", "Berlin"]
     is_live:        True, False, or None (=unknown). Whether this video is a
                     live stream that goes on instead of a fixed-length video.
+    start_time:     Time in seconds where the reproduction should start, as
+                    specified in the URL.
+    end_time:       Time in seconds where the reproduction should end, as
+                    specified in the URL.
 
     Unless mentioned otherwise, the fields should be Unicode strings.
 
@@ -499,7 +505,7 @@ class InfoExtractor(object):
     # Methods for following #608
     @staticmethod
     def url_result(url, ie=None, video_id=None, video_title=None):
-        """Returns a url that points to a page that should be processed"""
+        """Returns a URL that points to a page that should be processed"""
         # TODO: ie should be the class used for getting the info
         video_info = {'_type': 'url',
                       'url': url,
@@ -523,7 +529,7 @@ class InfoExtractor(object):
             video_info['description'] = playlist_description
         return video_info
 
-    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
+    def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None):
         """
         Perform a regex search on the given string, using a single or a list of
         patterns returning the first matching group.
@@ -549,7 +555,7 @@ class InfoExtractor(object):
                 return next(g for g in mobj.groups() if g is not None)
             else:
                 return mobj.group(group)
-        elif default is not _NO_DEFAULT:
+        elif default is not NO_DEFAULT:
             return default
         elif fatal:
             raise RegexNotFoundError('Unable to extract %s' % _name)
@@ -557,7 +563,7 @@ class InfoExtractor(object):
             self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message())
             return None
 
-    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
+    def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None):
         """
         Like _search_regex, but strips HTML tags and unescapes entities.
         """
@@ -633,7 +639,7 @@ class InfoExtractor(object):
         return unescapeHTML(escaped)
 
     def _og_search_thumbnail(self, html, **kargs):
-        return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs)
+        return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs)
 
     def _og_search_description(self, html, **kargs):
         return self._og_search_property('description', html, fatal=False, **kargs)
@@ -705,6 +711,25 @@ class InfoExtractor(object):
         return self._html_search_meta('twitter:player', html,
                                       'twitter card player')
 
+    @staticmethod
+    def _hidden_inputs(html):
+        return dict([
+            (input.group('name'), input.group('value')) for input in re.finditer(
+                r'''(?x)
+                    <input\s+
+                        type=(?P<q_hidden>["\'])hidden(?P=q_hidden)\s+
+                        name=(?P<q_name>["\'])(?P<name>.+?)(?P=q_name)\s+
+                        (?:id=(?P<q_id>["\']).+?(?P=q_id)\s+)?
+                        value=(?P<q_value>["\'])(?P<value>.*?)(?P=q_value)
+                ''', html)
+        ])
+
+    def _form_hidden_inputs(self, form_id, html):
+        form = self._search_regex(
+            r'(?s)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id,
+            html, '%s form' % form_id, group='form')
+        return self._hidden_inputs(form)
+
     def _sort_formats(self, formats, field_preference=None):
         if not formats:
             raise ExtractorError('No video formats found')
@@ -815,10 +840,14 @@ class InfoExtractor(object):
         self.to_screen(msg)
         time.sleep(timeout)
 
-    def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None):
+    def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None,
+                             transform_source=lambda s: fix_xml_ampersands(s).strip()):
         manifest = self._download_xml(
             manifest_url, video_id, 'Downloading f4m manifest',
-            'Unable to download f4m manifest')
+            'Unable to download f4m manifest',
+            # Some manifests may be malformed, e.g. prosiebensat1 generated manifests
+            # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244)
+            transform_source=transform_source)
 
         formats = []
         manifest_version = '1.0'
@@ -828,8 +857,19 @@ class InfoExtractor(object):
             media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media')
         for i, media_el in enumerate(media_nodes):
             if manifest_version == '2.0':
-                manifest_url = ('/'.join(manifest_url.split('/')[:-1]) + '/' +
-                                (media_el.attrib.get('href') or media_el.attrib.get('url')))
+                media_url = media_el.attrib.get('href') or media_el.attrib.get('url')
+                if not media_url:
+                    continue
+                manifest_url = (
+                    media_url if media_url.startswith('http://') or media_url.startswith('https://')
+                    else ('/'.join(manifest_url.split('/')[:-1]) + '/' + media_url))
+                # If media_url is itself a f4m manifest do the recursive extraction
+                # since bitrates in parent manifest (this one) and media_url manifest
+                # may differ leading to inability to resolve the format by requested
+                # bitrate in f4m downloader
+                if determine_ext(manifest_url) == 'f4m':
+                    formats.extend(self._extract_f4m_formats(manifest_url, video_id, preference, f4m_id))
+                    continue
             tbr = int_or_none(media_el.attrib.get('bitrate'))
             formats.append({
                 'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])),
@@ -846,7 +886,8 @@ class InfoExtractor(object):
 
     def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
                               entry_protocol='m3u8', preference=None,
-                              m3u8_id=None, note=None, errnote=None):
+                              m3u8_id=None, note=None, errnote=None,
+                              fatal=True):
 
         formats = [{
             'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
@@ -866,7 +907,10 @@ class InfoExtractor(object):
         m3u8_doc = self._download_webpage(
             m3u8_url, video_id,
             note=note or 'Downloading m3u8 information',
-            errnote=errnote or 'Failed to download m3u8 information')
+            errnote=errnote or 'Failed to download m3u8 information',
+            fatal=fatal)
+        if m3u8_doc is False:
+            return m3u8_doc
         last_info = None
         last_media = None
         kv_rex = re.compile(
@@ -956,7 +1000,7 @@ class InfoExtractor(object):
     def _parse_smil_video(self, video, video_id, base, rtmp_count):
         src = video.get('src')
         if not src:
-            return ([], rtmp_count)
+            return [], rtmp_count
         bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
         width = int_or_none(video.get('width'))
         height = int_or_none(video.get('height'))
@@ -969,7 +1013,7 @@ class InfoExtractor(object):
                     proto = 'http'
         ext = video.get('ext')
         if proto == 'm3u8':
-            return (self._extract_m3u8_formats(src, video_id, ext), rtmp_count)
+            return self._extract_m3u8_formats(src, video_id, ext), rtmp_count
         elif proto == 'rtmp':
             rtmp_count += 1
             streamer = video.get('streamer') or base
@@ -1076,7 +1120,7 @@ class InfoExtractor(object):
 class SearchInfoExtractor(InfoExtractor):
     """
     Base class for paged search queries extractors.
-    They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
+    They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query}
     Instances should define _SEARCH_KEY and _MAX_RESULTS.
     """
 

youtube_dl/extractor/crunchyroll.py

@@ -12,6 +12,7 @@ from math import pow, sqrt, floor
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urllib_request,
 )
 from ..utils import (
@@ -27,7 +28,7 @@ from ..aes import (
 
 
 class CrunchyrollIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
     _NETRC_MACHINE = 'crunchyroll'
     _TESTS = [{
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
@@ -45,6 +46,22 @@ class CrunchyrollIE(InfoExtractor):
             # rtmp
             'skip_download': True,
         },
+    }, {
+        'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1',
+        'info_dict': {
+            'id': '589804',
+            'ext': 'flv',
+            'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
+            'description': 'md5:fe2743efedb49d279552926d0bd0cd9e',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Danny Choo Network',
+            'upload_date': '20120213',
+        },
+        'params': {
+            # rtmp
+            'skip_download': True,
+        },
+
     }, {
         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
         'only_matching': True,
@@ -238,7 +255,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             video_upload_date = unified_strdate(video_upload_date)
         video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
 
-        playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
+        playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
         playerdata_req = compat_urllib_request.Request(playerdata_url)
         playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
         playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
@@ -251,16 +268,17 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
             stream_quality, stream_format = self._FORMAT_IDS[fmt]
             video_format = fmt + 'p'
-            streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
-            # urlencode doesn't work!
-            streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality=' + stream_quality + '&media%5Fid=' + stream_id + '&video%5Fformat=' + stream_format
+            streamdata_req = compat_urllib_request.Request(
+                'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
+                % (stream_id, stream_format, stream_quality),
+                compat_urllib_parse.urlencode({'current_page': url}).encode('utf-8'))
             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
-            streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
             streamdata = self._download_xml(
                 streamdata_req, video_id,
                 note='Downloading media info for %s' % video_format)
-            video_url = streamdata.find('./host').text
-            video_play_path = streamdata.find('./file').text
+            stream_info = streamdata.find('./{default}preload/stream_info')
+            video_url = stream_info.find('./host').text
+            video_play_path = stream_info.find('./file').text
             formats.append({
                 'url': video_url,
                 'play_path': video_play_path,

youtube_dl/extractor/ctsnews.py

@@ -6,6 +6,7 @@ from ..utils import parse_iso8601, ExtractorError
 
 
 class CtsNewsIE(InfoExtractor):
+    IE_DESC = '華視新聞'
     # https connection failed (Connection reset)
     _VALID_URL = r'http://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html'
     _TESTS = [{

youtube_dl/extractor/dailymotion.py

@@ -13,8 +13,10 @@ from ..compat import (
 )
 from ..utils import (
     ExtractorError,
+    determine_ext,
     int_or_none,
     orderedSet,
+    parse_iso8601,
     str_to_int,
     unescapeHTML,
 )
@@ -28,10 +30,16 @@ class DailymotionBaseInfoExtractor(InfoExtractor):
         request.add_header('Cookie', 'family_filter=off; ff=off')
         return request
 
+    def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
+        request = self._build_request(url)
+        return self._download_webpage_handle(request, *args, **kwargs)
+
+    def _download_webpage_no_ff(self, url, *args, **kwargs):
+        request = self._build_request(url)
+        return self._download_webpage(request, *args, **kwargs)
+
 
 class DailymotionIE(DailymotionBaseInfoExtractor):
-    """Information Extractor for Dailymotion"""
-
     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
     IE_NAME = 'dailymotion'
 
@@ -50,9 +58,17 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
             'info_dict': {
                 'id': 'x2iuewm',
                 'ext': 'mp4',
-                'uploader': 'IGN',
                 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
+                'description': 'Several come bundled with the Steam Controller.',
+                'thumbnail': 're:^https?:.*\.(?:jpg|png)$',
+                'duration': 74,
+                'timestamp': 1425657362,
                 'upload_date': '20150306',
+                'uploader': 'IGN',
+                'uploader_id': 'xijv66',
+                'age_limit': 0,
+                'view_count': int,
+                'comment_count': int,
             }
         },
         # Vevo video
@@ -86,38 +102,106 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        url = 'https://www.dailymotion.com/video/%s' % video_id
 
-        # Retrieve video webpage to extract further information
-        request = self._build_request(url)
-        webpage = self._download_webpage(request, video_id)
-
-        # Extract URL, uploader and title from webpage
-        self.report_extraction(video_id)
-
-        # It may just embed a vevo video:
-        m_vevo = re.search(
-            r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
-            webpage)
-        if m_vevo is not None:
-            vevo_id = m_vevo.group('id')
-            self.to_screen('Vevo video detected: %s' % vevo_id)
-            return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
+        webpage = self._download_webpage_no_ff(
+            'https://www.dailymotion.com/video/%s' % video_id, video_id)
 
         age_limit = self._rta_search(webpage)
 
-        video_upload_date = None
-        mobj = re.search(r'<meta property="video:release_date" content="([0-9]{4})-([0-9]{2})-([0-9]{2}).+?"/>', webpage)
-        if mobj is not None:
-            video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
+        description = self._og_search_description(webpage) or self._html_search_meta(
+            'description', webpage, 'description')
+
+        view_count = str_to_int(self._search_regex(
+            [r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"',
+             r'video_views_count[^>]+>\s+([\d\.,]+)'],
+            webpage, 'view count', fatal=False))
+        comment_count = int_or_none(self._search_regex(
+            r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
+            webpage, 'comment count', fatal=False))
+
+        player_v5 = self._search_regex(
+            r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
+            webpage, 'player v5', default=None)
+        if player_v5:
+            player = self._parse_json(player_v5, video_id)
+            metadata = player['metadata']
+            formats = []
+            for quality, media_list in metadata['qualities'].items():
+                for media in media_list:
+                    media_url = media.get('url')
+                    if not media_url:
+                        continue
+                    type_ = media.get('type')
+                    if type_ == 'application/vnd.lumberjack.manifest':
+                        continue
+                    if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8':
+                        formats.extend(self._extract_m3u8_formats(
+                            media_url, video_id, 'mp4', m3u8_id='hls'))
+                    else:
+                        f = {
+                            'url': media_url,
+                            'format_id': quality,
+                        }
+                        m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
+                        if m:
+                            f.update({
+                                'width': int(m.group('width')),
+                                'height': int(m.group('height')),
+                            })
+                        formats.append(f)
+            self._sort_formats(formats)
+
+            title = metadata['title']
+            duration = int_or_none(metadata.get('duration'))
+            timestamp = int_or_none(metadata.get('created_time'))
+            thumbnail = metadata.get('poster_url')
+            uploader = metadata.get('owner', {}).get('screenname')
+            uploader_id = metadata.get('owner', {}).get('id')
+
+            subtitles = {}
+            for subtitle_lang, subtitle in metadata.get('subtitles', {}).get('data', {}).items():
+                subtitles[subtitle_lang] = [{
+                    'ext': determine_ext(subtitle_url),
+                    'url': subtitle_url,
+                } for subtitle_url in subtitle.get('urls', [])]
+
+            return {
+                'id': video_id,
+                'title': title,
+                'description': description,
+                'thumbnail': thumbnail,
+                'duration': duration,
+                'timestamp': timestamp,
+                'uploader': uploader,
+                'uploader_id': uploader_id,
+                'age_limit': age_limit,
+                'view_count': view_count,
+                'comment_count': comment_count,
+                'formats': formats,
+                'subtitles': subtitles,
+            }
+
+        # vevo embed
+        vevo_id = self._search_regex(
+            r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
+            webpage, 'vevo embed', default=None)
+        if vevo_id:
+            return self.url_result('vevo:%s' % vevo_id, 'Vevo')
+
+        # fallback old player
+        embed_page = self._download_webpage_no_ff(
+            'https://www.dailymotion.com/embed/video/%s' % video_id,
+            video_id, 'Downloading embed page')
+
+        timestamp = parse_iso8601(self._html_search_meta(
+            'video:release_date', webpage, 'upload date'))
+
+        info = self._parse_json(
+            self._search_regex(
+                r'var info = ({.*?}),$', embed_page,
+                'video info', flags=re.MULTILINE),
+            video_id)
 
-        embed_url = 'https://www.dailymotion.com/embed/video/%s' % video_id
-        embed_request = self._build_request(embed_url)
-        embed_page = self._download_webpage(
-            embed_request, video_id, 'Downloading embed page')
-        info = self._search_regex(r'var info = ({.*?}),$', embed_page,
-                                  'video info', flags=re.MULTILINE)
-        info = json.loads(info)
         if info.get('error') is not None:
             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
             raise ExtractorError(msg, expected=True)
@@ -138,16 +222,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                     'width': width,
                     'height': height,
                 })
-        if not formats:
-            raise ExtractorError('Unable to extract video URL')
+        self._sort_formats(formats)
 
         # subtitles
         video_subtitles = self.extract_subtitles(video_id, webpage)
 
-        view_count = str_to_int(self._search_regex(
-            r'video_views_count[^>]+>\s+([\d\.,]+)',
-            webpage, 'view count', fatal=False))
-
         title = self._og_search_title(webpage, default=None)
         if title is None:
             title = self._html_search_regex(
@@ -158,12 +237,14 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
             'id': video_id,
             'formats': formats,
             'uploader': info['owner.screenname'],
-            'upload_date': video_upload_date,
+            'timestamp': timestamp,
             'title': title,
+            'description': description,
             'subtitles': video_subtitles,
             'thumbnail': info['thumbnail_url'],
             'age_limit': age_limit,
             'view_count': view_count,
+            'duration': info['duration']
         }
 
     def _get_subtitles(self, video_id, webpage):
@@ -198,10 +279,17 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 
     def _extract_entries(self, id):
         video_ids = []
+        processed_urls = set()
         for pagenum in itertools.count(1):
-            request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
-            webpage = self._download_webpage(request,
-                                             id, 'Downloading page %s' % pagenum)
+            page_url = self._PAGE_TEMPLATE % (id, pagenum)
+            webpage, urlh = self._download_webpage_handle_no_ff(
+                page_url, id, 'Downloading page %s' % pagenum)
+            if urlh.geturl() in processed_urls:
+                self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
+                    page_url, urlh.geturl()), id)
+                break
+
+            processed_urls.add(urlh.geturl())
 
             video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
 
@@ -225,7 +313,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 
 class DailymotionUserIE(DailymotionPlaylistIE):
     IE_NAME = 'dailymotion:user'
-    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?:(?:old/)?user/)?(?P<user>[^/]+)$'
+    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
     _TESTS = [{
         'url': 'https://www.dailymotion.com/user/nqtv',
@@ -234,6 +322,17 @@ class DailymotionUserIE(DailymotionPlaylistIE):
             'title': 'Rémi Gaillard',
         },
         'playlist_mincount': 100,
+    }, {
+        'url': 'http://www.dailymotion.com/user/UnderProject',
+        'info_dict': {
+            'id': 'UnderProject',
+            'title': 'UnderProject',
+        },
+        'playlist_mincount': 1800,
+        'expected_warnings': [
+            'Stopped at duplicated page',
+        ],
+        'skip': 'Takes too long time',
     }]
 
     def _real_extract(self, url):
@@ -251,3 +350,52 @@ class DailymotionUserIE(DailymotionPlaylistIE):
             'title': full_user,
             'entries': self._extract_entries(user),
         }
+
+
+class DailymotionCloudIE(DailymotionBaseInfoExtractor):
+    _VALID_URL_PREFIX = r'http://api\.dmcloud\.net/(?:player/)?embed/'
+    _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
+    _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
+
+    _TESTS = [{
+        # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
+        # Tested at FranceTvInfo_2
+        'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
+        'only_matching': True,
+    }, {
+        # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
+        'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def _extract_dmcloud_url(self, webpage):
+        mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage)
+        if mobj:
+            return mobj.group(1)
+
+        mobj = re.search(
+            r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL,
+            webpage)
+        if mobj:
+            return mobj.group(1)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage_no_ff(url, video_id)
+
+        title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
+
+        video_info = self._parse_json(self._search_regex(
+            r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
+
+        # TODO: parse ios_url, which is in fact a manifest
+        video_url = video_info['mp4_url']
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'thumbnail': video_info.get('thumbnail_url'),
+        }

youtube_dl/extractor/dfb.py

@@ -3,42 +3,47 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import unified_strdate
 
 
 class DFBIE(InfoExtractor):
     IE_NAME = 'tv.dfb.de'
-    _VALID_URL = r'https?://tv\.dfb\.de/video/[^/]+/(?P<id>\d+)'
+    _VALID_URL = r'https?://tv\.dfb\.de/video/(?P<display_id>[^/]+)/(?P<id>\d+)'
 
     _TEST = {
-        'url': 'http://tv.dfb.de/video/highlights-des-empfangs-in-berlin/9070/',
+        'url': 'http://tv.dfb.de/video/u-19-em-stimmen-zum-spiel-gegen-russland/11633/',
         # The md5 is different each time
         'info_dict': {
-            'id': '9070',
+            'id': '11633',
+            'display_id': 'u-19-em-stimmen-zum-spiel-gegen-russland',
             'ext': 'flv',
-            'title': 'Highlights des Empfangs in Berlin',
-            'upload_date': '20140716',
+            'title': 'U 19-EM: Stimmen zum Spiel gegen Russland',
+            'upload_date': '20150714',
         },
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, display_id)
         player_info = self._download_xml(
             'http://tv.dfb.de/server/hd_video.php?play=%s' % video_id,
-            video_id)
+            display_id)
         video_info = player_info.find('video')
 
-        f4m_info = self._download_xml(self._proto_relative_url(video_info.find('url').text.strip()), video_id)
+        f4m_info = self._download_xml(
+            self._proto_relative_url(video_info.find('url').text.strip()), display_id)
         token_el = f4m_info.find('token')
         manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0'
+        formats = self._extract_f4m_formats(manifest_url, display_id)
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'title': video_info.find('title').text,
-            'url': manifest_url,
-            'ext': 'flv',
             'thumbnail': self._og_search_thumbnail(webpage),
-            'upload_date': ''.join(video_info.find('time_date').text.split('.')[::-1]),
+            'upload_date': unified_strdate(video_info.find('time_date').text),
+            'formats': formats,
         }

youtube_dl/extractor/douyutv.py

@@ -9,6 +9,7 @@ from ..compat import (compat_str, compat_basestring)
 
 
 class DouyuTVIE(InfoExtractor):
+    IE_DESC = '斗鱼'
     _VALID_URL = r'http://(?:www\.)?douyutv\.com/(?P<id>[A-Za-z0-9]+)'
     _TESTS = [{
         'url': 'http://www.douyutv.com/iseven',

youtube_dl/extractor/dramafever.py

@@ -6,6 +6,8 @@ import itertools
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
+    compat_urllib_parse,
+    compat_urllib_request,
     compat_urlparse,
 )
 from ..utils import (
@@ -17,7 +19,54 @@ from ..utils import (
 )
 
