release 2016.02.22

Closes #8075

AUTHORS

@@ -155,3 +155,8 @@ Vignesh Venkat
 Tom Gijselinck
 Founder Fang
 Andrew Alexeyew
+Saso Bezlaj
+Erwin de Haan
+Jens Wille
+Robin Houtevelts
+Patrick Griffis

CONTRIBUTING.md

@@ -1,6 +1,6 @@
-**Please include the full output of youtube-dl when run with `-v`**, i.e. add `-v` flag to your command line, copy the **whole** output and post it in the issue body wrapped in \`\`\` for better formatting. It should look similar to this:
+**Please include the full output of youtube-dl when run with `-v`**, i.e. **add** `-v` flag to **your command line**, copy the **whole** output and post it in the issue body wrapped in \`\`\` for better formatting. It should look similar to this:
 ```
-$ youtube-dl -v http://www.youtube.com/watch?v=BaW_jenozKcj
+$ youtube-dl -v <your command line>
 [debug] System config: []
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']

Makefile

@@ -44,7 +44,7 @@ test:
 ot: offlinetest
 
 offlinetest: codetest
-	nosetests --verbose test --exclude test_download.py --exclude test_age_restriction.py --exclude test_subtitles.py --exclude test_write_annotations.py --exclude test_youtube_lists.py
+	nosetests --verbose test --exclude test_download.py --exclude test_age_restriction.py --exclude test_subtitles.py --exclude test_write_annotations.py --exclude test_youtube_lists.py --exclude test_iqiyi_sdk_interpreter.py
 
 tar: youtube-dl.tar.gz
 

README.md

@@ -173,6 +173,10 @@ which means you can modify it, redistribute it or use it however you like.
                                      expected filesize (experimental)
     --hls-prefer-native              Use the native HLS downloader instead of
                                      ffmpeg (experimental)
+    --hls-use-mpegts                 Use the mpegts container for HLS videos,
+                                     allowing to play the video while
+                                     downloading (some players may not be able
+                                     to play it)
     --external-downloader COMMAND    Use the specified external downloader.
                                      Currently supports
                                      aria2c,axel,curl,httpie,wget
@@ -438,28 +442,97 @@ On Windows you may also need to setup the `%HOME%` environment variable manually
 
 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 parentheses, followed by a lowercase S. Allowed names are:
 
- - `id`: The sequence will be replaced by the video identifier.
- - `url`: The sequence will be replaced by the video URL.
- - `uploader`: The sequence will be replaced by the nickname of the person who uploaded the video.
- - `upload_date`: The sequence will be replaced by the upload date in YYYYMMDD format.
- - `title`: The sequence will be replaced by the video title.
- - `ext`: The sequence will be replaced by the appropriate extension (like flv or mp4).
- - `epoch`: The sequence will be replaced by the Unix epoch when creating the file.
- - `autonumber`: The sequence will be replaced by a five-digit number that will be increased with each download, starting at zero.
- - `playlist`: The sequence will be replaced by the name or the id of the playlist that contains the video.
- - `playlist_index`: The sequence will be replaced by the index of the video in the playlist padded with leading zeros according to the total length of the playlist.
- - `format_id`: The sequence will be replaced by the format code specified by `--format`.
- - `duration`: The sequence will be replaced by the length of the video in seconds.
+ - `id`: Video identifier
+ - `title`: Video title
+ - `url`: Video URL
+ - `ext`: Video filename extension
+ - `alt_title`: A secondary title of the video
+ - `display_id`: An alternative identifier for the video
+ - `uploader`: Full name of the video uploader
+ - `creator`: The main artist who created the video
+ - `release_date`: The date (YYYYMMDD) when the video was released
+ - `timestamp`: UNIX timestamp of the moment the video became available
+ - `upload_date`: Video upload date (YYYYMMDD)
+ - `uploader_id`: Nickname or id of the video uploader
+ - `location`: Physical location where the video was filmed
+ - `duration`: Length of the video in seconds
+ - `view_count`: How many users have watched the video on the platform
+ - `like_count`: Number of positive ratings of the video
+ - `dislike_count`: Number of negative ratings of the video
+ - `repost_count`: Number of reposts of the video
+ - `average_rating`: Average rating give by users, the scale used depends on the webpage
+ - `comment_count`: Number of comments on the video
+ - `age_limit`: Age restriction for the video (years)
+ - `format`: A human-readable description of the format 
+ - `format_id`: Format code specified by `--format`
+ - `format_note`: Additional info about the format
+ - `width`: Width of the video
+ - `height`: Height of the video
+ - `resolution`: Textual description of width and height
+ - `tbr`: Average bitrate of audio and video in KBit/s
+ - `abr`: Average audio bitrate in KBit/s
+ - `acodec`: Name of the audio codec in use
+ - `asr`: Audio sampling rate in Hertz
+ - `vbr`: Average video bitrate in KBit/s
+ - `fps`: Frame rate
+ - `vcodec`: Name of the video codec in use