 
-class DramaFeverIE(InfoExtractor):
+class DramaFeverBaseIE(InfoExtractor):
+    _LOGIN_URL = 'https://www.dramafever.com/accounts/login/'
+    _NETRC_MACHINE = 'dramafever'
+
+    _CONSUMER_SECRET = 'DA59dtVXYLxajktV'
+
+    _consumer_secret = None
+
+    def _get_consumer_secret(self):
+        mainjs = self._download_webpage(
+            'http://www.dramafever.com/static/51afe95/df2014/scripts/main.js',
+            None, 'Downloading main.js', fatal=False)
+        if not mainjs:
+            return self._CONSUMER_SECRET
+        return self._search_regex(
+            r"var\s+cs\s*=\s*'([^']+)'", mainjs,
+            'consumer secret', default=self._CONSUMER_SECRET)
+
+    def _real_initialize(self):
+        self._login()
+        self._consumer_secret = self._get_consumer_secret()
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+
+        login_form = {
+            'username': username,
+            'password': password,
+        }
+
+        request = compat_urllib_request.Request(
+            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+        response = self._download_webpage(
+            request, None, 'Logging in as %s' % username)
+
+        if all(logout_pattern not in response
+               for logout_pattern in ['href="/accounts/logout/"', '>Log out<']):
+            error = self._html_search_regex(
+                r'(?s)class="hidden-xs prompt"[^>]*>(.+?)<',
+                response, 'error message', default=None)
+            if error:
+                raise ExtractorError('Unable to login: %s' % error, expected=True)
+            raise ExtractorError('Unable to log in')
+
+
+class DramaFeverIE(DramaFeverBaseIE):
     IE_NAME = 'dramafever'
     _VALID_URL = r'https?://(?:www\.)?dramafever\.com/drama/(?P<id>[0-9]+/[0-9]+)(?:/|$)'
     _TEST = {
@@ -85,6 +134,23 @@ class DramaFeverIE(InfoExtractor):
                 'url': href,
             }]
 
+        series_id, episode_number = video_id.split('.')
+        episode_info = self._download_json(
+            # We only need a single episode info, so restricting page size to one episode
+            # and dealing with page number as with episode number
+            r'http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_number=%s&page_size=1'
+            % (self._consumer_secret, series_id, episode_number),
+            video_id, 'Downloading episode info JSON', fatal=False)
+        if episode_info:
+            value = episode_info.get('value')
+            if value:
+                subfile = value[0].get('subfile') or value[0].get('new_subfile')
+                if subfile and subfile != 'http://www.dramafever.com/st/':
+                    subtitles.setdefault('English', []).append({
+                        'ext': 'srt',
+                        'url': subfile,
+                    })
+
         return {
             'id': video_id,
             'title': title,
@@ -97,7 +163,7 @@ class DramaFeverIE(InfoExtractor):
         }
 
 
-class DramaFeverSeriesIE(InfoExtractor):
+class DramaFeverSeriesIE(DramaFeverBaseIE):
     IE_NAME = 'dramafever:series'
     _VALID_URL = r'https?://(?:www\.)?dramafever\.com/drama/(?P<id>[0-9]+)(?:/(?:(?!\d+(?:/|$)).+)?)?$'
     _TESTS = [{
@@ -118,27 +184,14 @@ class DramaFeverSeriesIE(InfoExtractor):
         'playlist_count': 20,
     }]
 
-    _CONSUMER_SECRET = 'DA59dtVXYLxajktV'
     _PAGE_SIZE = 60  # max is 60 (see http://api.drama9.com/#get--api-4-episode-series-)
 
-    def _get_consumer_secret(self, video_id):
-        mainjs = self._download_webpage(
-            'http://www.dramafever.com/static/51afe95/df2014/scripts/main.js',
-            video_id, 'Downloading main.js', fatal=False)
-        if not mainjs:
-            return self._CONSUMER_SECRET
-        return self._search_regex(
-            r"var\s+cs\s*=\s*'([^']+)'", mainjs,
-            'consumer secret', default=self._CONSUMER_SECRET)
-
     def _real_extract(self, url):
         series_id = self._match_id(url)
 
-        consumer_secret = self._get_consumer_secret(series_id)
-
         series = self._download_json(
             'http://www.dramafever.com/api/4/series/query/?cs=%s&series_id=%s'
-            % (consumer_secret, series_id),
+            % (self._consumer_secret, series_id),
             series_id, 'Downloading series JSON')['series'][series_id]
 
         title = clean_html(series['name'])
@@ -148,11 +201,14 @@ class DramaFeverSeriesIE(InfoExtractor):
         for page_num in itertools.count(1):
             episodes = self._download_json(
                 'http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_size=%d&page_number=%d'
-                % (consumer_secret, series_id, self._PAGE_SIZE, page_num),
+                % (self._consumer_secret, series_id, self._PAGE_SIZE, page_num),
                 series_id, 'Downloading episodes JSON page #%d' % page_num)
             for episode in episodes.get('value', []):
+                episode_url = episode.get('episode_url')
+                if not episode_url:
+                    continue
                 entries.append(self.url_result(
-                    compat_urlparse.urljoin(url, episode['episode_url']),
+                    compat_urlparse.urljoin(url, episode_url),
                     'DramaFever', episode.get('guid')))
             if page_num == episodes['num_pages']:
                 break

youtube_dl/extractor/drbonanza.py

@@ -15,7 +15,6 @@ class DRBonanzaIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.dr.dk/bonanza/serie/portraetter/Talkshowet.htm?assetId=65517',
-        'md5': 'fe330252ddea607635cf2eb2c99a0af3',
         'info_dict': {
             'id': '65517',
             'ext': 'mp4',
@@ -26,6 +25,9 @@ class DRBonanzaIE(InfoExtractor):
             'upload_date': '20110120',
             'duration': 3664,
         },
+        'params': {
+            'skip_download': True,  # requires rtmp
+        },
     }, {
         'url': 'http://www.dr.dk/bonanza/radio/serie/sport/fodbold.htm?assetId=59410',
         'md5': '6dfe039417e76795fb783c52da3de11d',
@@ -93,6 +95,11 @@ class DRBonanzaIE(InfoExtractor):
                         'format_id': file['Type'].replace('Video', ''),
                         'preference': preferencemap.get(file['Type'], -10),
                     })
+                    if format['url'].startswith('rtmp'):
+                        rtmp_url = format['url']
+                        format['rtmp_live'] = True  # --resume does not work
+                        if '/bonanza/' in rtmp_url:
+                            format['play_path'] = rtmp_url.split('/bonanza/')[1]
                     formats.append(format)
                 elif file['Type'] == "Thumb":
                     thumbnail = file['Location']
@@ -111,9 +118,6 @@ class DRBonanzaIE(InfoExtractor):
         description = '%s\n%s\n%s\n' % (
             info['Description'], info['Actors'], info['Colophon'])
 
-        for f in formats:
-            f['url'] = f['url'].replace('rtmp://vod-bonanza.gss.dr.dk/bonanza/', 'http://vodfiles.dr.dk/')
-            f['url'] = f['url'].replace('mp4:bonanza', 'bonanza')
         self._sort_formats(formats)
 
         display_id = re.sub(r'[^\w\d-]', '', re.sub(r' ', '-', title.lower())) + '-' + asset_id

youtube_dl/extractor/drtuber.py

@@ -36,25 +36,24 @@ class DrTuberIE(InfoExtractor):
             r'<source src="([^"]+)"', webpage, 'video URL')
 
         title = self._html_search_regex(
-            [r'class="hd_title" style="[^"]+">([^<]+)</h1>', r'<title>([^<]+) - \d+'],
+            [r'<p[^>]+class="title_substrate">([^<]+)</p>', r'<title>([^<]+) - \d+'],
             webpage, 'title')
 
         thumbnail = self._html_search_regex(
             r'poster="([^"]+)"',
             webpage, 'thumbnail', fatal=False)
 
-        like_count = str_to_int(self._html_search_regex(
-            r'<span id="rate_likes">\s*<img[^>]+>\s*<span>([\d,\.]+)</span>',
-            webpage, 'like count', fatal=False))
-        dislike_count = str_to_int(self._html_search_regex(
-            r'<span id="rate_dislikes">\s*<img[^>]+>\s*<span>([\d,\.]+)</span>',
-            webpage, 'like count', fatal=False))
-        comment_count = str_to_int(self._html_search_regex(
-            r'<span class="comments_count">([\d,\.]+)</span>',
-            webpage, 'comment count', fatal=False))
+        def extract_count(id_, name):
+            return str_to_int(self._html_search_regex(
+                r'<span[^>]+(?:class|id)="%s"[^>]*>([\d,\.]+)</span>' % id_,
+                webpage, '%s count' % name, fatal=False))
+
+        like_count = extract_count('rate_likes', 'like')
+        dislike_count = extract_count('rate_dislikes', 'dislike')
+        comment_count = extract_count('comments_count', 'comment')
 
         cats_str = self._search_regex(
-            r'<span>Categories:</span><div>(.+?)</div>', webpage, 'categories', fatal=False)
+            r'<div[^>]+class="categories_list">(.+?)</div>', webpage, 'categories', fatal=False)
         categories = [] if not cats_str else re.findall(r'<a title="([^"]+)"', cats_str)
 
         return {

youtube_dl/extractor/ehow.py

@@ -1,9 +1,7 @@
 from __future__ import unicode_literals
 
-from ..compat import (
-    compat_urllib_parse,
-)
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
 
 
 class EHowIE(InfoExtractor):
@@ -26,7 +24,7 @@ class EHowIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         video_url = self._search_regex(
             r'(?:file|source)=(http[^\'"&]*)', webpage, 'video URL')
-        final_url = compat_urllib_parse.unquote(video_url)
+        final_url = compat_urllib_parse_unquote(video_url)
         uploader = self._html_search_meta('uploader', webpage)
         title = self._og_search_title(webpage).replace(' | eHow', '')
 

youtube_dl/extractor/empflix.py

@@ -1,31 +0,0 @@
-from __future__ import unicode_literals
-
-from .tnaflix import TNAFlixIE
-
-
-class EMPFlixIE(TNAFlixIE):
-    _VALID_URL = r'https?://(?:www\.)?empflix\.com/videos/(?P<display_id>.+?)-(?P<id>[0-9]+)\.html'
-
-    _TITLE_REGEX = r'name="title" value="(?P<title>[^"]*)"'
-    _DESCRIPTION_REGEX = r'name="description" value="([^"]*)"'
-    _CONFIG_REGEX = r'flashvars\.config\s*=\s*escape\("([^"]+)"'
-
-    _TESTS = [
-        {
-            'url': 'http://www.empflix.com/videos/Amateur-Finger-Fuck-33051.html',
-            'md5': 'b1bc15b6412d33902d6e5952035fcabc',
-            'info_dict': {
-                'id': '33051',
-                'display_id': 'Amateur-Finger-Fuck',
-                'ext': 'mp4',
-                'title': 'Amateur Finger Fuck',
-                'description': 'Amateur solo finger fucking.',
-                'thumbnail': 're:https?://.*\.jpg$',
-                'age_limit': 18,
-            }
-        },
-        {
-            'url': 'http://www.empflix.com/videos/[AROMA][ARMD-718]-Aoi-Yoshino-Sawa-25826.html',
-            'only_matching': True,
-        }
-    ]

youtube_dl/extractor/facebook.py

@@ -9,7 +9,7 @@ from ..compat import (
     compat_http_client,
     compat_str,
     compat_urllib_error,
-    compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urllib_request,
 )
 from ..utils import (
@@ -136,7 +136,7 @@ class FacebookIE(InfoExtractor):
             else:
                 raise ExtractorError('Cannot parse data')
         data = dict(json.loads(m.group(1)))
-        params_raw = compat_urllib_parse.unquote(data['params'])
+        params_raw = compat_urllib_parse_unquote(data['params'])
         params = json.loads(params_raw)
         video_data = params['video_data'][0]
 

youtube_dl/extractor/faz.py

@@ -6,9 +6,9 @@ from .common import InfoExtractor
 
 class FazIE(InfoExtractor):
     IE_NAME = 'faz.net'
-    _VALID_URL = r'https?://www\.faz\.net/multimedia/videos/.*?-(?P<id>\d+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?faz\.net/(?:[^/]+/)*.*?-(?P<id>\d+)\.html'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.faz.net/multimedia/videos/stockholm-chemie-nobelpreis-fuer-drei-amerikanische-forscher-12610585.html',
         'info_dict': {
             'id': '12610585',
@@ -16,7 +16,22 @@ class FazIE(InfoExtractor):
             'title': 'Stockholm: Chemie-Nobelpreis für drei amerikanische Forscher',
             'description': 'md5:1453fbf9a0d041d985a47306192ea253',
         },
-    }
+    }, {
+        'url': 'http://www.faz.net/aktuell/politik/berlin-gabriel-besteht-zerreissprobe-ueber-datenspeicherung-13659345.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.faz.net/berlin-gabriel-besteht-zerreissprobe-ueber-datenspeicherung-13659345.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.faz.net/-13659345.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.faz.net/aktuell/politik/-13659345.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.faz.net/foobarblafasel-13659345.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/francetv.py

@@ -6,18 +6,15 @@ import re
 import json
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse_urlparse,
-    compat_urlparse,
-)
+from ..compat import compat_urlparse
 from ..utils import (
     clean_html,
     ExtractorError,
     int_or_none,
-    float_or_none,
     parse_duration,
     determine_ext,
 )
+from .dailymotion import DailymotionCloudIE
 
 
 class FranceTVBaseInfoExtractor(InfoExtractor):
@@ -58,12 +55,12 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
                     # See https://github.com/rg3/youtube-dl/issues/3963
                     # m3u8 urls work fine
                     continue
-                video_url_parsed = compat_urllib_parse_urlparse(video_url)
                 f4m_url = self._download_webpage(
-                    'http://hdfauth.francetv.fr/esi/urltokengen2.html?url=%s' % video_url_parsed.path,
+                    'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url,
                     video_id, 'Downloading f4m manifest token', fatal=False)
                 if f4m_url:
-                    formats.extend(self._extract_f4m_formats(f4m_url, video_id, 1, format_id))
+                    formats.extend(self._extract_f4m_formats(
+                        f4m_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, 1, format_id))
             elif ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4', m3u8_id=format_id))
             elif video_url.startswith('rtmp'):
@@ -86,7 +83,7 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
             'title': info['titre'],
             'description': clean_html(info['synopsis']),
             'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
-            'duration': float_or_none(info.get('real_duration'), 1000) or parse_duration(info['duree']),
+            'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
             'timestamp': int_or_none(info['diffusion']['timestamp']),
             'formats': formats,
         }
@@ -131,12 +128,26 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
             'skip_download': 'HLS (reqires ffmpeg)'
         },
         'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.',
+    }, {
+        'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
+        'md5': 'f485bda6e185e7d15dbc69b72bae993e',
+        'info_dict': {
+            'id': '556e03339473995ee145930c',
+            'ext': 'mp4',
+            'title': 'Les entreprises familiales : le secret de la réussite',
+            'thumbnail': 're:^https?://.*\.jpe?g$',
+        }
     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         page_title = mobj.group('title')
         webpage = self._download_webpage(url, page_title)
+
+        dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
+        if dmcloud_url:
+            return self.url_result(dmcloud_url, 'DailymotionCloud')
+
         video_id, catalogue = self._search_regex(
             r'id-video=([^@]+@[^"]+)', webpage, 'video id').split('@')
         return self._extract_video(video_id, catalogue)
@@ -145,11 +156,21 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
 class FranceTVIE(FranceTVBaseInfoExtractor):
     IE_NAME = 'francetv'
     IE_DESC = 'France 2, 3, 4, 5 and Ô'
-    _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/
-        (?:
-            emissions/.*?/(videos|emissions)/(?P<id>[^/?]+)
-        |   (emissions?|jt)/(?P<key>[^/?]+)
-        )'''
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            (?:www\.)?france[2345o]\.fr/
+                                (?:
+                                    emissions/[^/]+/(?:videos|diffusions)|
+                                    emission/[^/]+|
+                                    videos|
+                                    jt
+                                )
+                            /|
+                            embed\.francetv\.fr/\?ue=
+                        )
+                        (?P<id>[^/?]+)
+                    '''
 
     _TESTS = [
         # france2
@@ -206,24 +227,46 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
         },
         # franceo
         {
-            'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013',
-            'md5': '52f0bfe202848b15915a2f39aaa8981b',
+            'url': 'http://www.franceo.fr/jt/info-soir/18-07-2015',
+            'md5': '47d5816d3b24351cdce512ad7ab31da8',
             'info_dict': {
-                'id': '108634970',
+                'id': '125377621',
                 'ext': 'flv',
-                'title': 'Infô Afrique',
-                'description': 'md5:ebf346da789428841bee0fd2a935ea55',
-                'upload_date': '20140915',
-                'timestamp': 1410822000,
+                'title': 'Infô soir',
+                'description': 'md5:01b8c6915a3d93d8bbbd692651714309',
+                'upload_date': '20150718',
+                'timestamp': 1437241200,
+                'duration': 414,
             },
         },
+        {
+            # francetv embed
+            'url': 'http://embed.francetv.fr/?ue=8d7d3da1e3047c42ade5a5d7dfd3fc87',
+            'info_dict': {
+                'id': 'EV_30231',
+                'ext': 'flv',
+                'title': 'Alcaline, le concert avec Calogero',
+                'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
+                'upload_date': '20150226',
+                'timestamp': 1424989860,
+                'duration': 5400,
+            },
+        },
+        {
+            'url': 'http://www.france4.fr/emission/highlander/diffusion-du-17-07-2015-04h05',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://www.franceo.fr/videos/125377617',
+            'only_matching': True,
+        }
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        webpage = self._download_webpage(url, mobj.group('key') or mobj.group('id'))
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
         video_id, catalogue = self._html_search_regex(
-            r'href="http://videos\.francetv\.fr/video/([^@]+@[^"]+)"',
+            r'href="http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
             webpage, 'video ID').split('@')
         return self._extract_video(video_id, catalogue)
 

youtube_dl/extractor/gamespot.py

@@ -5,7 +5,7 @@ import json
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urlparse,
 )
 from ..utils import (
@@ -75,7 +75,7 @@ class GameSpotIE(InfoExtractor):
         return {
             'id': data_video['guid'],
             'display_id': page_id,
-            'title': compat_urllib_parse.unquote(data_video['title']),
+            'title': compat_urllib_parse_unquote(data_video['title']),
             'formats': formats,
             'description': self._html_search_meta('description', webpage),
             'thumbnail': self._og_search_thumbnail(webpage),

youtube_dl/extractor/generic.py

@@ -8,7 +8,6 @@ import re
 from .common import InfoExtractor
 from .youtube import YoutubeIE
 from ..compat import (
-    compat_urllib_parse,
     compat_urllib_parse_unquote,
     compat_urllib_request,
     compat_urlparse,
@@ -37,12 +36,18 @@ from .rutv import RUTVIE
 from .tvc import TVCIE
 from .sportbox import SportBoxEmbedIE
 from .smotri import SmotriIE
+from .myvi import MyviIE
 from .condenast import CondeNastIE
 from .udn import UDNEmbedIE
 from .senateisvp import SenateISVPIE
 from .bliptv import BlipTVIE
 from .svt import SVTIE
 from .pornhub import PornHubIE
+from .xhamster import XHamsterEmbedIE
+from .vimeo import VimeoIE
+from .dailymotion import DailymotionCloudIE
+from .onionstudios import OnionStudiosIE
+from .snagfilms import SnagFilmsEmbedIE
 
 
 class GenericIE(InfoExtractor):
@@ -271,14 +276,6 @@ class GenericIE(InfoExtractor):
                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
             },
         },
-        # BBC iPlayer embeds
-        {
-            'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
-            'info_dict': {
-                'title': 'BBC - Blogs -  Adam Curtis - BUGGER',
-            },
-            'playlist_mincount': 18,
-        },
         # RUTV embed
         {
             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
@@ -333,6 +330,26 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        # Myvi.ru embed
+        {
+            'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
+            'info_dict': {
+                'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
+                'ext': 'mp4',
+                'title': 'Ужастики, русский трейлер (2015)',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'duration': 153,
+            }
+        },
+        # XHamster embed
+        {
+            'url': 'http://www.numisc.com/forum/showthread.php?11696-FM15-which-pumiscer-was-this-%28-vid-%29-%28-alfa-as-fuck-srx-%29&s=711f5db534502e22260dec8c5e2d66d8',
+            'info_dict': {
+                'id': 'showthread',
+                'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
+            },
+            'playlist_mincount': 7,
+        },
         # Embedded TED video
         {
             'url': 'http://en.support.wordpress.com/videos/ted-talks/',
@@ -382,6 +399,26 @@ class GenericIE(InfoExtractor):
                 'skip_download': 'Requires rtmpdump'
             }
         },
+        # francetv embed
+        {
+            'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
+            'info_dict': {
+                'id': 'EV_30231',
+                'ext': 'mp4',
+                'title': 'Alcaline, le concert avec Calogero',
+                'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
+                'upload_date': '20150226',
+                'timestamp': 1424989860,
+                'duration': 5400,
+            },
+            'params': {
+                # m3u8 downloads
+                'skip_download': True,
+            },
+            'expected_warnings': [
+                'Forbidden'
+            ]
+        },
         # Condé Nast embed
         {
             'url': 'http://www.wired.com/2014/04/honda-asimo/',
@@ -655,6 +692,18 @@ class GenericIE(InfoExtractor):
                 'title': 'John Carlson Postgame 2/25/15',
             },
         },
+        # Kaltura embed (different embed code)
+        {
+            'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
+            'info_dict': {
+                'id': '1_a52wc67y',
+                'ext': 'flv',
+                'upload_date': '20150127',
+                'uploader_id': 'PremierMedia',
+                'timestamp': int,
+                'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
+            },
+        },
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
@@ -812,6 +861,50 @@ class GenericIE(InfoExtractor):
                 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.',
                 'uploader': 'Rogers Sportsnet',
             },
+        },
+        # Dailymotion Cloud video
+        {
+            'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
+            'md5': '49444254273501a64675a7e68c502681',
+            'info_dict': {
+                'id': '5585de919473990de4bee11b',
+                'ext': 'mp4',
+                'title': 'Le débat',
+                'thumbnail': 're:^https?://.*\.jpe?g$',
+            }
+        },
+        # OnionStudios embed
+        {
+            'url': 'http://www.clickhole.com/video/dont-understand-bitcoin-man-will-mumble-explanatio-2537',
+            'info_dict': {
+                'id': '2855',
+                'ext': 'mp4',
+                'title': 'Don’t Understand Bitcoin? This Man Will Mumble An Explanation At You',
+                'thumbnail': 're:^https?://.*\.jpe?g$',
+                'uploader': 'ClickHole',
+                'uploader_id': 'clickhole',
+            }
+        },
+        # SnagFilms embed
+        {
+            'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
+            'info_dict': {
+                'id': '74849a00-85a9-11e1-9660-123139220831',
+                'ext': 'mp4',
+                'title': '#whilewewatch',
+            }
+        },
+        # AdobeTVVideo embed
+        {
+            'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
+            'md5': '43662b577c018ad707a63766462b1e87',
+            'info_dict': {
+                'id': '2456',
+                'ext': 'mp4',
+                'title': 'New experience with Acrobat DC',
+                'description': 'New experience with Acrobat DC',
+                'duration': 248.667,
+            },
         }
     ]
 