+ - `container`: Name of the container format
+ - `filesize`: The number of bytes, if known in advance
+ - `filesize_approx`: An estimate for the number of bytes
+ - `protocol`: The protocol that will be used for the actual download
+ - `extractor`: Name of the extractor
+ - `extractor_key`: Key name of the extractor
+ - `epoch`: Unix epoch when creating the file
+ - `autonumber`: Five-digit number that will be increased with each download, starting at zero
+ - `playlist`: Name or id of the playlist that contains the video
+ - `playlist_index`: Index of the video in the playlist padded with leading zeros according to the total length of the playlist
+
+Available for the video that belongs to some logical chapter or section:
+ - `chapter`: Name or title of the chapter the video belongs to
+ - `chapter_number`: Number of the chapter the video belongs to
+ - `chapter_id`: Id of the chapter the video belongs to
+
+Available for the video that is an episode of some series or programme:
+ - `series`: Title of the series or programme the video episode belongs to
+ - `season`: Title of the season the video episode belongs to
+ - `season_number`: Number of the season the video episode belongs to
+ - `season_id`: Id of the season the video episode belongs to
+ - `episode`: Title of the video episode
+ - `episode_number`: Number of the video episode within a season
+ - `episode_id`: Id of the video episode
+
+Each aforementioned sequence when referenced in output template will be replaced by the actual value corresponding to the sequence name. Note that some of the sequences are not guaranteed to be present since they depend on the metadata obtained by particular extractor, such sequences will be replaced with `NA`.
+
+For example for `-o %(title)s-%(id)s.%(ext)s` and mp4 video with title `youtube-dl test video` and id `BaW_jenozKcj` this will result in a `youtube-dl test video-BaW_jenozKcj.mp4` file created in the current directory.
+
+Output template can also contain arbitrary hierarchical path, e.g. `-o '%(playlist)s/%(playlist_index)s - %(title)s.%(ext)s'` that will result in downloading each video in a directory corresponding to this path template. Any missing directory will be automatically created for you.
+
+To specify percent literal in output template use `%%`. To output to stdout use `-o -`.
 
 The current default template is `%(title)s-%(id)s.%(ext)s`.
 
 In some cases, you don't want special characters such as 中, spaces, or &, such as when transferring the downloaded filename to a Windows system or the filename through an 8bit-unsafe channel. In these cases, add the `--restrict-filenames` flag to get a shorter title:
 
+Examples (note on Windows you may need to use double quotes instead of single):
+
 ```bash
-$ youtube-dl --get-filename -o "%(title)s.%(ext)s" BaW_jenozKc
+$ youtube-dl --get-filename -o '%(title)s.%(ext)s' BaW_jenozKc
 youtube-dl test video ''_ä↭𝕐.mp4    # All kinds of weird characters
-$ youtube-dl --get-filename -o "%(title)s.%(ext)s" BaW_jenozKc --restrict-filenames
+
+$ youtube-dl --get-filename -o '%(title)s.%(ext)s' BaW_jenozKc --restrict-filenames
 youtube-dl_test_video_.mp4          # A simple file name
+
+# Download YouTube playlist videos in separate directory indexed by video order in a playlist
+$ youtube-dl -o '%(playlist)s/%(playlist_index)s - %(title)s.%(ext)s' https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re
+
+# Download Udemy course keeping each chapter in separate directory under MyVideos directory in your home
+$ youtube-dl -u user -p password -o '~/MyVideos/%(playlist)s/%(chapter_number)s - %(chapter)s/%(title)s.%(ext)s' https://www.udemy.com/java-tutorial/
+
+# Download entire series season keeping each series and each season in separate directory under C:/MyVideos
+$ youtube-dl -o "C:/MyVideos/%(series)s/%(season_number)s - %(season)s/%(episode_number)s - %(episode)s.%(ext)s" http://videomore.ru/kino_v_detalayah/5_sezon/367617
+
+# Stream the video being downloaded to stdout
+$ youtube-dl -o - BaW_jenozKc
 ```
 
 # FORMAT SELECTION
@@ -862,9 +935,9 @@ with youtube_dl.YoutubeDL(ydl_opts) as ydl:
 
 Bugs and suggestions should be reported at: <https://github.com/rg3/youtube-dl/issues>. Unless you were prompted so or there is another pertinent reason (e.g. GitHub fails to accept the bug report), please do not send bug reports via personal email. For discussions, join us in the IRC channel [#youtube-dl](irc://chat.freenode.net/#youtube-dl) on freenode ([webchat](http://webchat.freenode.net/?randomnick=1&channels=youtube-dl)).
 
-**Please include the full output of youtube-dl when run with `-v`**, i.e. add `-v` flag to your command line, copy the **whole** output and post it in the issue body wrapped in \`\`\` for better formatting. It should look similar to this:
+**Please include the full output of youtube-dl when run with `-v`**, i.e. **add** `-v` flag to **your command line**, copy the **whole** output and post it in the issue body wrapped in \`\`\` for better formatting. It should look similar to this:
 ```
-$ youtube-dl -v http://www.youtube.com/watch?v=BaW_jenozKcj
+$ youtube-dl -v <your command line>
 [debug] System config: []
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']

docs/supportedsites.md

@@ -30,6 +30,7 @@
  - **AlJazeera**
  - **Allocine**
  - **AlphaPorno**
+ - **AnimeOnDemand**
  - **anitube.se**
  - **AnySex**
  - **Aparat**
@@ -49,12 +50,14 @@
  - **arte.tv:ddc**
  - **arte.tv:embed**
  - **arte.tv:future**
+ - **arte.tv:magazine**
  - **AtresPlayer**
  - **ATTTechChannel**
  - **AudiMedia**
  - **audiomack**
  - **audiomack:album**
  - **Azubu**
+ - **AzubuLive**
  - **BaiduVideo**: 百度视频
  - **bambuser**
  - **bambuser:channel**
@@ -88,8 +91,11 @@
  - **canalc2.tv**
  - **Canalplus**: canalplus.fr, piwiplus.fr and d8.tv
  - **Canvas**
+ - **CBC**
+ - **CBCPlayer**
  - **CBS**
  - **CBSNews**: CBS News
+ - **CBSNewsLiveVideo**: CBS News Live Videos
  - **CBSSports**
  - **CeskaTelevize**
  - **channel9**: Channel 9
@@ -118,6 +124,7 @@
  - **ComedyCentralShows**: The Daily Show / The Colbert Report
  - **CondeNast**: Condé Nast media group: Allure, Architectural Digest, Ars Technica, Bon Appétit, Brides, Condé Nast, Condé Nast Traveler, Details, Epicurious, GQ, Glamour, Golf Digest, SELF, Teen Vogue, The New Yorker, Vanity Fair, Vogue, W Magazine, WIRED
  - **Cracked**
+ - **Crackle**
  - **Criterion**
  - **CrooksAndLiars**
  - **Crunchyroll**
@@ -133,6 +140,8 @@
  - **DailymotionCloud**
  - **daum.net**
  - **daum.net:clip**
+ - **daum.net:playlist**
+ - **daum.net:user**
  - **DBTV**
  - **DCN**
  - **dcn:live**
@@ -180,6 +189,7 @@
  - **ExpoTV**
  - **ExtremeTube**
  - **facebook**
+ - **facebook:post**
  - **faz.net**
  - **fc2**
  - **Fczenit**
@@ -257,7 +267,7 @@
  - **Instagram**
  - **instagram:user**: Instagram user profile
  - **InternetVideoArchive**
- - **IPrima** (Currently broken)
+ - **IPrima**
  - **iqiyi**: 爱奇艺
  - **Ir90Tv**
  - **ivi**: ivi.ru
@@ -278,6 +288,7 @@
  - **KeezMovies**
  - **KhanAcademy**
  - **KickStarter**
+ - **KonserthusetPlay**
  - **kontrtube**: KontrTube.ru - Труба зовёт
  - **KrasView**: Красвью
  - **Ku6**
@@ -314,6 +325,7 @@
  - **mailru**: Видео@Mail.Ru
  - **MakerTV**
  - **Malemotion**
+ - **MatchTV**
  - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
  - **metacafe**
@@ -350,7 +362,7 @@
  - **MySpace:album**
  - **MySpass**
  - **Myvi**
- - **myvideo**
+ - **myvideo** (Currently broken)
  - **MyVidster**
  - **n-tv.de**
  - **NationalGeographic**
@@ -400,6 +412,7 @@
  - **NowTV** (Currently broken)
  - **NowTVList**
  - **nowvideo**: NowVideo
+ - **Noz**
  - **npo**: npo.nl and ntr.nl
  - **npo.nl:live**
  - **npo.nl:radio**
@@ -438,6 +451,7 @@
  - **PlanetaPlay**
  - **play.fm**
  - **played.to**
+ - **PlaysTV**
  - **Playtvak**: Playtvak.cz, iDNES.cz and Lidovky.cz
  - **Playvid**
  - **Playwire**
@@ -449,6 +463,7 @@
  - **PornHd**
  - **PornHub**
  - **PornHubPlaylist**
+ - **PornHubUserVideos**
  - **Pornotube**
  - **PornoVoisines**
  - **PornoXO**
@@ -506,10 +521,12 @@
  - **Sapo**: SAPO Vídeos
  - **savefrom.net**
  - **SBS**: sbs.com.au
+ - **schooltv**
  - **SciVee**
  - **screen.yahoo:search**: Yahoo screen search
  - **Screencast**
  - **ScreencastOMatic**
+ - **ScreenJunkies**
  - **ScreenwaveMedia**
  - **SenateISVP**
  - **ServingSys**
@@ -643,6 +660,7 @@
  - **twitch:video**
  - **twitch:vod**
  - **twitter**
+ - **twitter:amplify**
  - **twitter:card**
  - **Ubu**
  - **udemy**
@@ -674,7 +692,9 @@
  - **VideoPremium**
  - **VideoTt**: video.tt - Your True Tube (Currently broken)
  - **videoweed**: VideoWeed
- - **Vidme**
+ - **vidme**
+ - **vidme:user**
+ - **vidme:user:likes**
  - **Vidzi**
  - **vier**
  - **vier:videos**

test/test_YoutubeDL.py

@@ -14,6 +14,7 @@ from test.helper import FakeYDL, assertRegexpMatches
 from youtube_dl import YoutubeDL
 from youtube_dl.compat import compat_str, compat_urllib_error
 from youtube_dl.extractor import YoutubeIE
+from youtube_dl.extractor.common import InfoExtractor
 from youtube_dl.postprocessor.common import PostProcessor
 from youtube_dl.utils import ExtractorError, match_filter_func
 
@@ -221,9 +222,19 @@ class TestFormatSelection(unittest.TestCase):
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], 'dash-video-low')
 