@@ -979,7 +1072,9 @@ class GenericIE(InfoExtractor):
             }
 
         if not self._downloader.params.get('test', False) and not is_intentional:
-            self._downloader.report_warning('Falling back on generic information extractor.')
+            force = self._downloader.params.get('force_generic_extractor', False)
+            self._downloader.report_warning(
+                '%s on generic information extractor.' % ('Forcing' if force else 'Falling back'))
 
         if not full_response:
             request = compat_urllib_request.Request(url)
@@ -1031,7 +1126,7 @@ class GenericIE(InfoExtractor):
         # Sometimes embedded video player is hidden behind percent encoding
         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
         # Unescaping the whole page allows to handle those cases in a generic way
-        webpage = compat_urllib_parse.unquote(webpage)
+        webpage = compat_urllib_parse_unquote(webpage)
 
         # it's tempting to parse this further, but you would
         # have to take into account all the variations like
@@ -1089,18 +1184,15 @@ class GenericIE(InfoExtractor):
         if matches:
             return _playlist_from_matches(matches, ie='RtlNl')
 
-        # Look for embedded (iframe) Vimeo player
-        mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
-        if mobj:
-            player_url = unescapeHTML(mobj.group('url'))
-            surl = smuggle_url(player_url, {'Referer': url})
-            return self.url_result(surl)
-        # Look for embedded (swf embed) Vimeo player
-        mobj = re.search(
-            r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
-        if mobj:
-            return self.url_result(mobj.group(1))
+        vimeo_url = VimeoIE._extract_vimeo_url(url, webpage)
+        if vimeo_url is not None:
+            return self.url_result(vimeo_url)
+
+        vid_me_embed_url = self._search_regex(
+            r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
+            webpage, 'vid.me embed', default=None)
+        if vid_me_embed_url is not None:
+            return self.url_result(vid_me_embed_url, 'Vidme')
 
         # Look for embedded YouTube player
         matches = re.findall(r'''(?x)
@@ -1294,7 +1386,7 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'))
         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
         if mobj is not None:
-            return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
+            return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
 
         # Look for funnyordie embed
         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
@@ -1327,6 +1419,11 @@ class GenericIE(InfoExtractor):
         if pornhub_url:
             return self.url_result(pornhub_url, 'PornHub')
 
+        # Look for embedded XHamster player
+        xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
+        if xhamster_urls:
+            return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
+
         # Look for embedded Tvigle player
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
@@ -1352,11 +1449,23 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
 
+        # Look for embedded francetv player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'))
+
         # Look for embedded smotri.com player
         smotri_url = SmotriIE._extract_url(webpage)
         if smotri_url:
             return self.url_result(smotri_url, 'Smotri')
 
+        # Look for embedded Myvi.ru player
+        myvi_url = MyviIE._extract_url(webpage)
+        if myvi_url:
+            return self.url_result(myvi_url)
+
         # Look for embeded soundcloud player
         mobj = re.search(
             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
@@ -1436,8 +1545,8 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'), 'Zapiks')
 
         # Look for Kaltura embeds
-        mobj = re.search(
-            r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
+        mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage) or
+                re.search(r'(?s)(["\'])(?:https?:)?//cdnapisec\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?\1.*?entry_id\s*:\s*(["\'])(?P<id>[^\2]+?)\2', webpage))
         if mobj is not None:
             return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
 
@@ -1494,6 +1603,30 @@ class GenericIE(InfoExtractor):
         if senate_isvp_url:
             return self.url_result(senate_isvp_url, 'SenateISVP')
 
+        # Look for Dailymotion Cloud videos
+        dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
+        if dmcloud_url:
+            return self.url_result(dmcloud_url, 'DailymotionCloud')
+
+        # Look for OnionStudios embeds
+        onionstudios_url = OnionStudiosIE._extract_url(webpage)
+        if onionstudios_url:
+            return self.url_result(onionstudios_url)
+
+        # Look for SnagFilms embeds
+        snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
+        if snagfilms_url:
+            return self.url_result(snagfilms_url)
+
+        # Look for AdobeTVVideo embeds
+        mobj = re.search(
+            r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
+            webpage)
+        if mobj is not None:
+            return self.url_result(
+                self._proto_relative_url(unescapeHTML(mobj.group(1))),
+                'AdobeTVVideo')
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True
@@ -1561,7 +1694,7 @@ class GenericIE(InfoExtractor):
                 if refresh_header:
                     found = re.search(REDIRECT_REGEX, refresh_header)
             if found:
-                new_url = compat_urlparse.urljoin(url, found.group(1))
+                new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
                 self.report_following_redirect(new_url)
                 return {
                     '_type': 'url',
@@ -1573,7 +1706,7 @@ class GenericIE(InfoExtractor):
         entries = []
         for video_url in found:
             video_url = compat_urlparse.urljoin(url, video_url)
-            video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
+            video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
 
             # Sometimes, jwplayer extraction will result in a YouTube URL
             if YoutubeIE.suitable(video_url):

youtube_dl/extractor/gfycat.py

@@ -6,12 +6,13 @@ from ..utils import (
     int_or_none,
     float_or_none,
     qualities,
+    ExtractorError,
 )
 
 
 class GfycatIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?gfycat\.com/(?P<id>[^/?#]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?gfycat\.com/(?:ifr/)?(?P<id>[^/?#]+)'
+    _TESTS = [{
         'url': 'http://gfycat.com/DeadlyDecisiveGermanpinscher',
         'info_dict': {
             'id': 'DeadlyDecisiveGermanpinscher',
@@ -27,14 +28,33 @@ class GfycatIE(InfoExtractor):
             'categories': list,
             'age_limit': 0,
         }
-    }
+    }, {
+        'url': 'http://gfycat.com/ifr/JauntyTimelyAmazontreeboa',
+        'info_dict': {
+            'id': 'JauntyTimelyAmazontreeboa',
+            'ext': 'mp4',
+            'title': 'JauntyTimelyAmazontreeboa',
+            'timestamp': 1411720126,
+            'upload_date': '20140926',
+            'uploader': 'anonymous',
+            'duration': 3.52,
+            'view_count': int,
+            'like_count': int,
+            'dislike_count': int,
+            'categories': list,
+            'age_limit': 0,
+        }
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         gfy = self._download_json(
             'http://gfycat.com/cajax/get/%s' % video_id,
-            video_id, 'Downloading video info')['gfyItem']
+            video_id, 'Downloading video info')
+        if 'error' in gfy:
+            raise ExtractorError('Gfycat said: ' + gfy['error'], expected=True)
+        gfy = gfy['gfyItem']
 
         title = gfy.get('title') or gfy['gfyName']
         description = gfy.get('description')

youtube_dl/extractor/gorillavid.py

@@ -78,12 +78,7 @@ class GorillaVidIE(InfoExtractor):
         if re.search(self._FILE_NOT_FOUND_REGEX, webpage) is not None:
             raise ExtractorError('Video %s does not exist' % video_id, expected=True)
 
-        fields = dict(re.findall(r'''(?x)<input\s+
-            type="hidden"\s+
-            name="([^"]+)"\s+
-            (?:id="[^"]+"\s+)?
-            value="([^"]*)"
-            ''', webpage))
+        fields = self._hidden_inputs(webpage)
 
         if fields['op'] == 'download1':
             countdown = int_or_none(self._search_regex(

youtube_dl/extractor/hentaistigma.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 
 
@@ -19,20 +17,19 @@ class HentaiStigmaIE(InfoExtractor):
     }
 
     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)
 
         title = self._html_search_regex(
-            r'<h2 class="posttitle"><a[^>]*>([^<]+)</a>',
+            r'<h2[^>]+class="posttitle"[^>]*><a[^>]*>([^<]+)</a>',
             webpage, 'title')
         wrap_url = self._html_search_regex(
-            r'<iframe src="([^"]+mp4)"', webpage, 'wrapper url')
+            r'<iframe[^>]+src="([^"]+mp4)"', webpage, 'wrapper url')
         wrap_webpage = self._download_webpage(wrap_url, video_id)
 
         video_url = self._html_search_regex(
-            r'clip:\s*{\s*url: "([^"]*)"', wrap_webpage, 'video url')
+            r'file\s*:\s*"([^"]+)"', wrap_webpage, 'video url')
 
         return {
             'id': video_id,

youtube_dl/extractor/hostingbulk.py

@@ -58,11 +58,7 @@ class HostingBulkIE(InfoExtractor):
             r'<img src="([^"]+)".+?class="pic"',
             webpage, 'thumbnail', fatal=False)
 
-        fields = dict(re.findall(r'''(?x)<input\s+
-            type="hidden"\s+
-            name="([^"]+)"\s+
-            value="([^"]*)"
-            ''', webpage))
+        fields = self._hidden_inputs(webpage)
 
         request = compat_urllib_request.Request(url, urlencode_postdata(fields))
         request.add_header('Content-type', 'application/x-www-form-urlencoded')

youtube_dl/extractor/howcast.py

@@ -1,8 +1,7 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import parse_iso8601
 
 
 class HowcastIE(InfoExtractor):
@@ -13,29 +12,31 @@ class HowcastIE(InfoExtractor):
         'info_dict': {
             'id': '390161',
             'ext': 'mp4',
-            'description': 'The square knot, also known as the reef knot, is one of the oldest, most basic knots to tie, and can be used in many different ways. Here\'s the proper way to tie a square knot.',
             'title': 'How to Tie a Square Knot Properly',
-        }
+            'description': 'md5:dbe792e5f6f1489027027bf2eba188a3',
+            'timestamp': 1276081287,
+            'upload_date': '20100609',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
 
-        video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
 
-        self.report_extraction(video_id)
-
-        video_url = self._search_regex(r'\'?file\'?: "(http://mobile-media\.howcast\.com/[0-9]+\.mp4)',
-                                       webpage, 'video URL')
-
-        video_description = self._html_search_regex(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') name=\'description\'',
-                                                    webpage, 'description', fatal=False)
+        embed_code = self._search_regex(
+            r'<iframe[^>]+src="[^"]+\bembed_code=([^\b]+)\b',
+            webpage, 'ooyala embed code')
 
         return {
+            '_type': 'url_transparent',
+            'ie_key': 'Ooyala',
+            'url': 'ooyala:%s' % embed_code,
             'id': video_id,
-            'url': video_url,
-            'title': self._og_search_title(webpage),
-            'description': video_description,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'timestamp': parse_iso8601(self._html_search_meta(
+                'article:published_time', webpage, 'timestamp')),
         }

youtube_dl/extractor/howstuffworks.py

@@ -10,7 +10,7 @@ from ..utils import (
 
 
 class HowStuffWorksIE(InfoExtractor):
-    _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm'
+    _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*(?:\d+-)?(?P<id>.+?)-video\.htm'
     _TESTS = [
         {
             'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm',
@@ -46,6 +46,10 @@ class HowStuffWorksIE(InfoExtractor):
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
         },
+        {
+            'url': 'http://shows.howstuffworks.com/stuff-to-blow-your-mind/optical-illusions-video.htm',
+            'only_matching': True,
+        }
     ]
 
     def _real_extract(self, url):

youtube_dl/extractor/imdb.py

@@ -46,7 +46,7 @@ class ImdbIE(InfoExtractor):
             format_info = info['videoPlayerObject']['video']
             formats.append({
                 'format_id': f_id,
-                'url': format_info['url'],
+                'url': format_info['videoInfoList'][0]['videoUrl'],
             })
 
         return {

youtube_dl/extractor/ina.py

@@ -7,7 +7,7 @@ from .common import InfoExtractor
 
 
 class InaIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?ina\.fr/video/(?P<id>I?[A-Z0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?ina\.fr/video/(?P<id>I?[A-Z0-9]+)'
     _TEST = {
         'url': 'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
         'md5': 'a667021bf2b41f8dc6049479d9bb38a3',

youtube_dl/extractor/infoq.py

@@ -4,14 +4,15 @@ import base64
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse,
+    compat_urllib_parse_unquote,
+    compat_urlparse,
 )
 
 
 class InfoQIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?infoq\.com/[^/]+/(?P<id>[^/]+)$'
+    _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
         'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
         'info_dict': {
@@ -20,7 +21,10 @@ class InfoQIE(InfoExtractor):
             'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
             'title': 'A Few of My Favorite [Python] Things',
         },
-    }
+    }, {
+        'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -35,14 +39,14 @@ class InfoQIE(InfoExtractor):
         # Extract video URL
         encoded_id = self._search_regex(
             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id')
-        real_id = compat_urllib_parse.unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
+        real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
         playpath = 'mp4:' + real_id
 
         video_filename = playpath.split('/')[-1]
         video_id, extension = video_filename.split('.')
 
         http_base = self._search_regex(
-            r'EXPRESSINSTALL_SWF\s*=\s*"(https?://[^/"]+/)', webpage,
+            r'EXPRESSINSTALL_SWF\s*=\s*[^"]*"((?:https?:)?//[^/"]+/)', webpage,
             'HTTP base URL')
 
         formats = [{
@@ -52,7 +56,7 @@ class InfoQIE(InfoExtractor):
             'play_path': playpath,
         }, {
             'format_id': 'http',
-            'url': http_base + real_id,
+            'url': compat_urlparse.urljoin(url, http_base) + real_id,
         }]
         self._sort_formats(formats)
 

youtube_dl/extractor/iqiyi.py

@@ -3,23 +3,18 @@ from __future__ import unicode_literals
 
 import hashlib
 import math
-import os.path
 import random
-import re
 import time
 import uuid
-import zlib
 
 from .common import InfoExtractor
 from ..compat import compat_urllib_parse
-from ..utils import (
-    ExtractorError,
-    url_basename,
-)
+from ..utils import ExtractorError
 
 
 class IqiyiIE(InfoExtractor):
     IE_NAME = 'iqiyi'
+    IE_DESC = '爱奇艺'
 
     _VALID_URL = r'http://(?:www\.)iqiyi.com/v_.+?\.html'
 
@@ -38,62 +33,57 @@ class IqiyiIE(InfoExtractor):
             'title': '名侦探柯南第752集',
         },
         'playlist': [{
-            'md5': '7e49376fecaffa115d951634917fe105',
             'info_dict': {
                 'id': 'e3f585b550a280af23c98b6cb2be19fb_part1',
                 'ext': 'f4v',
                 'title': '名侦探柯南第752集',
             },
         }, {
-            'md5': '41b75ba13bb7ac0e411131f92bc4f6ca',
             'info_dict': {
                 'id': 'e3f585b550a280af23c98b6cb2be19fb_part2',
                 'ext': 'f4v',
                 'title': '名侦探柯南第752集',
             },
         }, {
-            'md5': '0cee1dd0a3d46a83e71e2badeae2aab0',
             'info_dict': {
                 'id': 'e3f585b550a280af23c98b6cb2be19fb_part3',
                 'ext': 'f4v',
                 'title': '名侦探柯南第752集',
             },
         }, {
-            'md5': '4f8ad72373b0c491b582e7c196b0b1f9',
             'info_dict': {
                 'id': 'e3f585b550a280af23c98b6cb2be19fb_part4',
                 'ext': 'f4v',
                 'title': '名侦探柯南第752集',
             },
         }, {
-            'md5': 'd89ad028bcfad282918e8098e811711d',
             'info_dict': {
                 'id': 'e3f585b550a280af23c98b6cb2be19fb_part5',
                 'ext': 'f4v',
                 'title': '名侦探柯南第752集',
             },
         }, {
-            'md5': '9cb1e5c95da25dff0660c32ae50903b7',
             'info_dict': {
                 'id': 'e3f585b550a280af23c98b6cb2be19fb_part6',
                 'ext': 'f4v',
                 'title': '名侦探柯南第752集',
             },
         }, {
-            'md5': '155116e0ff1867bbc9b98df294faabc9',
             'info_dict': {
                 'id': 'e3f585b550a280af23c98b6cb2be19fb_part7',
                 'ext': 'f4v',
                 'title': '名侦探柯南第752集',
             },
         }, {
-            'md5': '53f5db77622ae14fa493ed2a278a082b',
             'info_dict': {
                 'id': 'e3f585b550a280af23c98b6cb2be19fb_part8',
                 'ext': 'f4v',
                 'title': '名侦探柯南第752集',
             },
         }],
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     _FORMATS_MAP = [
@@ -211,20 +201,7 @@ class IqiyiIE(InfoExtractor):
         return raw_data
 
     def get_enc_key(self, swf_url, video_id):
-        filename, _ = os.path.splitext(url_basename(swf_url))
-        enc_key_json = self._downloader.cache.load('iqiyi-enc-key', filename)
-        if enc_key_json is not None:
-            return enc_key_json[0]
-
-        req = self._request_webpage(
-            swf_url, video_id, note='download swf content')
-        cn = req.read()
-        cn = zlib.decompress(cn[8:])
-        pt = re.compile(b'MixerRemote\x08(?P<enc_key>.+?)\$&vv')
-        enc_key = self._search_regex(pt, cn, 'enc_key').decode('utf8')
-
-        self._downloader.cache.store('iqiyi-enc-key', filename, [enc_key])
-
+        enc_key = '8e29ab5666d041c3a1ea76e06dabdffb'
         return enc_key
 
     def _real_extract(self, url):

youtube_dl/extractor/ir90tv.py

@@ -0,0 +1,42 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import remove_start
+
+
+class Ir90TvIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?90tv\.ir/video/(?P<id>[0-9]+)/.*'
+    _TESTS = [{
+        'url': 'http://90tv.ir/video/95719/%D8%B4%D8%A7%DB%8C%D8%B9%D8%A7%D8%AA-%D9%86%D9%82%D9%84-%D9%88-%D8%A7%D9%86%D8%AA%D9%82%D8%A7%D9%84%D8%A7%D8%AA-%D9%85%D9%87%D9%85-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7-940218',
+        'md5': '411dbd94891381960cb9e13daa47a869',
+        'info_dict': {
+            'id': '95719',
+            'ext': 'mp4',
+            'title': 'شایعات نقل و انتقالات مهم فوتبال اروپا 94/02/18',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }, {
+        'url': 'http://www.90tv.ir/video/95719/%D8%B4%D8%A7%DB%8C%D8%B9%D8%A7%D8%AA-%D9%86%D9%82%D9%84-%D9%88-%D8%A7%D9%86%D8%AA%D9%82%D8%A7%D9%84%D8%A7%D8%AA-%D9%85%D9%87%D9%85-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7-940218',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = remove_start(self._html_search_regex(
+            r'<title>([^<]+)</title>', webpage, 'title'), '90tv.ir :: ')
+
+        video_url = self._search_regex(
+            r'<source[^>]+src="([^"]+)"', webpage, 'video url')
+
+        thumbnail = self._search_regex(r'poster="([^"]+)"', webpage, 'thumbnail url', fatal=False)
+
+        return {
+            'url': video_url,
+            'id': video_id,
+            'title': title,
+            'video_url': video_url,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/jeuxvideo.py

@@ -8,9 +8,9 @@ from .common import InfoExtractor
 
 
 class JeuxVideoIE(InfoExtractor):
-    _VALID_URL = r'http://.*?\.jeuxvideo\.com/.*/(.*?)-\d+\.htm'
+    _VALID_URL = r'http://.*?\.jeuxvideo\.com/.*/(.*?)\.htm'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.jeuxvideo.com/reportages-videos-jeux/0004/00046170/tearaway-playstation-vita-gc-2013-tearaway-nous-presente-ses-papiers-d-identite-00115182.htm',
         'md5': '046e491afb32a8aaac1f44dd4ddd54ee',
         'info_dict': {
@@ -19,7 +19,10 @@ class JeuxVideoIE(InfoExtractor):
             'title': 'Tearaway : GC 2013 : Tearaway nous présente ses papiers d\'identité',
             'description': 'Lorsque les développeurs de LittleBigPlanet proposent un nouveau titre, on ne peut que s\'attendre à un résultat original et fort attrayant.',
         },
-    }
+    }, {
+        'url': 'http://www.jeuxvideo.com/videos/chroniques/434220/l-histoire-du-jeu-video-la-saturn.htm',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/karaoketv.py