+        formats = [
+            {'format_id': 'vid-vcodec-dot', 'ext': 'mp4', 'preference': 1, 'vcodec': 'avc1.123456', 'acodec': 'none', 'url': TEST_URL},
+        ]
+        info_dict = _make_result(formats)
+
+        ydl = YDL({'format': 'bestvideo[vcodec=avc1.123456]'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'vid-vcodec-dot')
+
     def test_youtube_format_selection(self):
         order = [
-            '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '36', '17', '13',
+            '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '17', '36', '13',
             # Apple HTTP Live Streaming
             '96', '95', '94', '93', '92', '132', '151',
             # 3D
@@ -237,6 +248,17 @@ class TestFormatSelection(unittest.TestCase):
 
         def format_info(f_id):
             info = YoutubeIE._formats[f_id].copy()
+
+            # XXX: In real cases InfoExtractor._parse_mpd_formats() fills up 'acodec'
+            # and 'vcodec', while in tests such information is incomplete since
+            # commit a6c2c24479e5f4827ceb06f64d855329c0a6f593
+            # test_YoutubeDL.test_youtube_format_selection is broken without
+            # this fix
+            if 'acodec' in info and 'vcodec' not in info:
+                info['vcodec'] = 'none'
+            elif 'vcodec' in info and 'acodec' not in info:
+                info['acodec'] = 'none'
+
             info['format_id'] = f_id
             info['url'] = 'url:' + f_id
             return info
@@ -636,6 +658,42 @@ class TestYoutubeDL(unittest.TestCase):
         ydl = YDL()
         self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd')
 
+    def test_do_not_override_ie_key_in_url_transparent(self):
+        ydl = YDL()
+
+        class Foo1IE(InfoExtractor):
+            _VALID_URL = r'foo1:'
+
+            def _real_extract(self, url):
+                return {
+                    '_type': 'url_transparent',
+                    'url': 'foo2:',
+                    'ie_key': 'Foo2',
+                }
+
+        class Foo2IE(InfoExtractor):
+            _VALID_URL = r'foo2:'
+
+            def _real_extract(self, url):
+                return {
+                    '_type': 'url',
+                    'url': 'foo3:',
+                    'ie_key': 'Foo3',
+                }
+
+        class Foo3IE(InfoExtractor):
+            _VALID_URL = r'foo3:'
+
+            def _real_extract(self, url):
+                return _make_result([{'url': TEST_URL}])
+
+        ydl.add_info_extractor(Foo1IE(ydl))
+        ydl.add_info_extractor(Foo2IE(ydl))
+        ydl.add_info_extractor(Foo3IE(ydl))
+        ydl.extract_info('foo1:')
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['url'], TEST_URL)
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_all_urls.py

@@ -56,7 +56,7 @@ class TestAllURLsMatching(unittest.TestCase):
         assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM/videos')
 
     def test_youtube_user_matching(self):
-        self.assertMatch('www.youtube.com/NASAgovVideo/videos', ['youtube:user'])
+        self.assertMatch('http://www.youtube.com/NASAgovVideo/videos', ['youtube:user'])
 
     def test_youtube_feeds(self):
         self.assertMatch('https://www.youtube.com/feed/watch_later', ['youtube:watchlater'])

test/test_iqiyi_sdk_interpreter.py

@@ -0,0 +1,47 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL
+from youtube_dl.extractor import IqiyiIE
+
+
+class IqiyiIEWithCredentials(IqiyiIE):
+    def _get_login_info(self):
+        return 'foo', 'bar'
+
+
+class WarningLogger(object):
+    def __init__(self):
+        self.messages = []
+
+    def warning(self, msg):
+        self.messages.append(msg)
+
+    def debug(self, msg):
+        pass
+
+    def error(self, msg):
+        pass
+
+
+class TestIqiyiSDKInterpreter(unittest.TestCase):
+    def test_iqiyi_sdk_interpreter(self):
+        '''
+        Test the functionality of IqiyiSDKInterpreter by trying to log in
+
+        If `sign` is incorrect, /validate call throws an HTTP 556 error
+        '''
+        logger = WarningLogger()
+        ie = IqiyiIEWithCredentials(FakeYDL({'logger': logger}))
+        ie._login()
+        self.assertTrue('unable to log in:' in logger.messages[0])
+
+if __name__ == '__main__':
+    unittest.main()

test/test_subtitles.py

@@ -21,7 +21,7 @@ from youtube_dl.extractor import (
     NPOIE,
     ComedyCentralIE,
     NRKTVIE,
-    RaiIE,
+    RaiTVIE,
     VikiIE,
     ThePlatformIE,
     ThePlatformFeedIE,
@@ -65,16 +65,16 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(len(subtitles.keys()), 13)
-        self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260')
-        self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d')
-        for lang in ['it', 'fr', 'de']:
+        self.assertEqual(md5(subtitles['en']), '3cb210999d3e021bd6c7f0ea751eab06')
+        self.assertEqual(md5(subtitles['it']), '6d752b98c31f1cf8d597050c7a2cb4b5')
+        for lang in ['fr', 'de']:
             self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
-    def test_youtube_subtitles_sbv_format(self):
+    def test_youtube_subtitles_ttml_format(self):
         self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitlesformat'] = 'sbv'
+        self.DL.params['subtitlesformat'] = 'ttml'
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b')
+        self.assertEqual(md5(subtitles['en']), 'e306f8c42842f723447d9f63ad65df54')
 
     def test_youtube_subtitles_vtt_format(self):
         self.DL.params['writesubtitles'] = True
@@ -260,7 +260,7 @@ class TestNRKSubtitles(BaseTestSubtitles):
 
 class TestRaiSubtitles(BaseTestSubtitles):
     url = 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html'
-    IE = RaiIE
+    IE = RaiTVIE
 
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True

test/test_utils.py

@@ -22,6 +22,7 @@ from youtube_dl.utils import (
     DateRange,
     detect_exe_version,
     determine_ext,
+    dict_get,
     encode_compat_str,
     encodeFilename,
     escape_rfc3986,
@@ -34,6 +35,7 @@ from youtube_dl.utils import (
     is_html,
     js_to_json,
     limit_length,
+    ohdave_rsa_encrypt,
     OnDemandPagedList,
     orderedSet,
     parse_duration,
@@ -450,6 +452,28 @@ class TestUtil(unittest.TestCase):
         data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'})
         self.assertTrue(isinstance(data, bytes))
 
+    def test_dict_get(self):
+        FALSE_VALUES = {
+            'none': None,
+            'false': False,
+            'zero': 0,
+            'empty_string': '',
+            'empty_list': [],
+        }
+        d = FALSE_VALUES.copy()
+        d['a'] = 42
+        self.assertEqual(dict_get(d, 'a'), 42)
+        self.assertEqual(dict_get(d, 'b'), None)
+        self.assertEqual(dict_get(d, 'b', 42), 42)
+        self.assertEqual(dict_get(d, ('a', )), 42)
+        self.assertEqual(dict_get(d, ('b', 'a', )), 42)
+        self.assertEqual(dict_get(d, ('b', 'c', 'a', 'd', )), 42)
+        self.assertEqual(dict_get(d, ('b', 'c', )), None)
+        self.assertEqual(dict_get(d, ('b', 'c', ), 42), 42)
+        for key, false_value in FALSE_VALUES.items():
+            self.assertEqual(dict_get(d, ('b', 'c', key, )), None)
+            self.assertEqual(dict_get(d, ('b', 'c', key, ), skip_false_values=False), false_value)
+
     def test_encode_compat_str(self):
         self.assertEqual(encode_compat_str(b'\xd1\x82\xd0\xb5\xd1\x81\xd1\x82', 'utf-8'), 'тест')
         self.assertEqual(encode_compat_str('тест', 'utf-8'), 'тест')
@@ -471,6 +495,10 @@ class TestUtil(unittest.TestCase):
         d = json.loads(stripped)
         self.assertEqual(d, {'STATUS': 'OK'})
 
+        stripped = strip_jsonp('ps.embedHandler({"status": "success"});')
+        d = json.loads(stripped)
+        self.assertEqual(d, {'status': 'success'})
+
     def test_uppercase_escape(self):
         self.assertEqual(uppercase_escape('aä'), 'aä')
         self.assertEqual(uppercase_escape('\\U0001d550'), '𝕐')
@@ -765,6 +793,13 @@ The first line
                 {'nocheckcertificate': False}, '--check-certificate', 'nocheckcertificate', 'false', 'true', '='),
             ['--check-certificate=true'])
 
+    def test_ohdave_rsa_encrypt(self):
+        N = 0xab86b6371b5318aaa1d3c9e612a9f1264f372323c8c0f19875b5fc3b3fd3afcc1e5bec527aa94bfa85bffc157e4245aebda05389a5357b75115ac94f074aefcd
+        e = 65537
+
+        self.assertEqual(
+            ohdave_rsa_encrypt(b'aa111222', e, N),
+            '726664bd9a23fd0c70f9f1b84aab5e3905ce1e45a584e9cbcf9bcc7510338fc1986d6c599ff990d923aa43c51c0d9013cd572e13bc58f4ae48f2ed8c0b0ba881')
 
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_lists.py

@@ -34,7 +34,7 @@ class TestYoutubeLists(unittest.TestCase):
         ie = YoutubePlaylistIE(dl)
         # TODO find a > 100 (paginating?) videos course
         result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
-        entries = result['entries']
+        entries = list(result['entries'])
         self.assertEqual(YoutubeIE().extract_id(entries[0]['url']), 'j9WZyLZCBzs')
         self.assertEqual(len(entries), 25)
         self.assertEqual(YoutubeIE().extract_id(entries[-1]['url']), 'rYefUsYuEp0')

youtube_dl/YoutubeDL.py

@@ -263,7 +263,7 @@ class YoutubeDL(object):
     the downloader (see youtube_dl/downloader/common.py):
     nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
     noresizebuffer, retries, continuedl, noprogress, consoletitle,
-    xattr_set_filesize, external_downloader_args.
+    xattr_set_filesize, external_downloader_args, hls_use_mpegts.
 
     The following options are used by the post processors:
     prefer_ffmpeg:     If True, use ffmpeg instead of avconv if both are available,
@@ -605,12 +605,12 @@ class YoutubeDL(object):
             if rejecttitle:
                 if re.search(rejecttitle, title, re.IGNORECASE):
                     return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
-        date = info_dict.get('upload_date', None)
+        date = info_dict.get('upload_date')
         if date is not None:
             dateRange = self.params.get('daterange', DateRange())
             if date not in dateRange:
                 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
-        view_count = info_dict.get('view_count', None)
+        view_count = info_dict.get('view_count')
         if view_count is not None:
             min_views = self.params.get('min_views')
             if min_views is not None and view_count < min_views:
@@ -707,7 +707,6 @@ class YoutubeDL(object):
         It will also download the videos if 'download'.
         Returns the resolved ie_result.
         """
-
         result_type = ie_result.get('_type', 'video')
 
         if result_type in ('url', 'url_transparent'):
@@ -736,7 +735,7 @@ class YoutubeDL(object):
 