@@ -2,7 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse
+from ..compat import compat_urllib_parse_unquote_plus
 from ..utils import (
     js_to_json,
 )
@@ -24,7 +24,7 @@ class KaraoketvIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         page_video_url = self._og_search_video_url(webpage, video_id)
-        config_json = compat_urllib_parse.unquote_plus(self._search_regex(
+        config_json = compat_urllib_parse_unquote_plus(self._search_regex(
             r'config=(.*)', page_video_url, 'configuration'))
 
         urls_info_json = self._download_json(

youtube_dl/extractor/kuwo.py

@@ -0,0 +1,314 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+import itertools
+
+from .common import InfoExtractor
+from ..utils import (
+    get_element_by_id,
+    clean_html,
+    ExtractorError,
+    remove_start,
+)
+
+
+class KuwoBaseIE(InfoExtractor):
+    _FORMATS = [
+        {'format': 'ape', 'ext': 'ape', 'preference': 100},
+        {'format': 'mp3-320', 'ext': 'mp3', 'br': '320kmp3', 'abr': 320, 'preference': 80},
+        {'format': 'mp3-192', 'ext': 'mp3', 'br': '192kmp3', 'abr': 192, 'preference': 70},
+        {'format': 'mp3-128', 'ext': 'mp3', 'br': '128kmp3', 'abr': 128, 'preference': 60},
+        {'format': 'wma', 'ext': 'wma', 'preference': 20},
+        {'format': 'aac', 'ext': 'aac', 'abr': 48, 'preference': 10}
+    ]
+
+    def _get_formats(self, song_id):
+        formats = []
+        for file_format in self._FORMATS:
+            song_url = self._download_webpage(
+                'http://antiserver.kuwo.cn/anti.s?format=%s&br=%s&rid=MUSIC_%s&type=convert_url&response=url' %
+                (file_format['ext'], file_format.get('br', ''), song_id),
+                song_id, note='Download %s url info' % file_format['format'],
+            )
+            if song_url.startswith('http://') or song_url.startswith('https://'):
+                formats.append({
+                    'url': song_url,
+                    'format_id': file_format['format'],
+                    'format': file_format['format'],
+                    'preference': file_format['preference'],
+                    'abr': file_format.get('abr'),
+                })
+        self._sort_formats(formats)
+        return formats
+
+
+class KuwoIE(KuwoBaseIE):
+    IE_NAME = 'kuwo:song'
+    IE_DESC = '酷我音乐'
+    _VALID_URL = r'http://www\.kuwo\.cn/yinyue/(?P<id>\d+?)/'
+    _TESTS = [{
+        'url': 'http://www.kuwo.cn/yinyue/635632/',
+        'info_dict': {
+            'id': '635632',
+            'ext': 'ape',
+            'title': '爱我别走',
+            'creator': '张震岳',
+            'upload_date': '20080122',
+            'description': 'md5:ed13f58e3c3bf3f7fd9fbc4e5a7aa75c'
+        },
+    }, {
+        'url': 'http://www.kuwo.cn/yinyue/6446136/',
+        'info_dict': {
+            'id': '6446136',
+            'ext': 'mp3',
+            'title': '心',
+            'creator': 'IU',
+            'upload_date': '20150518',
+        },
+        'params': {
+            'format': 'mp3-320'
+        },
+    }]
+
+    def _real_extract(self, url):
+        song_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, song_id, note='Download song detail info',
+            errnote='Unable to get song detail info')
+
+        song_name = self._html_search_regex(
+            r'<h1[^>]+title="([^"]+)">', webpage, 'song name')
+        singer_name = self._html_search_regex(
+            r'<div[^>]+class="s_img">\s*<a[^>]+title="([^>]+)"',
+            webpage, 'singer name', fatal=False)
+        lrc_content = clean_html(get_element_by_id('lrcContent', webpage))
+        if lrc_content == '暂无':     # indicates no lyrics
+            lrc_content = None
+
+        formats = self._get_formats(song_id)
+
+        album_id = self._html_search_regex(
+            r'<p[^>]+class="album"[^<]+<a[^>]+href="http://www\.kuwo\.cn/album/(\d+)/"',
+            webpage, 'album id', fatal=False)
+
+        publish_time = None
+        if album_id is not None:
+            album_info_page = self._download_webpage(
+                'http://www.kuwo.cn/album/%s/' % album_id, song_id,
+                note='Download album detail info',
+                errnote='Unable to get album detail info')
+
+            publish_time = self._html_search_regex(
+                r'发行时间：(\d{4}-\d{2}-\d{2})', album_info_page,
+                'publish time', fatal=False)
+            if publish_time:
+                publish_time = publish_time.replace('-', '')
+
+        return {
+            'id': song_id,
+            'title': song_name,
+            'creator': singer_name,
+            'upload_date': publish_time,
+            'description': lrc_content,
+            'formats': formats,
+        }
+
+
+class KuwoAlbumIE(InfoExtractor):
+    IE_NAME = 'kuwo:album'
+    IE_DESC = '酷我音乐 - 专辑'
+    _VALID_URL = r'http://www\.kuwo\.cn/album/(?P<id>\d+?)/'
+    _TEST = {
+        'url': 'http://www.kuwo.cn/album/502294/',
+        'info_dict': {
+            'id': '502294',
+            'title': 'M',
+            'description': 'md5:6a7235a84cc6400ec3b38a7bdaf1d60c',
+        },
+        'playlist_count': 2,
+    }
+
+    def _real_extract(self, url):
+        album_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            url, album_id, note='Download album info',
+            errnote='Unable to get album info')
+
+        album_name = self._html_search_regex(
+            r'<div[^>]+class="comm"[^<]+<h1[^>]+title="([^"]+)"', webpage,
+            'album name')
+        album_intro = remove_start(
+            clean_html(get_element_by_id('intro', webpage)),
+            '%s简介：' % album_name)
+
+        entries = [
+            self.url_result(song_url, 'Kuwo') for song_url in re.findall(
+                r'<p[^>]+class="listen"><a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+/)"',
+                webpage)
+        ]
+        return self.playlist_result(entries, album_id, album_name, album_intro)
+
+
+class KuwoChartIE(InfoExtractor):
+    IE_NAME = 'kuwo:chart'
+    IE_DESC = '酷我音乐 - 排行榜'
+    _VALID_URL = r'http://yinyue\.kuwo\.cn/billboard_(?P<id>[^.]+).htm'
+    _TEST = {
+        'url': 'http://yinyue.kuwo.cn/billboard_香港中文龙虎榜.htm',
+        'info_dict': {
+            'id': '香港中文龙虎榜',
+            'title': '香港中文龙虎榜',
+            'description': 're:\d{4}第\d{2}期',
+        },
+        'playlist_mincount': 10,
+    }
+
+    def _real_extract(self, url):
+        chart_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, chart_id, note='Download chart info',
+            errnote='Unable to get chart info')
+
+        chart_name = self._html_search_regex(
+            r'<h1[^>]+class="unDis">([^<]+)</h1>', webpage, 'chart name')
+
+        chart_desc = self._html_search_regex(
+            r'<p[^>]+class="tabDef">(\d{4}第\d{2}期)</p>', webpage, 'chart desc')
+
+        entries = [
+            self.url_result(song_url, 'Kuwo') for song_url in re.findall(
+                r'<a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+)/"', webpage)
+        ]
+        return self.playlist_result(entries, chart_id, chart_name, chart_desc)
+
+
+class KuwoSingerIE(InfoExtractor):
+    IE_NAME = 'kuwo:singer'
+    IE_DESC = '酷我音乐 - 歌手'
+    _VALID_URL = r'http://www\.kuwo\.cn/mingxing/(?P<id>[^/]+)'
+    _TESTS = [{
+        'url': 'http://www.kuwo.cn/mingxing/bruno+mars/',
+        'info_dict': {
+            'id': 'bruno+mars',
+            'title': 'Bruno Mars',
+        },
+        'playlist_count': 10,
+    }, {
+        'url': 'http://www.kuwo.cn/mingxing/Ali/music.htm',
+        'info_dict': {
+            'id': 'Ali',
+            'title': 'Ali',
+        },
+        'playlist_mincount': 95,
+    }]
+
+    def _real_extract(self, url):
+        singer_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, singer_id, note='Download singer info',
+            errnote='Unable to get singer info')
+
+        singer_name = self._html_search_regex(
+            r'<div class="title clearfix">\s*<h1>([^<]+)<span', webpage, 'singer name'
+        )
+
+        entries = []
+        first_page_only = False if re.search(r'/music(?:_\d+)?\.htm', url) else True
+        for page_num in itertools.count(1):
+            webpage = self._download_webpage(
+                'http://www.kuwo.cn/mingxing/%s/music_%d.htm' % (singer_id, page_num),
+                singer_id, note='Download song list page #%d' % page_num,
+                errnote='Unable to get song list page #%d' % page_num)
+
+            entries.extend([
+                self.url_result(song_url, 'Kuwo') for song_url in re.findall(
+                    r'<p[^>]+class="m_name"><a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+)/',
+                    webpage)
+            ][:10 if first_page_only else None])
+
+            if first_page_only or not re.search(r'<a[^>]+href="[^"]+">下一页</a>', webpage):
+                break
+
+        return self.playlist_result(entries, singer_id, singer_name)
+
+
+class KuwoCategoryIE(InfoExtractor):
+    IE_NAME = 'kuwo:category'
+    IE_DESC = '酷我音乐 - 分类'
+    _VALID_URL = r'http://yinyue\.kuwo\.cn/yy/cinfo_(?P<id>\d+?).htm'
+    _TEST = {
+        'url': 'http://yinyue.kuwo.cn/yy/cinfo_86375.htm',
+        'info_dict': {
+            'id': '86375',
+            'title': '八十年代精选',
+            'description': '这些都是属于八十年代的回忆！',
+        },
+        'playlist_count': 30,
+    }
+
+    def _real_extract(self, url):
+        category_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, category_id, note='Download category info',
+            errnote='Unable to get category info')
+
+        category_name = self._html_search_regex(
+            r'<h1[^>]+title="([^<>]+?)">[^<>]+?</h1>', webpage, 'category name')
+
+        category_desc = remove_start(
+            get_element_by_id('intro', webpage).strip(),
+            '%s简介：' % category_name)
+
+        jsonm = self._parse_json(self._html_search_regex(
+            r'var\s+jsonm\s*=\s*([^;]+);', webpage, 'category songs'), category_id)
+
+        entries = [
+            self.url_result('http://www.kuwo.cn/yinyue/%s/' % song['musicrid'], 'Kuwo')
+            for song in jsonm['musiclist']
+        ]
+        return self.playlist_result(entries, category_id, category_name, category_desc)
+
+
+class KuwoMvIE(KuwoBaseIE):
+    IE_NAME = 'kuwo:mv'
+    IE_DESC = '酷我音乐 - MV'
+    _VALID_URL = r'http://www\.kuwo\.cn/mv/(?P<id>\d+?)/'
+    _TEST = {
+        'url': 'http://www.kuwo.cn/mv/6480076/',
+        'info_dict': {
+            'id': '6480076',
+            'ext': 'mkv',
+            'title': '我们家MV',
+            'creator': '2PM',
+        },
+    }
+    _FORMATS = KuwoBaseIE._FORMATS + [
+        {'format': 'mkv', 'ext': 'mkv', 'preference': 250},
+        {'format': 'mp4', 'ext': 'mp4', 'preference': 200},
+    ]
+
+    def _real_extract(self, url):
+        song_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, song_id, note='Download mv detail info: %s' % song_id,
+            errnote='Unable to get mv detail info: %s' % song_id)
+
+        mobj = re.search(
+            r'<h1[^>]+title="(?P<song>[^"]+)">[^<]+<span[^>]+title="(?P<singer>[^"]+)"',
+            webpage)
+        if mobj:
+            song_name = mobj.group('song')
+            singer_name = mobj.group('singer')
+        else:
+            raise ExtractorError('Unable to find song or singer names')
+
+        formats = self._get_formats(song_id)
+
+        return {
+            'id': song_id,
+            'title': song_name,
+            'creator': singer_name,
+            'formats': formats,
+        }

youtube_dl/extractor/lecture2go.py

@@ -0,0 +1,62 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    parse_duration,
+    int_or_none,
+)
+
+
+class Lecture2GoIE(InfoExtractor):
+    _VALID_URL = r'https?://lecture2go\.uni-hamburg\.de/veranstaltungen/-/v/(?P<id>\d+)'
+    _TEST = {
+        'url': 'https://lecture2go.uni-hamburg.de/veranstaltungen/-/v/17473',
+        'md5': 'ac02b570883020d208d405d5a3fd2f7f',
+        'info_dict': {
+            'id': '17473',
+            'ext': 'flv',
+            'title': '2 - Endliche Automaten und reguläre Sprachen',
+            'creator': 'Frank Heitmann',
+            'duration': 5220,
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_regex(r'<em[^>]+class="title">(.+)</em>', webpage, 'title')
+
+        formats = []
+        for url in set(re.findall(r'"src","([^"]+)"', webpage)):
+            ext = determine_ext(url)
+            if ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(url, video_id))
+            elif ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(url, video_id))
+            else:
+                formats.append({
+                    'url': url,
+                })
+
+        self._sort_formats(formats)
+
+        creator = self._html_search_regex(
+            r'<div[^>]+id="description">([^<]+)</div>', webpage, 'creator', fatal=False)
+        duration = parse_duration(self._html_search_regex(
+            r'Duration:\s*</em>\s*<em[^>]*>([^<]+)</em>', webpage, 'duration', fatal=False))
+        view_count = int_or_none(self._html_search_regex(
+            r'Views:\s*</em>\s*<em[^>]+>(\d+)</em>', webpage, 'view count', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'creator': creator,
+            'duration': duration,
+            'view_count': view_count,
+        }

youtube_dl/extractor/letv.py

@@ -15,10 +15,12 @@ from ..utils import (
     determine_ext,
     ExtractorError,
     parse_iso8601,
+    int_or_none,
 )
 
 
 class LetvIE(InfoExtractor):
+    IE_DESC = '乐视网'
     _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P<id>\d+).html'
 
     _TESTS = [{
@@ -133,7 +135,7 @@ class LetvIE(InfoExtractor):
                 }
 
                 if format_id[-1:] == 'p':
-                    url_info_dict['height'] = format_id[:-1]
+                    url_info_dict['height'] = int_or_none(format_id[:-1])
 
                 urls.append(url_info_dict)
 

youtube_dl/extractor/lifenews.py

@@ -8,6 +8,7 @@ from ..compat import compat_urlparse
 from ..utils import (
     determine_ext,
     int_or_none,
+    remove_end,
     unified_strdate,
     ExtractorError,
 )
@@ -39,7 +40,6 @@ class LifeNewsIE(InfoExtractor):
             'title': 'В Сети появилось видео захвата «Правым сектором» колхозных полей ',
             'description': 'Жители двух поселков Днепропетровской области не простили радикалам угрозу лишения плодородных земель и пошли в лобовую. ',
             'upload_date': '20150402',
-            'uploader': 'embed.life.ru',
         }
     }, {
         'url': 'http://lifenews.ru/news/153461',
@@ -50,7 +50,6 @@ class LifeNewsIE(InfoExtractor):
             'title': 'В Москве спасли потерявшегося медвежонка, который спрятался на дереве',
             'description': 'Маленький хищник не смог найти дорогу домой и обрел временное убежище на тополе недалеко от жилого массива, пока его не нашла соседская собака.',
             'upload_date': '20150505',
-            'uploader': 'embed.life.ru',
         }
     }, {
         'url': 'http://lifenews.ru/video/13035',
@@ -72,20 +71,20 @@ class LifeNewsIE(InfoExtractor):
         if not videos and not iframe_link:
             raise ExtractorError('No media links available for %s' % video_id)
 
-        title = self._og_search_title(webpage)
-        TITLE_SUFFIX = ' - Первый по срочным новостям — LIFE | NEWS'
-        if title.endswith(TITLE_SUFFIX):
-            title = title[:-len(TITLE_SUFFIX)]
+        title = remove_end(
+            self._og_search_title(webpage),
+            ' - Первый по срочным новостям — LIFE | NEWS')
 
         description = self._og_search_description(webpage)
 
         view_count = self._html_search_regex(
             r'<div class=\'views\'>\s*(\d+)\s*</div>', webpage, 'view count', fatal=False)
         comment_count = self._html_search_regex(
-            r'<div class=\'comments\'>\s*<span class=\'counter\'>\s*(\d+)\s*</span>', webpage, 'comment count', fatal=False)
+            r'=\'commentCount\'[^>]*>\s*(\d+)\s*<',
+            webpage, 'comment count', fatal=False)
 
         upload_date = self._html_search_regex(
-            r'<time datetime=\'([^\']+)\'>', webpage, 'upload date', fatal=False)
+            r'<time[^>]*datetime=\'([^\']+)\'', webpage, 'upload date', fatal=False)
         if upload_date is not None:
             upload_date = unified_strdate(upload_date)
 

youtube_dl/extractor/lynda.py

@@ -30,13 +30,13 @@ class LyndaBaseIE(InfoExtractor):
             return
 
         login_form = {
-            'username': username,
-            'password': password,
+            'username': username.encode('utf-8'),
+            'password': password.encode('utf-8'),
             'remember': 'false',
             'stayPut': 'false'
         }
         request = compat_urllib_request.Request(
-            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
+            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
         login_page = self._download_webpage(
             request, None, 'Logging in as %s' % username)
 
@@ -65,7 +65,7 @@ class LyndaBaseIE(InfoExtractor):
                     'stayPut': 'false',
                 }
                 request = compat_urllib_request.Request(
-                    self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form))
+                    self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form).encode('utf-8'))
                 login_page = self._download_webpage(
                     request, None,
                     'Confirming log in and log out from another device')

youtube_dl/extractor/malemotion.py

@@ -2,9 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse,
-)
+from ..compat import compat_urllib_parse_unquote
 
 
 class MalemotionIE(InfoExtractor):
@@ -24,7 +22,7 @@ class MalemotionIE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        video_url = compat_urllib_parse.unquote(self._search_regex(
+        video_url = compat_urllib_parse_unquote(self._search_regex(
             r'<source type="video/mp4" src="(.+?)"', webpage, 'video URL'))
         video_title = self._html_search_regex(
             r'<title>(.*?)</title', webpage, 'title')

youtube_dl/extractor/metacafe.py

@@ -6,6 +6,7 @@ from .common import InfoExtractor
 from ..compat import (
     compat_parse_qs,
     compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urllib_request,
 )
 from ..utils import (
@@ -155,7 +156,7 @@ class MetacafeIE(InfoExtractor):
         video_url = None
         mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage)
         if mobj is not None:
-            mediaURL = compat_urllib_parse.unquote(mobj.group(1))
+            mediaURL = compat_urllib_parse_unquote(mobj.group(1))
             video_ext = mediaURL[-3:]
 
             # Extract gdaKey if available

youtube_dl/extractor/mitele.py

@@ -5,6 +5,7 @@ import json
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urlparse,
 )
 from ..utils import (
@@ -48,7 +49,7 @@ class MiTeleIE(InfoExtractor):
             domain = 'http://' + domain
         info_url = compat_urlparse.urljoin(
             domain,
-            compat_urllib_parse.unquote(embed_data['flashvars']['host'])
+            compat_urllib_parse_unquote(embed_data['flashvars']['host'])
         )
         info_el = self._download_xml(info_url, episode).find('./video/info')
 

youtube_dl/extractor/mixcloud.py

@@ -3,9 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse,
-)
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     ExtractorError,
     HEADRequest,
@@ -60,7 +58,7 @@ class MixcloudIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         uploader = mobj.group(1)
         cloudcast_name = mobj.group(2)
-        track_id = compat_urllib_parse.unquote('-'.join((uploader, cloudcast_name)))
+        track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name)))
 
         webpage = self._download_webpage(url, track_id)
 

youtube_dl/extractor/mofosex.py

@@ -5,9 +5,9 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_urllib_parse_unquote,
     compat_urllib_parse_urlparse,
     compat_urllib_request,
-    compat_urllib_parse,
 )
 
 
@@ -34,7 +34,7 @@ class MofosexIE(InfoExtractor):
         webpage = self._download_webpage(req, video_id)
 
         video_title = self._html_search_regex(r'<h1>(.+?)<', webpage, 'title')
-        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'flashvars.video_url = \'([^\']+)', webpage, 'video_url'))
+        video_url = compat_urllib_parse_unquote(self._html_search_regex(r'flashvars.video_url = \'([^\']+)', webpage, 'video_url'))
         path = compat_urllib_parse_urlparse(video_url).path
         extension = os.path.splitext(path)[1][1:]
         format = path.split('/')[5].split('_')[:2]

youtube_dl/extractor/myspass.py

@@ -35,7 +35,8 @@ class MySpassIE(InfoExtractor):
 
         # get metadata
         metadata_url = META_DATA_URL_TEMPLATE % video_id
-        metadata = self._download_xml(metadata_url, video_id)
+        metadata = self._download_xml(
+            metadata_url, video_id, transform_source=lambda s: s.strip())
 