             force_properties = dict(
                 (k, v) for k, v in ie_result.items() if v is not None)
-            for f in ('_type', 'url'):
+            for f in ('_type', 'url', 'ie_key'):
                 if f in force_properties:
                     del force_properties[f]
             new_result = info.copy()
@@ -748,18 +747,18 @@ class YoutubeDL(object):
                 new_result, download=download, extra_info=extra_info)
         elif result_type == 'playlist' or result_type == 'multi_video':
             # We process each entry in the playlist
-            playlist = ie_result.get('title', None) or ie_result.get('id', None)
+            playlist = ie_result.get('title') or ie_result.get('id')
             self.to_screen('[download] Downloading playlist: %s' % playlist)
 
             playlist_results = []
 
             playliststart = self.params.get('playliststart', 1) - 1
-            playlistend = self.params.get('playlistend', None)
+            playlistend = self.params.get('playlistend')
             # For backwards compatibility, interpret -1 as whole list
             if playlistend == -1:
                 playlistend = None
 
-            playlistitems_str = self.params.get('playlist_items', None)
+            playlistitems_str = self.params.get('playlist_items')
             playlistitems = None
             if playlistitems_str is not None:
                 def iter_playlistitems(format):
@@ -783,7 +782,7 @@ class YoutubeDL(object):
                     entries = ie_entries[playliststart:playlistend]
                 n_entries = len(entries)
                 self.to_screen(
-                    "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
+                    '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
                     (ie_result['extractor'], playlist, n_all_entries, n_entries))
             elif isinstance(ie_entries, PagedList):
                 if playlistitems:
@@ -797,7 +796,7 @@ class YoutubeDL(object):
                         playliststart, playlistend)
                 n_entries = len(entries)
                 self.to_screen(
-                    "[%s] playlist %s: Downloading %d videos" %
+                    '[%s] playlist %s: Downloading %d videos' %
                     (ie_result['extractor'], playlist, n_entries))
             else:  # iterable
                 if playlistitems:
@@ -808,7 +807,7 @@ class YoutubeDL(object):
                         ie_entries, playliststart, playlistend))
                 n_entries = len(entries)
                 self.to_screen(
-                    "[%s] playlist %s: Downloading %d videos" %
+                    '[%s] playlist %s: Downloading %d videos' %
                     (ie_result['extractor'], playlist, n_entries))
 
             if self.params.get('playlistreverse', False):
@@ -906,7 +905,7 @@ class YoutubeDL(object):
             str_operator_rex = re.compile(r'''(?x)
                 \s*(?P<key>ext|acodec|vcodec|container|protocol)
                 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
-                \s*(?P<value>[a-zA-Z0-9_-]+)
+                \s*(?P<value>[a-zA-Z0-9._-]+)
                 \s*$
                 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
             m = str_operator_rex.search(filter_spec)
@@ -1289,6 +1288,9 @@ class YoutubeDL(object):
 
             if format.get('format_id') is None:
                 format['format_id'] = compat_str(i)
+            else:
+                # Sanitize format_id from characters used in format selector expression
+                format['format_id'] = re.sub('[\s,/+\[\]()]', '_', format['format_id'])
             format_id = format['format_id']
             if format_id not in formats_dict:
                 formats_dict[format_id] = []
@@ -1339,7 +1341,6 @@ 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'] and
                     not info_dict.get('is_live')):
                 merger = FFmpegMergerPP(self)
                 if merger.available and merger.can_merge():
@@ -1796,7 +1797,7 @@ class YoutubeDL(object):
             else:
                 res = '%sp' % format['height']
         elif format.get('width') is not None:
-            res = '?x%d' % format['width']
+            res = '%dx?' % format['width']
         else:
             res = default
         return res

youtube_dl/__init__.py

@@ -369,6 +369,7 @@ def _real_main(argv=None):
         'no_color': opts.no_color,
         'ffmpeg_location': opts.ffmpeg_location,
         'hls_prefer_native': opts.hls_prefer_native,
+        'hls_use_mpegts': opts.hls_use_mpegts,
         'external_downloader_args': external_downloader_args,
         'postprocessor_args': postprocessor_args,
         'cn_verification_proxy': opts.cn_verification_proxy,

youtube_dl/__main__.py

@@ -7,7 +7,7 @@ from __future__ import unicode_literals
 
 import sys
 
-if __package__ is None and not hasattr(sys, "frozen"):
+if __package__ is None and not hasattr(sys, 'frozen'):
     # direct call of __main__.py
     import os.path
     path = os.path.realpath(os.path.abspath(__file__))

youtube_dl/aes.py

@@ -161,7 +161,7 @@ def aes_decrypt_text(data, password, key_size_bytes):
     nonce = data[:NONCE_LENGTH_BYTES]
     cipher = data[NONCE_LENGTH_BYTES:]
 
-    class Counter:
+    class Counter(object):
         __value = nonce + [0] * (BLOCK_SIZE_BYTES - NONCE_LENGTH_BYTES)
 
         def next_value(self):

youtube_dl/compat.py

@@ -181,20 +181,20 @@ except ImportError:  # Python < 3.4
             # parameter := attribute "=" value
             url = req.get_full_url()
 
-            scheme, data = url.split(":", 1)
-            mediatype, data = data.split(",", 1)
+            scheme, data = url.split(':', 1)
+            mediatype, data = data.split(',', 1)
 
             # even base64 encoded data URLs might be quoted so unquote in any case:
             data = compat_urllib_parse_unquote_to_bytes(data)
-            if mediatype.endswith(";base64"):
+            if mediatype.endswith(';base64'):
                 data = binascii.a2b_base64(data)
                 mediatype = mediatype[:-7]
 
             if not mediatype:
-                mediatype = "text/plain;charset=US-ASCII"
+                mediatype = 'text/plain;charset=US-ASCII'
 
             headers = email.message_from_string(
-                "Content-type: %s\nContent-length: %d\n" % (mediatype, len(data)))
+                'Content-type: %s\nContent-length: %d\n' % (mediatype, len(data)))
 
             return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url)
 
@@ -268,7 +268,7 @@ except ImportError:  # Python 2
             nv = name_value.split('=', 1)
             if len(nv) != 2:
                 if strict_parsing:
-                    raise ValueError("bad query field: %r" % (name_value,))
+                    raise ValueError('bad query field: %r' % (name_value,))
                 # Handle case of a control-name with no equal sign
                 if keep_blank_values:
                     nv.append('')
@@ -466,7 +466,7 @@ if sys.version_info < (2, 7):
         if err is not None:
             raise err
         else:
-            raise socket.error("getaddrinfo returns an empty list")
+            raise socket.error('getaddrinfo returns an empty list')
 else:
     compat_socket_create_connection = socket.create_connection
 

youtube_dl/downloader/common.py

@@ -45,6 +45,7 @@ class FileDownloader(object):
                         (experimental)
     external_downloader_args:  A list of additional command-line arguments for the
                         external downloader.
+    hls_use_mpegts:     Use the mpegts container for HLS videos.
 
     Subclasses of this one must re-define the real_download method.
     """
@@ -156,7 +157,7 @@ class FileDownloader(object):
 
     def slow_down(self, start_time, now, byte_counter):
         """Sleep if the download speed is over the rate limit."""
-        rate_limit = self.params.get('ratelimit', None)
+        rate_limit = self.params.get('ratelimit')
         if rate_limit is None or byte_counter == 0:
             return
         if now is None:

youtube_dl/downloader/dash.py

@@ -1,66 +1,59 @@
 from __future__ import unicode_literals
 
+import os
 import re
 
-from .common import FileDownloader
-from ..utils import sanitized_Request
+from .fragment import FragmentFD
+from ..utils import (
+    sanitize_open,
+    encodeFilename,
+)
 
 
-class DashSegmentsFD(FileDownloader):
+class DashSegmentsFD(FragmentFD):
     """
     Download segments in a DASH manifest
     """
+
+    FD_NAME = 'dashsegments'
+
     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']
+        segment_urls = [info_dict['segment_urls'][0]] if self.params.get('test', False) else info_dict['segment_urls']
+        initialization_url = info_dict.get('initialization_url')
 
-        is_test = self.params.get('test', False)
-        remaining_bytes = self._TEST_FILE_SIZE if is_test else None
-        byte_counter = 0
+        ctx = {
+            'filename': filename,
+            'total_frags': len(segment_urls) + (1 if initialization_url else 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 = sanitized_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)
+        self._prepare_and_start_frag_download(ctx)
 
         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
+        segments_filenames = []
 
-        self.try_rename(tmpfilename, filename)
+        def append_url_to_file(target_url, target_filename):
+            success = ctx['dl'].download(target_filename, {'url': combine_url(base_url, target_url)})
+            if not success:
+                return False
+            down, target_sanitized = sanitize_open(target_filename, 'rb')
+            ctx['dest_stream'].write(down.read())
+            down.close()
+            segments_filenames.append(target_sanitized)
 
-        self._hook_progress({
-            'downloaded_bytes': byte_counter,
-            'total_bytes': byte_counter,
-            'filename': filename,
-            'status': 'finished',
-        })
+        if initialization_url:
+            append_url_to_file(initialization_url, ctx['tmpfilename'] + '-Init')
+        for i, segment_url in enumerate(segment_urls):
+            segment_filename = '%s-Seg%d' % (ctx['tmpfilename'], i)
+            append_url_to_file(segment_url, segment_filename)
+
+        self._finish_frag_download(ctx)
+
+        for segment_file in segments_filenames:
+            os.remove(encodeFilename(segment_file))
 
         return True

youtube_dl/downloader/f4m.py

@@ -273,15 +273,21 @@ class F4mFD(FragmentFD):
         return fragments_list
 
     def _parse_bootstrap_node(self, node, base_url):
-        if node.text is None:
+        # Sometimes non empty inline bootstrap info can be specified along
+        # with bootstrap url attribute (e.g. dummy inline bootstrap info
+        # contains whitespace characters in [1]). We will prefer bootstrap
+        # url over inline bootstrap info when present.
+        # 1. http://live-1-1.rutube.ru/stream/1024/HDS/SD/C2NKsS85HQNckgn5HdEmOQ/1454167650/S-s604419906/move/four/dirs/upper/1024-576p.f4m
+        bootstrap_url = node.get('url')
+        if bootstrap_url:
             bootstrap_url = compat_urlparse.urljoin(
-                base_url, node.attrib['url'])
+                base_url, bootstrap_url)
             boot_info = self._get_bootstrap_from_url(bootstrap_url)
         else:
             bootstrap_url = None
             bootstrap = base64.b64decode(node.text.encode('ascii'))
             boot_info = read_bootstrap_info(bootstrap)
-        return (boot_info, bootstrap_url)
+        return boot_info, bootstrap_url
 