         # extract values from metadata
         url_flv_el = metadata.find('url_flv')

youtube_dl/extractor/myvi.py

@@ -0,0 +1,60 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .vimple import SprutoBaseIE
+
+
+class MyviIE(SprutoBaseIE):
+    _VALID_URL = r'''(?x)
+                    https?://
+                        myvi\.(?:ru/player|tv)/
+                            (?:
+                                (?:
+                                    embed/html|
+                                    flash|
+                                    api/Video/Get
+                                )/|
+                                content/preloader\.swf\?.*\bid=
+                            )
+                            (?P<id>[\da-zA-Z_-]+)
+                    '''
+    _TESTS = [{
+        'url': 'http://myvi.ru/player/embed/html/oOy4euHA6LVwNNAjhD9_Jq5Ha2Qf0rtVMVFMAZav8wObeRTZaCATzucDQIDph8hQU0',
+        'md5': '571bbdfba9f9ed229dc6d34cc0f335bf',
+        'info_dict': {
+            'id': 'f16b2bbd-cde8-481c-a981-7cd48605df43',
+            'ext': 'mp4',
+            'title': 'хозяин жизни',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 25,
+        },
+    }, {
+        'url': 'http://myvi.ru/player/content/preloader.swf?id=oOy4euHA6LVwNNAjhD9_Jq5Ha2Qf0rtVMVFMAZav8wOYf1WFpPfc_bWTKGVf_Zafr0',
+        'only_matching': True,
+    }, {
+        'url': 'http://myvi.ru/player/api/Video/Get/oOy4euHA6LVwNNAjhD9_Jq5Ha2Qf0rtVMVFMAZav8wObeRTZaCATzucDQIDph8hQU0',
+        'only_matching': True,
+    }, {
+        'url': 'http://myvi.tv/embed/html/oTGTNWdyz4Zwy_u1nraolwZ1odenTd9WkTnRfIL9y8VOgHYqOHApE575x4_xxS9Vn0?ap=0',
+        'only_matching': True,
+    }, {
+        'url': 'http://myvi.ru/player/flash/ocp2qZrHI-eZnHKQBK4cZV60hslH8LALnk0uBfKsB-Q4WnY26SeGoYPi8HWHxu0O30',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def _extract_url(cls, webpage):
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//myvi\.(?:ru/player|tv)/(?:embed/html|flash)/[^"]+)\1', webpage)
+        if mobj:
+            return mobj.group('url')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        spruto = self._download_json(
+            'http://myvi.ru/player/api/Video/Get/%s?sig' % video_id, video_id)['sprutoData']
+
+        return self._extract_spruto(spruto, video_id)

youtube_dl/extractor/myvideo.py

@@ -10,6 +10,7 @@ from .common import InfoExtractor
 from ..compat import (
     compat_ord,
     compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urllib_request,
 )
 from ..utils import (
@@ -107,7 +108,7 @@ class MyVideoIE(InfoExtractor):
             if not a == '_encxml':
                 params[a] = b
             else:
-                encxml = compat_urllib_parse.unquote(b)
+                encxml = compat_urllib_parse_unquote(b)
         if not params.get('domain'):
             params['domain'] = 'www.myvideo.de'
         xmldata_url = '%s?%s' % (encxml, compat_urllib_parse.urlencode(params))
@@ -135,7 +136,7 @@ class MyVideoIE(InfoExtractor):
         video_url = None
         mobj = re.search('connectionurl=\'(.*?)\'', dec_data)
         if mobj:
-            video_url = compat_urllib_parse.unquote(mobj.group(1))
+            video_url = compat_urllib_parse_unquote(mobj.group(1))
             if 'myvideo2flash' in video_url:
                 self.report_warning(
                     'Rewriting URL to use unencrypted rtmp:// ...',
@@ -147,10 +148,10 @@ class MyVideoIE(InfoExtractor):
             mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data)
             if mobj is None:
                 raise ExtractorError('unable to extract url')
-            video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2))
+            video_url = compat_urllib_parse_unquote(mobj.group(1)) + compat_urllib_parse_unquote(mobj.group(2))
 
         video_file = self._search_regex('source=\'(.*?)\'', dec_data, 'video file')
-        video_file = compat_urllib_parse.unquote(video_file)
+        video_file = compat_urllib_parse_unquote(video_file)
 
         if not video_file.endswith('f4m'):
             ppath, prefix = video_file.split('.')
@@ -159,7 +160,7 @@ class MyVideoIE(InfoExtractor):
             video_playpath = ''
 
         video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, 'swfobj')
-        video_swfobj = compat_urllib_parse.unquote(video_swfobj)
+        video_swfobj = compat_urllib_parse_unquote(video_swfobj)
 
         video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>",
                                               webpage, 'title')

youtube_dl/extractor/nationalgeographic.py

@@ -8,25 +8,40 @@ from ..utils import (
 
 
 class NationalGeographicIE(InfoExtractor):
-    _VALID_URL = r'http://video\.nationalgeographic\.com/video/.*?'
+    _VALID_URL = r'http://video\.nationalgeographic\.com/.*?'
 
-    _TEST = {
-        'url': 'http://video.nationalgeographic.com/video/news/150210-news-crab-mating-vin?source=featuredvideo',
-        'info_dict': {
-            'id': '4DmDACA6Qtk_',
-            'ext': 'flv',
-            'title': 'Mating Crabs Busted by Sharks',
-            'description': 'md5:16f25aeffdeba55aaa8ec37e093ad8b3',
+    _TESTS = [
+        {
+            'url': 'http://video.nationalgeographic.com/video/news/150210-news-crab-mating-vin?source=featuredvideo',
+            'info_dict': {
+                'id': '4DmDACA6Qtk_',
+                'ext': 'flv',
+                'title': 'Mating Crabs Busted by Sharks',
+                'description': 'md5:16f25aeffdeba55aaa8ec37e093ad8b3',
+            },
+            'add_ie': ['ThePlatform'],
         },
-        'add_ie': ['ThePlatform'],
-    }
+        {
+            'url': 'http://video.nationalgeographic.com/wild/when-sharks-attack/the-real-jaws',
+            'info_dict': {
+                'id': '_JeBD_D7PlS5',
+                'ext': 'flv',
+                'title': 'The Real Jaws',
+                'description': 'md5:8d3e09d9d53a85cd397b4b21b2c77be6',
+            },
+            'add_ie': ['ThePlatform'],
+        },
+    ]
 
     def _real_extract(self, url):
         name = url_basename(url)
 
         webpage = self._download_webpage(url, name)
-        feed_url = self._search_regex(r'data-feed-url="([^"]+)"', webpage, 'feed url')
-        guid = self._search_regex(r'data-video-guid="([^"]+)"', webpage, 'guid')
+        feed_url = self._search_regex(
+            r'data-feed-url="([^"]+)"', webpage, 'feed url')
+        guid = self._search_regex(
+            r'id="(?:videoPlayer|player-container)"[^>]+data-guid="([^"]+)"',
+            webpage, 'guid')
 
         feed = self._download_xml('%s?byGuid=%s' % (feed_url, guid), name)
         content = feed.find('.//{http://search.yahoo.com/mrss/}content')
@@ -34,5 +49,6 @@ class NationalGeographicIE(InfoExtractor):
 
         return self.url_result(smuggle_url(
             'http://link.theplatform.com/s/ngs/%s?format=SMIL&formats=MPEG4&manifest=f4m' % theplatform_id,
-            # For some reason, the normal links don't work and we must force the use of f4m
+            # For some reason, the normal links don't work and we must force
+            # the use of f4m
             {'force_smil_url': True}))

youtube_dl/extractor/neteasemusic.py

@@ -0,0 +1,459 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from hashlib import md5
+from base64 import b64encode
+from datetime import datetime
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_request,
+    compat_urllib_parse,
+    compat_str,
+    compat_itertools_count,
+)
+
+
+class NetEaseMusicBaseIE(InfoExtractor):
+    _FORMATS = ['bMusic', 'mMusic', 'hMusic']
+    _NETEASE_SALT = '3go8&$8*3*3h0k(2)2'
+    _API_BASE = 'http://music.163.com/api/'
+
+    @classmethod
+    def _encrypt(cls, dfsid):
+        salt_bytes = bytearray(cls._NETEASE_SALT.encode('utf-8'))
+        string_bytes = bytearray(compat_str(dfsid).encode('ascii'))
+        salt_len = len(salt_bytes)
+        for i in range(len(string_bytes)):
+            string_bytes[i] = string_bytes[i] ^ salt_bytes[i % salt_len]
+        m = md5()
+        m.update(bytes(string_bytes))
+        result = b64encode(m.digest()).decode('ascii')
+        return result.replace('/', '_').replace('+', '-')
+
+    @classmethod
+    def extract_formats(cls, info):
+        formats = []
+        for song_format in cls._FORMATS:
+            details = info.get(song_format)
+            if not details:
+                continue
+            formats.append({
+                'url': 'http://m1.music.126.net/%s/%s.%s' %
+                       (cls._encrypt(details['dfsId']), details['dfsId'],
+                        details['extension']),
+                'ext': details.get('extension'),
+                'abr': details.get('bitrate', 0) / 1000,
+                'format_id': song_format,
+                'filesize': details.get('size'),
+                'asr': details.get('sr')
+            })
+        return formats
+
+    @classmethod
+    def convert_milliseconds(cls, ms):
+        return int(round(ms / 1000.0))
+
+    def query_api(self, endpoint, video_id, note):
+        req = compat_urllib_request.Request('%s%s' % (self._API_BASE, endpoint))
+        req.add_header('Referer', self._API_BASE)
+        return self._download_json(req, video_id, note)
+
+
+class NetEaseMusicIE(NetEaseMusicBaseIE):
+    IE_NAME = 'netease:song'
+    IE_DESC = '网易云音乐'
+    _VALID_URL = r'https?://music\.163\.com/(#/)?song\?id=(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://music.163.com/#/song?id=32102397',
+        'md5': 'f2e97280e6345c74ba9d5677dd5dcb45',
+        'info_dict': {
+            'id': '32102397',
+            'ext': 'mp3',
+            'title': 'Bad Blood (feat. Kendrick Lamar)',
+            'creator': 'Taylor Swift / Kendrick Lamar',
+            'upload_date': '20150517',
+            'timestamp': 1431878400,
+            'description': 'md5:a10a54589c2860300d02e1de821eb2ef',
+        },
+    }, {
+        'note': 'No lyrics translation.',
+        'url': 'http://music.163.com/#/song?id=29822014',
+        'info_dict': {
+            'id': '29822014',
+            'ext': 'mp3',
+            'title': '听见下雨的声音',
+            'creator': '周杰伦',
+            'upload_date': '20141225',
+            'timestamp': 1419523200,
+            'description': 'md5:a4d8d89f44656af206b7b2555c0bce6c',
+        },
+    }, {
+        'note': 'No lyrics.',
+        'url': 'http://music.163.com/song?id=17241424',
+        'info_dict': {
+            'id': '17241424',
+            'ext': 'mp3',
+            'title': 'Opus 28',
+            'creator': 'Dustin O\'Halloran',
+            'upload_date': '20080211',
+            'timestamp': 1202745600,
+        },
+    }, {
+        'note': 'Has translated name.',
+        'url': 'http://music.163.com/#/song?id=22735043',
+        'info_dict': {
+            'id': '22735043',
+            'ext': 'mp3',
+            'title': '소원을 말해봐 (Genie)',
+            'creator': '少女时代',
+            'description': 'md5:79d99cc560e4ca97e0c4d86800ee4184',
+            'upload_date': '20100127',
+            'timestamp': 1264608000,
+            'alt_title': '说出愿望吧(Genie)',
+        }
+    }]
+
+    def _process_lyrics(self, lyrics_info):
+        original = lyrics_info.get('lrc', {}).get('lyric')
+        translated = lyrics_info.get('tlyric', {}).get('lyric')
+
+        if not translated:
+            return original
+
+        lyrics_expr = r'(\[[0-9]{2}:[0-9]{2}\.[0-9]{2,}\])([^\n]+)'
+        original_ts_texts = re.findall(lyrics_expr, original)
+        translation_ts_dict = dict(
+            (time_stamp, text) for time_stamp, text in re.findall(lyrics_expr, translated)
+        )
+        lyrics = '\n'.join([
+            '%s%s / %s' % (time_stamp, text, translation_ts_dict.get(time_stamp, ''))
+            for time_stamp, text in original_ts_texts
+        ])
+        return lyrics
+
+    def _real_extract(self, url):
+        song_id = self._match_id(url)
+
+        params = {
+            'id': song_id,
+            'ids': '[%s]' % song_id
+        }
+        info = self.query_api(
+            'song/detail?' + compat_urllib_parse.urlencode(params),
+            song_id, 'Downloading song info')['songs'][0]
+
+        formats = self.extract_formats(info)
+        self._sort_formats(formats)
+
+        lyrics_info = self.query_api(
+            'song/lyric?id=%s&lv=-1&tv=-1' % song_id,
+            song_id, 'Downloading lyrics data')
+        lyrics = self._process_lyrics(lyrics_info)
+
+        alt_title = None
+        if info.get('transNames'):
+            alt_title = '/'.join(info.get('transNames'))
+
+        return {
+            'id': song_id,
+            'title': info['name'],
+            'alt_title': alt_title,
+            'creator': ' / '.join([artist['name'] for artist in info.get('artists', [])]),
+            'timestamp': self.convert_milliseconds(info.get('album', {}).get('publishTime')),
+            'thumbnail': info.get('album', {}).get('picUrl'),
+            'duration': self.convert_milliseconds(info.get('duration', 0)),
+            'description': lyrics,
+            'formats': formats,
+        }
+
+
+class NetEaseMusicAlbumIE(NetEaseMusicBaseIE):
+    IE_NAME = 'netease:album'
+    IE_DESC = '网易云音乐 - 专辑'
+    _VALID_URL = r'https?://music\.163\.com/(#/)?album\?id=(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://music.163.com/#/album?id=220780',
+        'info_dict': {
+            'id': '220780',
+            'title': 'B\'day',
+        },
+        'playlist_count': 23,
+    }
+
+    def _real_extract(self, url):
+        album_id = self._match_id(url)
+
+        info = self.query_api(
+            'album/%s?id=%s' % (album_id, album_id),
+            album_id, 'Downloading album data')['album']
+
+        name = info['name']
+        desc = info.get('description')
+        entries = [
+            self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
+                            'NetEaseMusic', song['id'])
+            for song in info['songs']
+        ]
+        return self.playlist_result(entries, album_id, name, desc)
+
+
+class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
+    IE_NAME = 'netease:singer'
+    IE_DESC = '网易云音乐 - 歌手'
+    _VALID_URL = r'https?://music\.163\.com/(#/)?artist\?id=(?P<id>[0-9]+)'
+    _TESTS = [{
+        'note': 'Singer has aliases.',
+        'url': 'http://music.163.com/#/artist?id=10559',
+        'info_dict': {
+            'id': '10559',
+            'title': '张惠妹 - aMEI;阿密特',
+        },
+        'playlist_count': 50,
+    }, {
+        'note': 'Singer has translated name.',
+        'url': 'http://music.163.com/#/artist?id=124098',
+        'info_dict': {
+            'id': '124098',
+            'title': '李昇基 - 이승기',
+        },
+        'playlist_count': 50,
+    }]
+
+    def _real_extract(self, url):
+        singer_id = self._match_id(url)
+
+        info = self.query_api(
+            'artist/%s?id=%s' % (singer_id, singer_id),
+            singer_id, 'Downloading singer data')
+
+        name = info['artist']['name']
+        if info['artist']['trans']:
+            name = '%s - %s' % (name, info['artist']['trans'])
+        if info['artist']['alias']:
+            name = '%s - %s' % (name, ';'.join(info['artist']['alias']))
+
+        entries = [
+            self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
+                            'NetEaseMusic', song['id'])
+            for song in info['hotSongs']
+        ]
+        return self.playlist_result(entries, singer_id, name)
+
+
+class NetEaseMusicListIE(NetEaseMusicBaseIE):
+    IE_NAME = 'netease:playlist'
+    IE_DESC = '网易云音乐 - 歌单'
+    _VALID_URL = r'https?://music\.163\.com/(#/)?(playlist|discover/toplist)\?id=(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://music.163.com/#/playlist?id=79177352',
+        'info_dict': {
+            'id': '79177352',
+            'title': 'Billboard 2007 Top 100',
+            'description': 'md5:12fd0819cab2965b9583ace0f8b7b022'
+        },
+        'playlist_count': 99,
+    }, {
+        'note': 'Toplist/Charts sample',
+        'url': 'http://music.163.com/#/discover/toplist?id=3733003',
+        'info_dict': {
+            'id': '3733003',
+            'title': 're:韩国Melon排行榜周榜 [0-9]{4}-[0-9]{2}-[0-9]{2}',
+            'description': 'md5:73ec782a612711cadc7872d9c1e134fc',
+        },
+        'playlist_count': 50,
+    }]
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+
+        info = self.query_api(
+            'playlist/detail?id=%s&lv=-1&tv=-1' % list_id,
+            list_id, 'Downloading playlist data')['result']
+
+        name = info['name']
+        desc = info.get('description')
+
+        if info.get('specialType') == 10:  # is a chart/toplist
+            datestamp = datetime.fromtimestamp(
+                self.convert_milliseconds(info['updateTime'])).strftime('%Y-%m-%d')
+            name = '%s %s' % (name, datestamp)
+
+        entries = [
+            self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
+                            'NetEaseMusic', song['id'])
+            for song in info['tracks']
+        ]
+        return self.playlist_result(entries, list_id, name, desc)
+
+
+class NetEaseMusicMvIE(NetEaseMusicBaseIE):
+    IE_NAME = 'netease:mv'
+    IE_DESC = '网易云音乐 - MV'
+    _VALID_URL = r'https?://music\.163\.com/(#/)?mv\?id=(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://music.163.com/#/mv?id=415350',
+        'info_dict': {
+            'id': '415350',
+            'ext': 'mp4',
+            'title': '이럴거면 그러지말지',
+            'description': '白雅言自作曲唱甜蜜爱情',
+            'creator': '白雅言',
+            'upload_date': '20150520',
+        },
+    }
+
+    def _real_extract(self, url):
+        mv_id = self._match_id(url)
+
+        info = self.query_api(
+            'mv/detail?id=%s&type=mp4' % mv_id,
+            mv_id, 'Downloading mv info')['data']
+
+        formats = [
+            {'url': mv_url, 'ext': 'mp4', 'format_id': '%sp' % brs, 'height': int(brs)}
+            for brs, mv_url in info['brs'].items()
+        ]
+        self._sort_formats(formats)
+
+        return {
+            'id': mv_id,
+            'title': info['name'],
+            'description': info.get('desc') or info.get('briefDesc'),
+            'creator': info['artistName'],
+            'upload_date': info['publishTime'].replace('-', ''),
+            'formats': formats,
+            'thumbnail': info.get('cover'),
+            'duration': self.convert_milliseconds(info.get('duration', 0)),
+        }
+
+
+class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
+    IE_NAME = 'netease:program'
+    IE_DESC = '网易云音乐 - 电台节目'
+    _VALID_URL = r'https?://music\.163\.com/(#/?)program\?id=(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://music.163.com/#/program?id=10109055',
+        'info_dict': {
+            'id': '10109055',
+            'ext': 'mp3',
+            'title': '不丹足球背后的故事',
+            'description': '喜马拉雅人的足球梦 ...',
+            'creator': '大话西藏',
+            'timestamp': 1434179342,
+            'upload_date': '20150613',
+            'duration': 900,
+        },
+    }, {
+        'note': 'This program has accompanying songs.',
+        'url': 'http://music.163.com/#/program?id=10141022',
+        'info_dict': {
+            'id': '10141022',
+            'title': '25岁，你是自在如风的少年<27°C>',
+            'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
+        },
+        'playlist_count': 4,
+    }, {
+        'note': 'This program has accompanying songs.',
+        'url': 'http://music.163.com/#/program?id=10141022',
+        'info_dict': {
+            'id': '10141022',
+            'ext': 'mp3',
+            'title': '25岁，你是自在如风的少年<27°C>',
+            'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
+            'timestamp': 1434450841,
+            'upload_date': '20150616',
+        },
+        'params': {
+            'noplaylist': True
+        }
+    }]
+
+    def _real_extract(self, url):
+        program_id = self._match_id(url)
+
+        info = self.query_api(
+            'dj/program/detail?id=%s' % program_id,
+            program_id, 'Downloading program info')['program']
+
+        name = info['name']
+        description = info['description']
+
+        if not info['songs'] or self._downloader.params.get('noplaylist'):
+            if info['songs']:
+                self.to_screen(
+                    'Downloading just the main audio %s because of --no-playlist'
+                    % info['mainSong']['id'])
+
+            formats = self.extract_formats(info['mainSong'])
+            self._sort_formats(formats)
+
+            return {
+                'id': program_id,
+                'title': name,
+                'description': description,
+                'creator': info['dj']['brand'],
+                'timestamp': self.convert_milliseconds(info['createTime']),
+                'thumbnail': info['coverUrl'],
+                'duration': self.convert_milliseconds(info.get('duration', 0)),
+                'formats': formats,
+            }
+
+        self.to_screen(
+            'Downloading playlist %s - add --no-playlist to just download the main audio %s'
+            % (program_id, info['mainSong']['id']))
+
+        song_ids = [info['mainSong']['id']]
+        song_ids.extend([song['id'] for song in info['songs']])
+        entries = [
+            self.url_result('http://music.163.com/#/song?id=%s' % song_id,
+                            'NetEaseMusic', song_id)
+            for song_id in song_ids
+        ]
+        return self.playlist_result(entries, program_id, name, description)
+
+
+class NetEaseMusicDjRadioIE(NetEaseMusicBaseIE):
+    IE_NAME = 'netease:djradio'
+    IE_DESC = '网易云音乐 - 电台'
+    _VALID_URL = r'https?://music\.163\.com/(#/)?djradio\?id=(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://music.163.com/#/djradio?id=42',
+        'info_dict': {
+            'id': '42',
+            'title': '声音蔓延',
+            'description': 'md5:766220985cbd16fdd552f64c578a6b15'
+        },
+        'playlist_mincount': 40,
+    }
+    _PAGE_SIZE = 1000
+
+    def _real_extract(self, url):
+        dj_id = self._match_id(url)
+
+        name = None
+        desc = None
+        entries = []
+        for offset in compat_itertools_count(start=0, step=self._PAGE_SIZE):
+            info = self.query_api(
+                'dj/program/byradio?asc=false&limit=%d&radioId=%s&offset=%d'
+                % (self._PAGE_SIZE, dj_id, offset),
+                dj_id, 'Downloading dj programs - %d' % offset)
+
+            entries.extend([
+                self.url_result(
+                    'http://music.163.com/#/program?id=%s' % program['id'],
+                    'NetEaseMusicProgram', program['id'])
+                for program in info['programs']
+            ])
+
+            if name is None:
+                radio = info['programs'][0]['radio']
+                name = radio['name']
+                desc = radio['desc']
+
+            if not info['more']:
+                break
+
+        return self.playlist_result(entries, dj_id, name, desc)

youtube_dl/extractor/newstube.py

@@ -31,7 +31,7 @@ class NewstubeIE(InfoExtractor):
         page = self._download_webpage(url, video_id, 'Downloading page')
 
         video_guid = self._html_search_regex(
-            r'<meta property="og:video" content="https?://(?:www\.)?newstube\.ru/freshplayer\.swf\?guid=(?P<guid>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
+            r'<meta property="og:video:url" content="https?://(?:www\.)?newstube\.ru/freshplayer\.swf\?guid=(?P<guid>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
             page, 'video GUID')
 
         player = self._download_xml(

youtube_dl/extractor/nextmedia.py

@@ -6,6 +6,7 @@ from ..utils import parse_iso8601
 
 
 class NextMediaIE(InfoExtractor):
+    IE_DESC = '蘋果日報'
     _VALID_URL = r'http://hk.apple.nextmedia.com/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)'
     _TESTS = [{
         'url': 'http://hk.apple.nextmedia.com/realtime/news/20141108/53109199',
@@ -66,6 +67,7 @@ class NextMediaIE(InfoExtractor):
 
 
 class NextMediaActionNewsIE(NextMediaIE):
+    IE_DESC = '蘋果日報 - 動新聞'
     _VALID_URL = r'http://hk.dv.nextmedia.com/actionnews/[^/]+/(?P<date>\d+)/(?P<id>\d+)/\d+'
     _TESTS = [{
         'url': 'http://hk.dv.nextmedia.com/actionnews/hit/20150121/19009428/20061460',
@@ -90,6 +92,7 @@ class NextMediaActionNewsIE(NextMediaIE):
 
 
 class AppleDailyIE(NextMediaIE):
+    IE_DESC = '臺灣蘋果日報'
     _VALID_URL = r'http://(www|ent).appledaily.com.tw/(?:animation|appledaily|enews|realtimenews)/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)(/.*)?'
     _TESTS = [{
         'url': 'http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694',

youtube_dl/extractor/noco.py

@@ -195,7 +195,7 @@ class NocoIE(InfoExtractor):
         if episode_number:
             title += ' #' + compat_str(episode_number)
         if episode:
-            title += ' - ' + episode
+            title += ' - ' + compat_str(episode)
 
         description = show.get('show_resume') or show.get('family_resume')
 

youtube_dl/extractor/nowtv.py

@@ -133,7 +133,7 @@ class NowTVIE(InfoExtractor):
         station = mobj.group('station')
 
         info = self._download_json(
-            'https://api.nowtv.de/v3/movies/%s?fields=*,format,files' % display_id,
+            'https://api.nowtv.de/v3/movies/%s?fields=id,title,free,geoblocked,articleLong,articleShort,broadcastStartDate,seoUrl,duration,format,files' % display_id,
             display_id)
 
         video_id = compat_str(info['id'])

youtube_dl/extractor/npo.py

@@ -1,5 +1,7 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     fix_xml_ampersands,
@@ -7,7 +9,6 @@ from ..utils import (
     qualities,
     strip_jsonp,
     unified_strdate,
-    url_basename,
 )
 
 
@@ -16,13 +17,42 @@ class NPOBaseIE(InfoExtractor):
         token_page = self._download_webpage(
             'http://ida.omroep.nl/npoplayer/i.js',
             video_id, note='Downloading token')
-        return self._search_regex(
+        token = self._search_regex(
             r'npoplayer\.token = "(.+?)"', token_page, 'token')
+        # Decryption algorithm extracted from http://npoplayer.omroep.nl/csjs/npoplayer-min.js
+        token_l = list(token)
+        first = second = None
+        for i in range(5, len(token_l) - 4):
+            if token_l[i].isdigit():
+                if first is None:
+                    first = i
+                elif second is None:
+                    second = i
+        if first is None or second is None:
+            first = 12
+            second = 13
+
+        token_l[first], token_l[second] = token_l[second], token_l[first]
+
+        return ''.join(token_l)
 
 
 class NPOIE(NPOBaseIE):
-    IE_NAME = 'npo.nl'
-    _VALID_URL = r'https?://(?:www\.)?npo\.nl/(?!live|radio)[^/]+/[^/]+/(?P<id>[^/?]+)'
+    IE_NAME = 'npo'
+    IE_DESC = 'npo.nl and ntr.nl'
+    _VALID_URL = r'''(?x)
+                    (?:
+                        npo:|
+                        https?://
+                            (?:www\.)?
+                            (?:
+                                npo\.nl/(?!live|radio)(?:[^/]+/){2}|
+                                ntr\.nl/(?:[^/]+/){2,}|
+                                omroepwnl\.nl/video/fragment/[^/]+__
+                            )
+                        )
+                        (?P<id>[^/?#]+)
+                '''
 
     _TESTS = [
         {
@@ -42,7 +72,7 @@ class NPOIE(NPOBaseIE):
             'info_dict': {
                 'id': 'VARA_101191800',
                 'ext': 'm4v',
-                'title': 'De Mega Mike & Mega Thomas show',
+                'title': 'De Mega Mike & Mega Thomas show: The best of.',
                 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
                 'upload_date': '20090227',
                 'duration': 2400,
@@ -54,8 +84,8 @@ class NPOIE(NPOBaseIE):
             'info_dict': {
                 'id': 'VPWON_1169289',
                 'ext': 'm4v',
-                'title': 'Tegenlicht',
-                'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
+                'title': 'Tegenlicht: De toekomst komt uit Afrika',
+                'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
                 'upload_date': '20130225',
                 'duration': 3000,
             },
@@ -84,6 +114,30 @@ class NPOIE(NPOBaseIE):
                 'title': 'Hoe gaat Europa verder na Parijs?',
             },
         },
+        {
+            'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
+            'md5': '01c6a2841675995da1f0cf776f03a9c3',
+            'info_dict': {
+                'id': 'VPWON_1233944',
+                'ext': 'm4v',
+                'title': 'Aap, poot, pies',
+                'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
+                'upload_date': '20150508',
+                'duration': 599,
+            },
+        },
+        {
+            'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
+            'md5': 'd30cd8417b8b9bca1fdff27428860d08',
+            'info_dict': {
+                'id': 'POW_00996502',
+                'ext': 'm4v',
+                'title': '''"Dit is wel een 'landslide'..."''',
+                'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
+                'upload_date': '20150508',
+                'duration': 462,
+            },
+        }
     ]
 
     def _real_extract(self, url):
@@ -92,12 +146,24 @@ class NPOIE(NPOBaseIE):
 
     def _get_info(self, video_id):
         metadata = self._download_json(
-            'http://e.omroep.nl/metadata/aflevering/%s' % video_id,
+            'http://e.omroep.nl/metadata/%s' % video_id,
             video_id,
             # We have to remove the javascript callback
             transform_source=strip_jsonp,
         )
 
+        # For some videos actual video id (prid) is different (e.g. for
+        # http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698
+        # video id is POMS_WNL_853698 but prid is POW_00996502)
+        video_id = metadata.get('prid') or video_id
+
+        # titel is too generic in some cases so utilize aflevering_titel as well
+        # when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html)
+        title = metadata['titel']
+        sub_title = metadata.get('aflevering_titel')
+        if sub_title and sub_title != title:
+            title += ': %s' % sub_title
+
         token = self._get_token(video_id)
 
         formats = []
@@ -170,8 +236,8 @@ class NPOIE(NPOBaseIE):
 
         return {
             'id': video_id,
-            'title': metadata['titel'],
-            'description': metadata['info'],
+            'title': title,
+            'description': metadata.get('info'),
             'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
             'upload_date': unified_strdate(metadata.get('gidsdatum')),
             'duration': parse_duration(metadata.get('tijdsduur')),
@@ -340,9 +406,8 @@ class NPORadioFragmentIE(InfoExtractor):
         }
 
 
-class TegenlichtVproIE(NPOIE):
-    IE_NAME = 'tegenlicht.vpro.nl'
-    _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'
+class VPROIE(NPOIE):
+    _VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html'
 
     _TESTS = [
         {
@@ -351,17 +416,72 @@ class TegenlichtVproIE(NPOIE):
             'info_dict': {
                 'id': 'VPWON_1169289',
                 'ext': 'm4v',
-                'title': 'Tegenlicht',
-                'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
+                'title': 'De toekomst komt uit Afrika',
+                'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
                 'upload_date': '20130225',
             },
         },
+        {
+            'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
+            'info_dict': {
+                'id': 'sergio-herman',
+                'title': 'Sergio Herman: Fucking perfect',
+            },
+            'playlist_count': 2,
+        },
+        {
+            # playlist with youtube embed
+            'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html',
+            'info_dict': {
+                'id': 'education-education',
+                'title': '2Doc',
+            },
+            'playlist_count': 2,
+        }
     ]
 
     def _real_extract(self, url):
-        name = url_basename(url)
-        webpage = self._download_webpage(url, name)
-        urn = self._html_search_meta('mediaurn', webpage)
-        info_page = self._download_json(
-            'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name)
-        return self._get_info(info_page['mid'])
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        entries = [
+            self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id)
+            for video_id in re.findall(r'data-media-id="([^"]+)"', webpage)
+        ]
+
+        playlist_title = self._search_regex(
+            r'<title>\s*([^>]+?)\s*-\s*Teledoc\s*-\s*VPRO\s*</title>',
+            webpage, 'playlist title', default=None) or self._og_search_title(webpage)
+
+        return self.playlist_result(entries, playlist_id, playlist_title)
+
+
+class WNLIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
+
+    _TEST = {
+        'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
+        'info_dict': {
+            'id': 'vandaag-de-dag-6-mei',
+            'title': 'Vandaag de Dag 6 mei',
+        },
+        'playlist_count': 4,
+    }
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        entries = [
+            self.url_result('npo:%s' % video_id, 'NPO')
+            for video_id, part in re.findall(
+                r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>(Deel \d+)', webpage)
+        ]
+
+        playlist_title = self._html_search_regex(
+            r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>',
+            webpage, 'playlist title')
+
+        return self.playlist_result(entries, playlist_id, playlist_title)

youtube_dl/extractor/nrk.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 
 class NRKIE(InfoExtractor):
-    _VALID_URL = r'(?:nrk:|http://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
+    _VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
 
     _TESTS = [
         {
@@ -76,7 +76,7 @@ class NRKIE(InfoExtractor):
 
 
 class NRKPlaylistIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?nrk\.no/(?!video)(?:[^/]+/)+(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video)(?:[^/]+/)+(?P<id>[^/]+)'
 
     _TESTS = [{
         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
@@ -116,11 +116,12 @@ class NRKPlaylistIE(InfoExtractor):
 
 
 class NRKTVIE(InfoExtractor):
-    _VALID_URL = r'(?P<baseurl>http://tv\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
+    IE_DESC = 'NRK TV and NRK Radio'
+    _VALID_URL = r'(?P<baseurl>https?://(?:tv|radio)\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
 
     _TESTS = [
         {
-            'url': 'http://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
+            'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
             'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
             'info_dict': {
                 'id': 'MUHH48000314',
@@ -132,7 +133,7 @@ class NRKTVIE(InfoExtractor):
             },
         },
         {
-            'url': 'http://tv.nrk.no/program/mdfp15000514',
+            'url': 'https://tv.nrk.no/program/mdfp15000514',
             'md5': '383650ece2b25ecec996ad7b5bb2a384',
             'info_dict': {
                 'id': 'mdfp15000514',
@@ -145,7 +146,7 @@ class NRKTVIE(InfoExtractor):
         },
         {
             # single playlist video
-            'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
+            'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
             'md5': 'adbd1dbd813edaf532b0a253780719c2',
             'info_dict': {
                 'id': 'MSPO40010515-part2',
@@ -157,7 +158,7 @@ class NRKTVIE(InfoExtractor):
             'skip': 'Only works from Norway',
         },
         {
-            'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
+            'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
             'playlist': [
                 {
                     'md5': '9480285eff92d64f06e02a5367970a7a',
@@ -188,6 +189,10 @@ class NRKTVIE(InfoExtractor):
                 'duration': 6947.5199999999995,
             },
             'skip': 'Only works from Norway',
+        },
+        {
+            'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
+            'only_matching': True,
         }
     ]
 
@@ -206,7 +211,8 @@ class NRKTVIE(InfoExtractor):
         ]}
 
     def _extract_f4m(self, manifest_url, video_id):
-        return self._extract_f4m_formats(manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id)
+        return self._extract_f4m_formats(
+            manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id, f4m_id='hds')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -268,7 +274,7 @@ class NRKTVIE(InfoExtractor):
 
         m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
         if m3u8_url:
-            formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4'))
+            formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4', m3u8_id='hls'))
         self._sort_formats(formats)
 
         subtitles_url = self._html_search_regex(

youtube_dl/extractor/odnoklassniki.py

@@ -2,7 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     unified_strdate,
     int_or_none,
@@ -62,7 +62,7 @@ class OdnoklassnikiIE(InfoExtractor):
             metadata = self._parse_json(metadata, video_id)
         else:
             metadata = self._download_json(
-                compat_urllib_parse.unquote(flashvars['metadataUrl']),
+                compat_urllib_parse_unquote(flashvars['metadataUrl']),
                 video_id, 'Downloading metadata JSON')
 
         movie = metadata['movie']

youtube_dl/extractor/onionstudios.py

@@ -0,0 +1,76 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import determine_ext
+
+
+class OnionStudiosIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?onionstudios\.com/(?:videos/[^/]+-|embed\?.*\bid=)(?P<id>\d+)(?!-)'
+
+    _TESTS = [{
+        'url': 'http://www.onionstudios.com/videos/hannibal-charges-forward-stops-for-a-cocktail-2937',
+        'md5': 'd4851405d31adfadf71cd7a487b765bb',
+        'info_dict': {
+            'id': '2937',
+            'ext': 'mp4',
+            'title': 'Hannibal charges forward, stops for a cocktail',
+            'description': 'md5:545299bda6abf87e5ec666548c6a9448',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'The A.V. Club',
+            'uploader_id': 'TheAVClub',
+        },
+    }, {
+        'url': 'http://www.onionstudios.com/embed?id=2855&autoplay=true',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?onionstudios\.com/embed.+?)\1', webpage)
+        if mobj:
+            return mobj.group('url')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://www.onionstudios.com/embed?id=%s' % video_id, video_id)
+
+        formats = []
+        for src in re.findall(r'<source[^>]+src="([^"]+)"', webpage):
+            if determine_ext(src) != 'm3u8':  # m3u8 always results in 403
+                formats.append({
+                    'url': src,
+                })
+        self._sort_formats(formats)
+
+        title = self._search_regex(
+            r'share_title\s*=\s*(["\'])(?P<title>[^\1]+?)\1',
+            webpage, 'title', group='title')
+        description = self._search_regex(
+            r'share_description\s*=\s*(["\'])(?P<description>[^\1]+?)\1',
+            webpage, 'description', default=None, group='description')
+        thumbnail = self._search_regex(
+            r'poster\s*=\s*(["\'])(?P<thumbnail>[^\1]+?)\1',
+            webpage, 'thumbnail', default=False, group='thumbnail')
+
+        uploader_id = self._search_regex(
+            r'twitter_handle\s*=\s*(["\'])(?P<uploader_id>[^\1]+?)\1',
+            webpage, 'uploader id', fatal=False, group='uploader_id')
+        uploader = self._search_regex(
+            r'window\.channelName\s*=\s*(["\'])Embedded:(?P<uploader>[^\1]+?)\1',
+            webpage, 'uploader', default=False, group='uploader')
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'formats': formats,
+        }

youtube_dl/extractor/openfilm.py

@@ -3,9 +3,9 @@ from __future__ import unicode_literals
 import json
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote_plus
 from ..utils import (
     parse_iso8601,
-    compat_urllib_parse,
     parse_age_limit,
     int_or_none,
 )
@@ -37,7 +37,7 @@ class OpenFilmIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        player = compat_urllib_parse.unquote_plus(
+        player = compat_urllib_parse_unquote_plus(
             self._og_search_video_url(webpage))
 
         video = json.loads(self._search_regex(

youtube_dl/extractor/pbs.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -31,10 +32,13 @@ class PBSIE(InfoExtractor):
             'info_dict': {
                 'id': '2365006249',
                 'ext': 'mp4',
-                'title': 'A More Perfect Union',
+                'title': 'Constitution USA with Peter Sagal - A More Perfect Union',
                 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
                 'duration': 3190,
             },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
@@ -42,10 +46,13 @@ class PBSIE(InfoExtractor):
             'info_dict': {
                 'id': '2365297690',
                 'ext': 'mp4',
-                'title': 'Losing Iraq',
+                'title': 'FRONTLINE - Losing Iraq',
                 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
                 'duration': 5050,
             },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            }
         },
         {
             'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
@@ -53,7 +60,7 @@ class PBSIE(InfoExtractor):
             'info_dict': {
                 'id': '2201174722',
                 'ext': 'mp4',
-                'title': 'Cyber Schools Gain Popularity, but Quality Questions Persist',
+                'title': 'PBS NewsHour - Cyber Schools Gain Popularity, but Quality Questions Persist',
                 'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28',
                 'duration': 801,
             },
@@ -65,10 +72,13 @@ class PBSIE(InfoExtractor):
                 'id': '2365297708',
                 'ext': 'mp4',
                 'description': 'md5:68d87ef760660eb564455eb30ca464fe',
-                'title': 'Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full',
+                'title': 'Great Performances - Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full',
                 'duration': 6559,
                 'thumbnail': 're:^https?://.*\.jpg$',
-            }
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/nova/earth/killer-typhoon.html',
@@ -78,11 +88,14 @@ class PBSIE(InfoExtractor):
                 'display_id': 'killer-typhoon',
                 'ext': 'mp4',
                 'description': 'md5:c741d14e979fc53228c575894094f157',
-                'title': 'Killer Typhoon',
+                'title': 'NOVA - Killer Typhoon',
                 'duration': 3172,
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'upload_date': '20140122',
-            }
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/',
@@ -90,6 +103,36 @@ class PBSIE(InfoExtractor):
                 'id': 'united-states-of-secrets',
             },
             'playlist_count': 2,
+        },
+        {
+            'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/',
+            'info_dict': {
+                'id': '2280706814',
+                'display_id': 'player',
+                'ext': 'mp4',
+                'title': 'American Experience - Death and the Civil War',
+                'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.',
+                'duration': 6705,
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
+        },
+        {
+            'url': 'http://video.pbs.org/video/2365367186/',
+            'info_dict': {
+                'id': '2365367186',
+                'display_id': '2365367186',
+                'ext': 'mp4',
+                'title': 'To Catch A Comet - Full Episode',
+                'description': 'On November 12, 2014, billions of kilometers from Earth, spacecraft orbiter Rosetta and lander Philae did what no other had dared to attempt \u2014 land on the volatile surface of a comet as it zooms around the sun at 67,000 km/hr. The European Space Agency hopes this mission can help peer into our past and unlock secrets of our origins.',
+                'duration': 3342,
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         }
     ]
 
@@ -123,7 +166,7 @@ class PBSIE(InfoExtractor):
                 return media_id, presumptive_id, upload_date
 
             url = self._search_regex(
-                r'<iframe\s+(?:class|id)=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>',
+                r'<iframe\s+[^>]*\s+src=["\']([^\'"]+partnerplayer[^\'"]+)["\']',
                 webpage, 'player URL')
             mobj = re.match(self._VALID_URL, url)
 
@@ -196,6 +239,20 @@ class PBSIE(InfoExtractor):
             rating_str = rating_str.rpartition('-')[2]
         age_limit = US_RATINGS.get(rating_str)
 
+        subtitles = {}
+        closed_captions_url = info.get('closed_captions_url')
+        if closed_captions_url:
+            subtitles['en'] = [{
+                'ext': 'ttml',
+                'url': closed_captions_url,
+            }]
+
+        # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc)
+        # Try turning it to 'program - title' naming scheme if possible
+        alt_title = info.get('program', {}).get('title')
+        if alt_title:
+            info['title'] = alt_title + ' - ' + re.sub(r'^' + alt_title + '[\s\-:]+', '', info['title'])
+
         return {
             'id': video_id,
             'display_id': display_id,
@@ -206,4 +263,5 @@ class PBSIE(InfoExtractor):
             'age_limit': age_limit,
             'upload_date': upload_date,
             'formats': formats,
+            'subtitles': subtitles,
         }

youtube_dl/extractor/photobucket.py

@@ -4,7 +4,7 @@ import json
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse
+from ..compat import compat_urllib_parse_unquote
 
 
 class PhotobucketIE(InfoExtractor):
@@ -34,7 +34,7 @@ class PhotobucketIE(InfoExtractor):
         info_json = self._search_regex(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (.*?)\);',
                                        webpage, 'info json')
         info = json.loads(info_json)
-        url = compat_urllib_parse.unquote(self._html_search_regex(r'file=(.+\.mp4)', info['linkcodes']['html'], 'url'))
+        url = compat_urllib_parse_unquote(self._html_search_regex(r'file=(.+\.mp4)', info['linkcodes']['html'], 'url'))
         return {
             'id': video_id,
             'url': url,

youtube_dl/extractor/pinkbike.py

@@ -0,0 +1,96 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    remove_end,
+    remove_start,
+    str_to_int,
+    unified_strdate,
+)
+
+
+class PinkbikeIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:(?:www\.)?pinkbike\.com/video/|es\.pinkbike\.org/i/kvid/kvid-y5\.swf\?id=)(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://www.pinkbike.com/video/402811/',
+        'md5': '4814b8ca7651034cd87e3361d5c2155a',
+        'info_dict': {
+            'id': '402811',
+            'ext': 'mp4',
+            'title': 'Brandon Semenuk - RAW 100',
+            'description': 'Official release: www.redbull.ca/rupertwalker',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 100,
+            'upload_date': '20150406',
+            'uploader': 'revelco',
+            'location': 'Victoria, British Columbia, Canada',
+            'view_count': int,
+            'comment_count': int,
+        }
+    }, {
+        'url': 'http://es.pinkbike.org/i/kvid/kvid-y5.swf?id=406629',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://www.pinkbike.com/video/%s' % video_id, video_id)
+
+        formats = []
+        for _, format_id, src in re.findall(
+                r'data-quality=((?:\\)?["\'])(.+?)\1[^>]+src=\1(.+?)\1', webpage):
+            height = int_or_none(self._search_regex(
+                r'^(\d+)[pP]$', format_id, 'height', default=None))
+            formats.append({
+                'url': src,
+                'format_id': format_id,
+                'height': height,
+            })
+        self._sort_formats(formats)
+
+        title = remove_end(self._og_search_title(webpage), ' Video - Pinkbike')
+        description = self._html_search_regex(
+            r'(?s)id="media-description"[^>]*>(.+?)<',
+            webpage, 'description', default=None) or remove_start(
+            self._og_search_description(webpage), title + '. ')
+        thumbnail = self._og_search_thumbnail(webpage)
+        duration = int_or_none(self._html_search_meta(
+            'video:duration', webpage, 'duration'))
+
+        uploader = self._search_regex(
+            r'un:\s*"([^"]+)"', webpage, 'uploader', fatal=False)
+        upload_date = unified_strdate(self._search_regex(
+            r'class="fullTime"[^>]+title="([^"]+)"',
+            webpage, 'upload date', fatal=False))
+
+        location = self._html_search_regex(
+            r'(?s)<dt>Location</dt>\s*<dd>(.+?)<',
+            webpage, 'location', fatal=False)
+
+        def extract_count(webpage, label):
+            return str_to_int(self._search_regex(
+                r'<span[^>]+class="stat-num"[^>]*>([\d,.]+)</span>\s*<span[^>]+class="stat-label"[^>]*>%s' % label,
+                webpage, label, fatal=False))
+
+        view_count = extract_count(webpage, 'Views')
+        comment_count = extract_count(webpage, 'Comments')
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'upload_date': upload_date,
+            'uploader': uploader,
+            'location': location,
+            'view_count': view_count,
+            'comment_count': comment_count,
+            'formats': formats
+        }

youtube_dl/extractor/planetaplay.py

@@ -18,7 +18,8 @@ class PlanetaPlayIE(InfoExtractor):
             'id': '3586',
             'ext': 'flv',
             'title': 'md5:e829428ee28b1deed00de90de49d1da1',
-        }
+        },
+        'skip': 'Not accessible from Travis CI server',
     }
 
     _SONG_FORMATS = {

youtube_dl/extractor/played.py

@@ -38,9 +38,7 @@ class PlayedIE(InfoExtractor):
         if m_error:
             raise ExtractorError(m_error.group('msg'), expected=True)
 
-        fields = re.findall(
-            r'type="hidden" name="([^"]+)"\s+value="([^"]+)">', orig_webpage)
-        data = dict(fields)
+        data = self._hidden_inputs(orig_webpage)
 
         self._sleep(2, video_id)
 

youtube_dl/extractor/playvid.py

@@ -4,7 +4,8 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse,
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_unquote_plus,
 )
 from ..utils import (
     clean_html,
@@ -44,7 +45,7 @@ class PlayvidIE(InfoExtractor):
         flashvars = self._html_search_regex(
             r'flashvars="(.+?)"', webpage, 'flashvars')
 
-        infos = compat_urllib_parse.unquote(flashvars).split(r'&')
+        infos = compat_urllib_parse_unquote(flashvars).split(r'&')
         for info in infos:
             videovars_match = re.match(r'^video_vars\[(.+?)\]=(.+?)$', info)
             if videovars_match:
@@ -52,7 +53,7 @@ class PlayvidIE(InfoExtractor):
                 val = videovars_match.group(2)
 
                 if key == 'title':
-                    video_title = compat_urllib_parse.unquote_plus(val)
+                    video_title = compat_urllib_parse_unquote_plus(val)
                 if key == 'duration':
                     try:
                         duration = int(val)

youtube_dl/extractor/pornhub.py

@@ -5,7 +5,8 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse,
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_unquote_plus,
     compat_urllib_parse_urlparse,
     compat_urllib_request,
 )
@@ -19,8 +20,8 @@ from ..aes import (
 
 
 class PornHubIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-f]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-z]+)'
+    _TESTS = [{
         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
         'md5': '882f488fa1f0026f023f33576004a2ed',
         'info_dict': {
@@ -30,7 +31,10 @@ class PornHubIE(InfoExtractor):
             "title": "Seductive Indian beauty strips down and fingers her pink pussy",
             "age_limit": 18
         }
-    }
+    }, {
+        'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
+        'only_matching': True,
+    }]
 
     @classmethod
     def _extract_url(cls, webpage):
@@ -66,7 +70,7 @@ class PornHubIE(InfoExtractor):
             webpage, 'uploader', fatal=False)
         thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, 'thumbnail', fatal=False)
         if thumbnail:
-            thumbnail = compat_urllib_parse.unquote(thumbnail)
+            thumbnail = compat_urllib_parse_unquote(thumbnail)
 
         view_count = self._extract_count(
             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
@@ -77,9 +81,9 @@ class PornHubIE(InfoExtractor):
         comment_count = self._extract_count(
             r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment')
 
-        video_urls = list(map(compat_urllib_parse.unquote, re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage)))
+        video_urls = list(map(compat_urllib_parse_unquote, re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage)))
         if webpage.find('"encrypted":true') != -1:
-            password = compat_urllib_parse.unquote_plus(
+            password = compat_urllib_parse_unquote_plus(
                 self._search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
 

youtube_dl/extractor/primesharetv.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
@@ -31,12 +29,7 @@ class PrimeShareTVIE(InfoExtractor):
         if '>File not exist<' in webpage:
             raise ExtractorError('Video %s does not exist' % video_id, expected=True)
 
-        fields = dict(re.findall(r'''(?x)<input\s+
-            type="hidden"\s+
-            name="([^"]+)"\s+
-            (?:id="[^"]+"\s+)?
-            value="([^"]*)"
-            ''', webpage))
+        fields = self._hidden_inputs(webpage)
 
         headers = {
             'Referer': url,

youtube_dl/extractor/promptfile.py

@@ -35,10 +35,7 @@ class PromptFileIE(InfoExtractor):
             raise ExtractorError('Video %s does not exist' % video_id,
                                  expected=True)
 
-        fields = dict(re.findall(r'''(?x)type="hidden"\s+
-            name="(.+?)"\s+
-            value="(.*?)"
-            ''', webpage))
+        fields = self._hidden_inputs(webpage)
         post = compat_urllib_parse.urlencode(fields)
         req = compat_urllib_request.Request(url, post)
         req.add_header('Content-type', 'application/x-www-form-urlencoded')

youtube_dl/extractor/prosiebensat1.py

@@ -9,8 +9,11 @@ from ..compat import (
     compat_urllib_parse,
 )
 from ..utils import (
-    unified_strdate,
+    ExtractorError,
+    determine_ext,
+    float_or_none,
     int_or_none,
+    unified_strdate,
 )
 
 
@@ -21,6 +24,11 @@ class ProSiebenSat1IE(InfoExtractor):
 
     _TESTS = [
         {
+            # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
+            # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
+            # - malformed f4m manifest support
+            # - proper handling of URLs starting with `https?://` in 2.0 manifests
+            # - recursive child f4m manifests extraction
             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
             'info_dict': {
                 'id': '2104602',
@@ -208,7 +216,7 @@ class ProSiebenSat1IE(InfoExtractor):
         clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id')
 
         access_token = 'prosieben'
-        client_name = 'kolibri-1.12.6'
+        client_name = 'kolibri-2.0.19-splec4'
         client_location = url
 
         videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
@@ -218,10 +226,13 @@ class ProSiebenSat1IE(InfoExtractor):
             'ids': clip_id,
         })
 
-        videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
+        video = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')[0]
 
-        duration = float(videos[0]['duration'])
-        source_ids = [source['id'] for source in videos[0]['sources']]
+        if video.get('is_protected') is True:
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        duration = float_or_none(video.get('duration'))
+        source_ids = [source['id'] for source in video['sources']]
         source_ids_str = ','.join(map(str, source_ids))
 
         g = '01!8d8F_)r9]4s[qeuXfP%'
@@ -275,8 +286,9 @@ class ProSiebenSat1IE(InfoExtractor):
 
         for source in urls_sources:
             protocol = source['protocol']
+            source_url = source['url']
             if protocol == 'rtmp' or protocol == 'rtmpe':
-                mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source['url'])
+                mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
                 if not mobj:
                     continue
                 path = mobj.group('path')
@@ -293,9 +305,11 @@ class ProSiebenSat1IE(InfoExtractor):
                     'ext': 'mp4',
                     'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
                 })
+            elif 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
+                formats.extend(self._extract_f4m_formats(source_url, clip_id))
             else:
                 formats.append({
-                    'url': source['url'],
+                    'url': source_url,
                     'vbr': fix_bitrate(source['bitrate']),
                 })
 

youtube_dl/extractor/qqmusic.py

@@ -9,12 +9,14 @@ from .common import InfoExtractor
 from ..utils import (
     strip_jsonp,
     unescapeHTML,
+    clean_html,
 )
 from ..compat import compat_urllib_request
 
 
 class QQMusicIE(InfoExtractor):
     IE_NAME = 'qqmusic'
+    IE_DESC = 'QQ音乐'
     _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
     _TESTS = [{
         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
@@ -26,6 +28,20 @@ class QQMusicIE(InfoExtractor):
             'upload_date': '20141227',
             'creator': '林俊杰',
             'description': 'md5:d327722d0361576fde558f1ac68a7065',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }, {
+        'note': 'There is no mp3-320 version of this song.',
+        'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV',
+        'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
+        'info_dict': {
+            'id': '004MsGEo3DdNxV',
+            'ext': 'mp3',
+            'title': '如果',
+            'upload_date': '20050626',
+            'creator': '李季美',
+            'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
+            'thumbnail': 're:^https?://.*\.jpg$',
         }
     }]
 
@@ -68,6 +84,14 @@ class QQMusicIE(InfoExtractor):
         if lrc_content:
             lrc_content = lrc_content.replace('\\n', '\n')
 
+        thumbnail_url = None
+        albummid = self._search_regex(
+            [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
+            detail_info_page, 'album mid', default=None)
+        if albummid:
+            thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
+                            % (albummid[-2:-1], albummid[-1], albummid)
+
         guid = self.m_r_get_ruin()
 
         vkey = self._download_json(
@@ -85,6 +109,7 @@ class QQMusicIE(InfoExtractor):
                 'preference': details['preference'],
                 'abr': details.get('abr'),
             })
+        self._check_formats(formats, mid)
         self._sort_formats(formats)
 
         return {
@@ -94,6 +119,7 @@ class QQMusicIE(InfoExtractor):
             'upload_date': publish_time,
             'creator': singer,
             'description': lrc_content,
+            'thumbnail': thumbnail_url,
         }
 
 
@@ -117,6 +143,7 @@ class QQPlaylistBaseIE(InfoExtractor):
 
 class QQMusicSingerIE(QQPlaylistBaseIE):
     IE_NAME = 'qqmusic:singer'
+    IE_DESC = 'QQ音乐 - 歌手'
     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
     _TEST = {
         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
@@ -161,39 +188,50 @@ class QQMusicSingerIE(QQPlaylistBaseIE):
 
 class QQMusicAlbumIE(QQPlaylistBaseIE):
     IE_NAME = 'qqmusic:album'
+    IE_DESC = 'QQ音乐 - 专辑'
     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 
-    _TEST = {
-        'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
+    _TESTS = [{
+        'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1',
         'info_dict': {
             'id': '000gXCTb2AhRR1',
             'title': '我们都是这样长大的',
-            'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
+            'description': 'md5:179c5dce203a5931970d306aa9607ea6',
         },
         'playlist_count': 4,
-    }
+    }, {
+        'url': 'http://y.qq.com/#type=album&mid=002Y5a3b3AlCu3',
+        'info_dict': {
+            'id': '002Y5a3b3AlCu3',
+            'title': '그리고...',
+            'description': 'md5:a48823755615508a95080e81b51ba729',
+        },
+        'playlist_count': 8,
+    }]
 
     def _real_extract(self, url):
         mid = self._match_id(url)
 
-        album_page = self._download_webpage(
-            self.qq_static_url('album', mid), mid, 'Download album page')
+        album = self._download_json(
+            'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
+            mid, 'Download album page')['data']
 
-        entries = self.get_entries_from_page(album_page)
-
-        album_name = self._html_search_regex(
-            r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
-            default=None)
-
-        album_detail = self._html_search_regex(
-            r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
-            album_page, 'album details', default=None)
+        entries = [
+            self.url_result(
+                'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
+            ) for song in album['list']
+        ]
+        album_name = album.get('name')
+        album_detail = album.get('desc')
+        if album_detail is not None:
+            album_detail = album_detail.strip()
 
         return self.playlist_result(entries, mid, album_name, album_detail)
 
 
 class QQMusicToplistIE(QQPlaylistBaseIE):
     IE_NAME = 'qqmusic:toplist'
+    IE_DESC = 'QQ音乐 - 排行榜'
     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 
     _TESTS = [{
@@ -243,3 +281,37 @@ class QQMusicToplistIE(QQPlaylistBaseIE):
         list_name = topinfo.get('ListName')
         list_description = topinfo.get('info')
         return self.playlist_result(entries, list_id, list_name, list_description)
+
+
+class QQMusicPlaylistIE(QQPlaylistBaseIE):
+    IE_NAME = 'qqmusic:playlist'
+    IE_DESC = 'QQ音乐 - 歌单'
+    _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
+
+    _TEST = {
+        'url': 'http://y.qq.com/#type=taoge&id=3462654915',
+        'info_dict': {
+            'id': '3462654915',
+            'title': '韩国5月新歌精选下旬',
+            'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
+        },
+        'playlist_count': 40,
+    }
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+
+        list_json = self._download_json(
+            'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg?type=1&json=1&utf8=1&onlysong=0&disstid=%s'
+            % list_id, list_id, 'Download list page',
+            transform_source=strip_jsonp)['cdlist'][0]
+
+        entries = [
+            self.url_result(
+                'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
+            ) for song in list_json['songlist']
+        ]
+
+        list_name = list_json.get('dissname')
+        list_description = clean_html(unescapeHTML(list_json.get('desc')))
+        return self.playlist_result(entries, list_id, list_name, list_description)

youtube_dl/extractor/quickvid.py

@@ -24,6 +24,7 @@ class QuickVidIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.(?:png|jpg|gif)$',
             'view_count': int,
         },
+        'skip': 'Not accessible from Travis CI server',
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/rds.py

@@ -0,0 +1,73 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    parse_iso8601,
+)
+
+
+class RDSIE(InfoExtractor):
+    IE_DESC = 'RDS.ca'
+    _VALID_URL = r'https?://(?:www\.)?rds\.ca/vid(?:[eé]|%C3%A9)os/(?:[^/]+/)*(?P<display_id>[^/]+)-(?P<id>\d+\.\d+)'
+
+    _TESTS = [{
+        'url': 'http://www.rds.ca/videos/football/nfl/fowler-jr-prend-la-direction-de-jacksonville-3.1132799',
+        'info_dict': {
+            'id': '3.1132799',
+            'display_id': 'fowler-jr-prend-la-direction-de-jacksonville',
+            'ext': 'mp4',
+            'title': 'Fowler Jr. prend la direction de Jacksonville',
+            'description': 'Dante Fowler Jr. est le troisième choix du repêchage 2015 de la NFL. ',
+            'timestamp': 1430397346,
+            'upload_date': '20150430',
+            'duration': 154.354,
+            'age_limit': 0,
+        }
+    }, {
+        'url': 'http://www.rds.ca/vid%C3%A9os/un-voyage-positif-3.877934',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        # TODO: extract f4m from 9c9media.com
+        video_url = self._search_regex(
+            r'<span[^>]+itemprop="contentURL"[^>]+content="([^"]+)"',
+            webpage, 'video url')
+
+        title = self._og_search_title(webpage) or self._html_search_meta(
+            'title', webpage, 'title', fatal=True)
+        description = self._og_search_description(webpage) or self._html_search_meta(
+            'description', webpage, 'description')
+        thumbnail = self._og_search_thumbnail(webpage) or self._search_regex(
+            [r'<link[^>]+itemprop="thumbnailUrl"[^>]+href="([^"]+)"',
+             r'<span[^>]+itemprop="thumbnailUrl"[^>]+content="([^"]+)"'],
+            webpage, 'thumbnail', fatal=False)
+        timestamp = parse_iso8601(self._search_regex(
+            r'<span[^>]+itemprop="uploadDate"[^>]+content="([^"]+)"',
+            webpage, 'upload date', fatal=False))
+        duration = parse_duration(self._search_regex(
+            r'<span[^>]+itemprop="duration"[^>]+content="([^"]+)"',
+            webpage, 'duration', fatal=False))
+        age_limit = self._family_friendly_search(webpage)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'duration': duration,
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/rtlnl.py

@@ -43,6 +43,25 @@ class RtlNlIE(InfoExtractor):
             'upload_date': '20150215',
             'description': 'Er zijn nieuwe beelden vrijgegeven die vlak na de aanslag in Kopenhagen zijn gemaakt. Op de video is goed te zien hoe omstanders zich bekommeren om één van de slachtoffers, terwijl de eerste agenten ter plaatse komen.',
         }
+    }, {
+        # empty synopsis and missing episodes (see https://github.com/rg3/youtube-dl/issues/6275)
+        'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a/autoplay=false',
+        'info_dict': {
+            'id': 'f536aac0-1dc3-4314-920e-3bd1c5b3811a',
+            'ext': 'mp4',
+            'title': 'RTL Nieuws - Meer beelden van overval juwelier',
+            'thumbnail': 're:^https?://screenshots\.rtl\.nl/system/thumb/sz=[0-9]+x[0-9]+/uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a$',
+            'timestamp': 1437233400,
+            'upload_date': '20150718',
+            'duration': 30.474,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # encrypted m3u8 streams, georestricted
+        'url': 'http://www.rtlxl.nl/#!/afl-2-257632/52a74543-c504-4cde-8aa8-ec66fe8d68a7',
+        'only_matching': True,
     }, {
         'url': 'http://www.rtl.nl/system/videoplayer/derden/embed.html#!/uuid=bb0353b0-d6a4-1dad-90e9-18fe75b8d1f0',
         'only_matching': True,
@@ -51,21 +70,33 @@ class RtlNlIE(InfoExtractor):
     def _real_extract(self, url):
         uuid = self._match_id(url)
         info = self._download_json(
-            'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=flash/' % uuid,
+            'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=adaptive/' % uuid,
             uuid)
 
         material = info['material'][0]
-        progname = info['abstracts'][0]['name']
-        subtitle = material['title'] or info['episodes'][0]['name']
-        description = material.get('synopsis') or info['episodes'][0]['synopsis']
+        title = info['abstracts'][0]['name']
+        subtitle = material.get('title')
+        if subtitle:
+            title += ' - %s' % subtitle
+        description = material.get('synopsis')
 
-        # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118)
-        videopath = material['videopath'].replace('.f4m', '.m3u8')
-        m3u8_url = 'http://manifest.us.rtl.nl' + videopath
+        meta = info.get('meta', {})
+
+        # m3u8 streams are encrypted and may not be handled properly by older ffmpeg/avconv.
+        # To workaround this previously adaptive -> flash trick was used to obtain
+        # unencrypted m3u8 streams (see https://github.com/rg3/youtube-dl/issues/4118)
+        # and bypass georestrictions as well.
+        # Currently, unencrypted m3u8 playlists are (intentionally?) invalid and therefore
+        # unusable albeit can be fixed by simple string replacement (see
+        # https://github.com/rg3/youtube-dl/pull/6337)
+        # Since recent ffmpeg and avconv handle encrypted streams just fine encrypted
+        # streams are used now.
+        videopath = material['videopath']
+        m3u8_url = meta.get('videohost', 'http://manifest.us.rtl.nl') + videopath
 
         formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4')
 
-        video_urlpart = videopath.split('/flash/')[1][:-5]
+        video_urlpart = videopath.split('/adaptive/')[1][:-5]
         PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4'
 
         formats.extend([
@@ -82,7 +113,7 @@ class RtlNlIE(InfoExtractor):
         self._sort_formats(formats)
 
         thumbnails = []
-        meta = info.get('meta', {})
+
         for p in ('poster_base_url', '"thumb_base_url"'):
             if not meta.get(p):
                 continue
@@ -98,7 +129,7 @@ class RtlNlIE(InfoExtractor):
 
         return {
             'id': uuid,
-            'title': '%s - %s' % (progname, subtitle),
+            'title': title,
             'formats': formats,
             'timestamp': material['original_date'],
             'description': description,

youtube_dl/extractor/rts.py

@@ -19,7 +19,16 @@ from ..utils import (
 
 class RTSIE(InfoExtractor):
     IE_DESC = 'RTS.ch'
-    _VALID_URL = r'https?://(?:www\.)?rts\.ch/(?:(?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html|play/tv/[^/]+/video/(?P<display_id_new>.+?)\?id=(?P<id_new>[0-9]+))'
+    _VALID_URL = r'''(?x)
+                    (?:
+                        rts:(?P<rts_id>\d+)|
+                        https?://
+                            (?:www\.)?rts\.ch/
+                            (?:
+                                (?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html|
+                                play/tv/[^/]+/video/(?P<display_id_new>.+?)\?id=(?P<id_new>[0-9]+)
+                            )
+                    )'''
 
     _TESTS = [
         {
@@ -122,6 +131,15 @@ class RTSIE(InfoExtractor):
                 'view_count': int,
             },
         },
+        {
+            # article with videos on rhs
+            'url': 'http://www.rts.ch/sport/hockey/6693917-hockey-davos-decroche-son-31e-titre-de-champion-de-suisse.html',
+            'info_dict': {
+                'id': '6693917',
+                'title': 'Hockey: Davos décroche son 31e titre de champion de Suisse',
+            },
+            'playlist_mincount': 5,
+        },
         {
             'url': 'http://www.rts.ch/play/tv/le-19h30/video/le-chantier-du-nouveau-parlement-vaudois-a-permis-une-trouvaille-historique?id=6348280',
             'only_matching': True,
@@ -130,7 +148,7 @@ class RTSIE(InfoExtractor):
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
-        video_id = m.group('id') or m.group('id_new')
+        video_id = m.group('rts_id') or m.group('id') or m.group('id_new')
         display_id = m.group('display_id') or m.group('display_id_new')
 
         def download_json(internal_id):
@@ -143,6 +161,15 @@ class RTSIE(InfoExtractor):
         # video_id extracted out of URL is not always a real id
         if 'video' not in all_info and 'audio' not in all_info:
             page = self._download_webpage(url, display_id)
+
+            # article with videos on rhs
+            videos = re.findall(
+                r'<article[^>]+class="content-item"[^>]*>\s*<a[^>]+data-video-urn="urn:rts:video:(\d+)"',
+                page)
+            if videos:
+                entries = [self.url_result('rts:%s' % video_urn, 'RTS') for video_urn in videos]
+                return self.playlist_result(entries, video_id, self._og_search_title(page))
+
             internal_id = self._html_search_regex(
                 r'<(?:video|audio) data-id="([0-9]+)"', page,
                 'internal video id')

youtube_dl/extractor/safari.py

@@ -83,7 +83,7 @@ class SafariIE(SafariBaseIE):
                                     library/view/[^/]+|
                                     api/v1/book
                                 )/
-                                (?P<course_id>\d+)/
+                                (?P<course_id>[^/]+)/
                                     (?:chapter(?:-content)?/)?
                                 (?P<part>part\d+)\.html
     '''
@@ -100,6 +100,10 @@ class SafariIE(SafariBaseIE):
     }, {
         'url': 'https://www.safaribooksonline.com/api/v1/book/9780133392838/chapter/part00.html',
         'only_matching': True,
+    }, {
+        # non-digits in course id
+        'url': 'https://www.safaribooksonline.com/library/view/create-a-nodejs/100000006A0210/part00.html',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -122,7 +126,7 @@ class SafariCourseIE(SafariBaseIE):
     IE_NAME = 'safari:course'
     IE_DESC = 'safaribooksonline.com online courses'
 
-    _VALID_URL = r'https?://(?:www\.)?safaribooksonline\.com/(?:library/view/[^/]+|api/v1/book)/(?P<id>\d+)/?(?:[#?]|$)'
+    _VALID_URL = r'https?://(?:www\.)?safaribooksonline\.com/(?:library/view/[^/]+|api/v1/book)/(?P<id>[^/]+)/?(?:[#?]|$)'
 
     _TESTS = [{
         'url': 'https://www.safaribooksonline.com/library/view/hadoop-fundamentals-livelessons/9780133392838/',

youtube_dl/extractor/sbs.py

@@ -1,17 +1,12 @@
 # -*- coding: utf-8 -*-
 from __future__ import unicode_literals
 
-import re
 from .common import InfoExtractor
-from ..utils import (
-    js_to_json,
-    remove_end,
-)
 
 
 class SBSIE(InfoExtractor):
     IE_DESC = 'sbs.com.au'
-    _VALID_URL = r'https?://(?:www\.)?sbs\.com\.au/ondemand/video/(?:single/)?(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?sbs\.com\.au/(?:ondemand|news)/video/(?:single/)?(?P<id>[0-9]+)'
 
     _TESTS = [{
         # Original URL is handled by the generic IE which finds the iframe:
@@ -21,39 +16,36 @@ class SBSIE(InfoExtractor):
         'info_dict': {
             'id': '320403011771',
             'ext': 'mp4',
-            'title': 'Dingo Conservation',
-            'description': 'Dingoes are on the brink of extinction; most of the animals we think are dingoes are in fact crossbred with wild dogs. This family run a dingo conservation park to prevent their extinction',
+            'title': 'Dingo Conservation (The Feed)',
+            'description': 'md5:f250a9856fca50d22dec0b5b8015f8a5',
             'thumbnail': 're:http://.*\.jpg',
+            'duration': 308,
         },
-        'add_ies': ['generic'],
     }, {
         'url': 'http://www.sbs.com.au/ondemand/video/320403011771/Dingo-Conservation-The-Feed',
         'only_matching': True,
+    }, {
+        'url': 'http://www.sbs.com.au/news/video/471395907773/The-Feed-July-9',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(
+            'http://www.sbs.com.au/ondemand/video/single/%s?context=web' % video_id, video_id)
 
-        player = self._search_regex(
-            r'(?s)playerParams\.releaseUrls\s*=\s*(\{.*?\n\});\n',
-            webpage, 'player')
-        player = re.sub(r"'\s*\+\s*[\da-zA-Z_]+\s*\+\s*'", '', player)
+        player_params = self._parse_json(
+            self._search_regex(
+                r'(?s)var\s+playerParams\s*=\s*({.+?});', webpage, 'playerParams'),
+            video_id)
 
-        release_urls = self._parse_json(js_to_json(player), video_id)
-
-        theplatform_url = release_urls.get('progressive') or release_urls['standard']
-
-        title = remove_end(self._og_search_title(webpage), ' (The Feed)')
-        description = self._html_search_meta('description', webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
+        urls = player_params['releaseUrls']
+        theplatform_url = (urls.get('progressive') or urls.get('standard') or
+                           urls.get('html') or player_params['relatedItemsURL'])
 
         return {
             '_type': 'url_transparent',
             'id': video_id,
             'url': theplatform_url,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
         }

youtube_dl/extractor/shared.py

@@ -1,6 +1,5 @@
 from __future__ import unicode_literals
 
-import re
 import base64
 
 from .common import InfoExtractor
@@ -35,8 +34,7 @@ class SharedIE(InfoExtractor):
             raise ExtractorError(
                 'Video %s does not exist' % video_id, expected=True)
 
-        download_form = dict(re.findall(
-            r'<input type="hidden" name="([^"]+)" value="([^"]*)"', webpage))
+        download_form = self._hidden_inputs(webpage)
         request = compat_urllib_request.Request(
             url, compat_urllib_parse.urlencode(download_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')

youtube_dl/extractor/smotri.py

@@ -53,7 +53,7 @@ class SmotriIE(InfoExtractor):
                 'thumbnail': 'http://frame4.loadup.ru/03/ed/57591.2.3.jpg',
             },
         },
-        # video-password
+        # video-password, not approved by moderator
         {
             'url': 'http://smotri.com/video/view/?id=v1390466a13c',
             'md5': 'f6331cef33cad65a0815ee482a54440b',
@@ -71,7 +71,24 @@ class SmotriIE(InfoExtractor):
             },
             'skip': 'Video is not approved by moderator',
         },
-        # age limit + video-password
+        # video-password
+        {
+            'url': 'http://smotri.com/video/view/?id=v6984858774#',
+            'md5': 'f11e01d13ac676370fc3b95b9bda11b0',
+            'info_dict': {
+                'id': 'v6984858774',
+                'ext': 'mp4',
+                'title': 'Дача Солженицина ПАРОЛЬ 223322',
+                'uploader': 'psavari1',
+                'uploader_id': 'psavari1',
+                'upload_date': '20081103',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                'videopassword': '223322',
+            },
+        },
+        # age limit + video-password, not approved by moderator
         {
             'url': 'http://smotri.com/video/view/?id=v15408898bcf',
             'md5': '91e909c9f0521adf5ee86fbe073aad70',
@@ -90,19 +107,22 @@ class SmotriIE(InfoExtractor):
             },
             'skip': 'Video is not approved by moderator',
         },
-        # not approved by moderator, but available
+        # age limit + video-password
         {
-            'url': 'http://smotri.com/video/view/?id=v28888533b73',
-            'md5': 'f44bc7adac90af518ef1ecf04893bb34',
+            'url': 'http://smotri.com/video/view/?id=v7780025814',
+            'md5': 'b4599b068422559374a59300c5337d72',
             'info_dict': {
-                'id': 'v28888533b73',
+                'id': 'v7780025814',
                 'ext': 'mp4',
-                'title': 'Russian Spies Killed By ISIL Child Soldier',
-                'uploader': 'Mopeder',
-                'uploader_id': 'mopeder',
-                'duration': 71,
-                'thumbnail': 'http://frame9.loadup.ru/d7/32/2888853.2.3.jpg',
-                'upload_date': '20150114',
+                'title': 'Sexy Beach (пароль 123)',
+                'uploader': 'вАся',
+                'uploader_id': 'asya_prosto',
+                'upload_date': '20081218',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'age_limit': 18,
+            },
+            'params': {
+                'videopassword': '123'
             },
         },
         # swf player
@@ -152,6 +172,10 @@ class SmotriIE(InfoExtractor):
             'getvideoinfo': '1',
         }
 
+        video_password = self._downloader.params.get('videopassword', None)
+        if video_password:
+            video_form['pass'] = hashlib.md5(video_password.encode('utf-8')).hexdigest()
+
         request = compat_urllib_request.Request(
             'http://smotri.com/video/view/url/bot/', compat_urllib_parse.urlencode(video_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
@@ -161,13 +185,18 @@ class SmotriIE(InfoExtractor):
         video_url = video.get('_vidURL') or video.get('_vidURL_mp4')
 
         if not video_url:
-            if video.get('_moderate_no') or not video.get('moderated'):
+            if video.get('_moderate_no'):
                 raise ExtractorError(
                     'Video %s has not been approved by moderator' % video_id, expected=True)
 
             if video.get('error'):
                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
 
+            if video.get('_pass_protected') == 1:
+                msg = ('Invalid video password' if video_password
+                       else 'This video is protected by a password, use the --video-password option')
+                raise ExtractorError(msg, expected=True)
+
         title = video['title']
         thumbnail = video['_imgURL']
         upload_date = unified_strdate(video['added'])

youtube_dl/extractor/snagfilms.py

@@ -0,0 +1,181 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    clean_html,
+    determine_ext,
+    int_or_none,
+    js_to_json,
+    parse_duration,
+)
+
+
+class SnagFilmsEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:(?:www|embed)\.)?snagfilms\.com/embed/player\?.*\bfilmId=(?P<id>[\da-f-]{36})'
+    _TESTS = [{
+        'url': 'http://embed.snagfilms.com/embed/player?filmId=74849a00-85a9-11e1-9660-123139220831&w=500',
+        'md5': '2924e9215c6eff7a55ed35b72276bd93',
+        'info_dict': {
+            'id': '74849a00-85a9-11e1-9660-123139220831',
+            'ext': 'mp4',
+            'title': '#whilewewatch',
+        }
+    }, {
+        # invalid labels, 360p is better that 480p
+        'url': 'http://www.snagfilms.com/embed/player?filmId=17ca0950-a74a-11e0-a92a-0026bb61d036',
+        'md5': '882fca19b9eb27ef865efeeaed376a48',
+        'info_dict': {
+            'id': '17ca0950-a74a-11e0-a92a-0026bb61d036',
+            'ext': 'mp4',
+            'title': 'Life in Limbo',
+        }
+    }, {
+        'url': 'http://www.snagfilms.com/embed/player?filmId=0000014c-de2f-d5d6-abcf-ffef58af0017',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:embed\.)?snagfilms\.com/embed/player.+?)\1',
+            webpage)
+        if mobj:
+            return mobj.group('url')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        if '>This film is not playable in your area.<' in webpage:
+            raise ExtractorError(
+                'Film %s is not playable in your area.' % video_id, expected=True)
+
+        formats = []
+        for source in self._parse_json(js_to_json(self._search_regex(
+                r'(?s)sources:\s*(\[.+?\]),', webpage, 'json')), video_id):
+            file_ = source.get('file')
+            if not file_:
+                continue
+            type_ = source.get('type')
+            ext = determine_ext(file_)
+            format_id = source.get('label') or ext
+            if all(v == 'm3u8' for v in (type_, ext)):
+                formats.extend(self._extract_m3u8_formats(
+                    file_, video_id, 'mp4', m3u8_id='hls'))
+            else:
+                bitrate = int_or_none(self._search_regex(
+                    [r'(\d+)kbps', r'_\d{1,2}x\d{1,2}_(\d{3,})\.%s' % ext],
+                    file_, 'bitrate', default=None))
+                height = int_or_none(self._search_regex(
+                    r'^(\d+)[pP]$', format_id, 'height', default=None))
+                formats.append({
+                    'url': file_,
+                    'format_id': format_id,
+                    'tbr': bitrate,
+                    'height': height,
+                })
+        self._sort_formats(formats)
+
+        title = self._search_regex(
+            [r"title\s*:\s*'([^']+)'", r'<title>([^<]+)</title>'],
+            webpage, 'title')
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+        }
+
+
+class SnagFilmsIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?snagfilms\.com/(?:films/title|show)/(?P<id>[^?#]+)'
+    _TESTS = [{
+        'url': 'http://www.snagfilms.com/films/title/lost_for_life',
+        'md5': '19844f897b35af219773fd63bdec2942',
+        'info_dict': {
+            'id': '0000014c-de2f-d5d6-abcf-ffef58af0017',
+            'display_id': 'lost_for_life',
+            'ext': 'mp4',
+            'title': 'Lost for Life',
+            'description': 'md5:fbdacc8bb6b455e464aaf98bc02e1c82',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 4489,
+            'categories': ['Documentary', 'Crime', 'Award Winning', 'Festivals']
+        }
+    }, {
+        'url': 'http://www.snagfilms.com/show/the_world_cut_project/india',
+        'md5': 'e6292e5b837642bbda82d7f8bf3fbdfd',
+        'info_dict': {
+            'id': '00000145-d75c-d96e-a9c7-ff5c67b20000',
+            'display_id': 'the_world_cut_project/india',
+            'ext': 'mp4',
+            'title': 'India',
+            'description': 'md5:5c168c5a8f4719c146aad2e0dfac6f5f',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 979,
+            'categories': ['Documentary', 'Sports', 'Politics']
+        }
+    }, {
+        # Film is not playable in your area.
+        'url': 'http://www.snagfilms.com/films/title/inside_mecca',
+        'only_matching': True,
+    }, {
+        # Film is not available.
+        'url': 'http://www.snagfilms.com/show/augie_alone/flirting',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        if ">Sorry, the Film you're looking for is not available.<" in webpage:
+            raise ExtractorError(
+                'Film %s is not available.' % display_id, expected=True)
+
+        film_id = self._search_regex(r'filmId=([\da-f-]{36})"', webpage, 'film id')
+
+        snag = self._parse_json(
+            self._search_regex(
+                'Snag\.page\.data\s*=\s*(\[.+?\]);', webpage, 'snag'),
+            display_id)
+
+        for item in snag:
+            if item.get('data', {}).get('film', {}).get('id') == film_id:
+                data = item['data']['film']
+                title = data['title']
+                description = clean_html(data.get('synopsis'))
+                thumbnail = data.get('image')
+                duration = int_or_none(data.get('duration') or data.get('runtime'))
+                categories = [
+                    category['title'] for category in data.get('categories', [])
+                    if category.get('title')]
+                break
+        else:
+            title = self._search_regex(
+                r'itemprop="title">([^<]+)<', webpage, 'title')
+            description = self._html_search_regex(
+                r'(?s)<div itemprop="description" class="film-synopsis-inner ">(.+?)</div>',
+                webpage, 'description', default=None) or self._og_search_description(webpage)
+            thumbnail = self._og_search_thumbnail(webpage)
+            duration = parse_duration(self._search_regex(
+                r'<span itemprop="duration" class="film-duration strong">([^<]+)<',
+                webpage, 'duration', fatal=False))
+            categories = re.findall(r'<a href="/movies/[^"]+">([^<]+)</a>', webpage)
+
+        return {
+            '_type': 'url_transparent',
+            'url': 'http://embed.snagfilms.com/embed/player?filmId=%s' % film_id,
+            'id': film_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'categories': categories,
+        }

youtube_dl/extractor/sohu.py

@@ -6,9 +6,12 @@ import re
 from .common import InfoExtractor
 from ..compat import (
     compat_str,
-    compat_urllib_request
+    compat_urllib_request,
+    compat_urllib_parse,
+)
+from ..utils import (
+    ExtractorError,
 )
-from ..utils import ExtractorError
 
 
 class SohuIE(InfoExtractor):
@@ -26,7 +29,7 @@ class SohuIE(InfoExtractor):
         'skip': 'On available in China',
     }, {
         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
-        'md5': 'ac9a5d322b4bf9ae184d53e4711e4f1a',
+        'md5': '699060e75cf58858dd47fb9c03c42cfb',
         'info_dict': {
             'id': '409385080',
             'ext': 'mp4',
@@ -34,7 +37,7 @@ class SohuIE(InfoExtractor):
         }
     }, {
         'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
-        'md5': '49308ff6dafde5ece51137d04aec311e',
+        'md5': '9bf34be48f2f4dadcb226c74127e203c',
         'info_dict': {
             'id': '78693464',
             'ext': 'mp4',
@@ -48,7 +51,7 @@ class SohuIE(InfoExtractor):
             'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
         },
         'playlist': [{
-            'md5': '492923eac023ba2f13ff69617c32754a',
+            'md5': 'bdbfb8f39924725e6589c146bc1883ad',
             'info_dict': {
                 'id': '78910339_part1',
                 'ext': 'mp4',
@@ -56,7 +59,7 @@ class SohuIE(InfoExtractor):
                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
             }
         }, {
-            'md5': 'de604848c0e8e9c4a4dde7e1347c0637',
+            'md5': '3e1f46aaeb95354fd10e7fca9fc1804e',
             'info_dict': {
                 'id': '78910339_part2',
                 'ext': 'mp4',
@@ -64,7 +67,7 @@ class SohuIE(InfoExtractor):
                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
             }
         }, {
-            'md5': '93584716ee0657c0b205b8aa3d27aa13',
+            'md5': '8407e634175fdac706766481b9443450',
             'info_dict': {
                 'id': '78910339_part3',
                 'ext': 'mp4',
@@ -139,21 +142,42 @@ class SohuIE(InfoExtractor):
         for i in range(part_count):
             formats = []
             for format_id, format_data in formats_json.items():
-                data = format_data['data']
+                allot = format_data['allot']
+
+                data = format_data['data']
+                clips_url = data['clipsURL']
+                su = data['su']
 
-                # URLs starts with http://newflv.sohu.ccgslb.net/ is not usable
-                # so retry until got a working URL
                 video_url = 'newflv.sohu.ccgslb.net'
+                cdnId = None
                 retries = 0
-                while 'newflv.sohu.ccgslb.net' in video_url and retries < 5:
-                    download_note = 'Download information from CDN gateway for format ' + format_id
+
+                while 'newflv.sohu.ccgslb.net' in video_url:
+                    params = {
+                        'prot': 9,
+                        'file': clips_url[i],
+                        'new': su[i],
+                        'prod': 'flash',
+                    }
+
+                    if cdnId is not None:
+                        params['idc'] = cdnId
+
+                    download_note = 'Downloading %s video URL part %d of %d' % (
+                        format_id, i + 1, part_count)
+
                     if retries > 0:
                         download_note += ' (retry #%d)' % retries
+                    part_info = self._parse_json(self._download_webpage(
+                        'http://%s/?%s' % (allot, compat_urllib_parse.urlencode(params)),
+                        video_id, download_note), video_id)
+
+                    video_url = part_info['url']
+                    cdnId = part_info.get('nid')
+
                     retries += 1
-                    cdn_info = self._download_json(
-                        'http://data.vod.itc.cn/cdnList?new=' + data['su'][i],
-                        video_id, download_note)
-                    video_url = cdn_info['url']
+                    if retries > 5:
+                        raise ExtractorError('Failed to get video URL')
 
                 formats.append({
                     'url': video_url,