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
@@ -316,7 +322,8 @@ class F4mFD(FragmentFD):
             metadata = None
 
         fragments_list = build_fragments_list(boot_info)
-        if self.params.get('test', False):
+        test = self.params.get('test', False)
+        if test:
             # We only download the first fragment
             fragments_list = fragments_list[:1]
         total_frags = len(fragments_list)
@@ -326,6 +333,7 @@ class F4mFD(FragmentFD):
         ctx = {
             'filename': filename,
             'total_frags': total_frags,
+            'live': live,
         }
 
         self._prepare_frag_download(ctx)
@@ -380,7 +388,7 @@ class F4mFD(FragmentFD):
                 else:
                     raise
 
-            if not fragments_list and live and bootstrap_url:
+            if not fragments_list and not test and live and bootstrap_url:
                 fragments_list = self._update_live_fragments(bootstrap_url, frag_i)
                 total_frags += len(fragments_list)
                 if fragments_list and (fragments_list[0][1] > frag_i + 1):

youtube_dl/downloader/fragment.py

@@ -26,7 +26,11 @@ class FragmentFD(FileDownloader):
         self._start_frag_download(ctx)
 
     def _prepare_frag_download(self, ctx):
-        self.to_screen('[%s] Total fragments: %d' % (self.FD_NAME, ctx['total_frags']))
+        if 'live' not in ctx:
+            ctx['live'] = False
+        self.to_screen(
+            '[%s] Total fragments: %s'
+            % (self.FD_NAME, ctx['total_frags'] if not ctx['live'] else 'unknown (live)'))
         self.report_destination(ctx['filename'])
         dl = HttpQuietDownloader(
             self.ydl,
@@ -34,7 +38,7 @@ class FragmentFD(FileDownloader):
                 'continuedl': True,
                 'quiet': True,
                 'noprogress': True,
-                'ratelimit': self.params.get('ratelimit', None),
+                'ratelimit': self.params.get('ratelimit'),
                 'retries': self.params.get('retries', 0),
                 'test': self.params.get('test', False),
             }
@@ -74,14 +78,14 @@ class FragmentFD(FileDownloader):
             if s['status'] not in ('downloading', 'finished'):
                 return
 
-            frag_total_bytes = s.get('total_bytes') or 0
-
-            estimated_size = (
-                (ctx['complete_frags_downloaded_bytes'] + frag_total_bytes) /
-                (state['frag_index'] + 1) * total_frags)
             time_now = time.time()
-            state['total_bytes_estimate'] = estimated_size
             state['elapsed'] = time_now - start
+            frag_total_bytes = s.get('total_bytes') or 0
+            if not ctx['live']:
+                estimated_size = (
+                    (ctx['complete_frags_downloaded_bytes'] + frag_total_bytes) /
+                    (state['frag_index'] + 1) * total_frags)
+                state['total_bytes_estimate'] = estimated_size
 
             if s['status'] == 'finished':
                 state['frag_index'] += 1
@@ -91,9 +95,10 @@ class FragmentFD(FileDownloader):
             else:
                 frag_downloaded_bytes = s['downloaded_bytes']
                 state['downloaded_bytes'] += frag_downloaded_bytes - ctx['prev_frag_downloaded_bytes']
-                state['eta'] = self.calc_eta(
-                    start, time_now, estimated_size,
-                    state['downloaded_bytes'])
+                if not ctx['live']:
+                    state['eta'] = self.calc_eta(
+                        start, time_now, estimated_size,
+                        state['downloaded_bytes'])
                 state['speed'] = s.get('speed')
                 ctx['prev_frag_downloaded_bytes'] = frag_downloaded_bytes
             self._hook_progress(state)

youtube_dl/downloader/hls.py

@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 import os
 import re
 import subprocess
+import sys
 
 from .common import FileDownloader
 from .fragment import FragmentFD
@@ -39,7 +40,11 @@ class HlsFD(FileDownloader):
                 '-headers',
                 ''.join('%s: %s\r\n' % (key, val) for key, val in headers.items())]
 
-        args += ['-i', url, '-f', 'mp4', '-c', 'copy', '-bsf:a', 'aac_adtstoasc']
+        args += ['-i', url, '-c', 'copy']
+        if self.params.get('hls_use_mpegts', False):
+            args += ['-f', 'mpegts']
+        else:
+            args += ['-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
 
         args = [encodeArgument(opt) for opt in args]
         args.append(encodeFilename(ffpp._ffmpeg_filename_argument(tmpfilename), True))
@@ -53,8 +58,10 @@ class HlsFD(FileDownloader):
             # subprocces.run would send the SIGKILL signal to ffmpeg and the
             # mp4 file couldn't be played, but if we ask ffmpeg to quit it
             # produces a file that is playable (this is mostly useful for live
-            # streams)
-            proc.communicate(b'q')
+            # streams). Note that Windows is not affected and produces playable
+            # files (see https://github.com/rg3/youtube-dl/issues/8300).
+            if sys.platform != 'win32':
+                proc.communicate(b'q')
             raise
         if retval == 0:
             fsize = os.path.getsize(encodeFilename(tmpfilename))

youtube_dl/downloader/http.py

@@ -140,8 +140,8 @@ class HttpFD(FileDownloader):
 
         if data_len is not None:
             data_len = int(data_len) + resume_len
-            min_data_len = self.params.get("min_filesize", None)
-            max_data_len = self.params.get("max_filesize", None)
+            min_data_len = self.params.get('min_filesize')
+            max_data_len = self.params.get('max_filesize')
             if min_data_len is not None and data_len < min_data_len:
                 self.to_screen('\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
                 return False

youtube_dl/downloader/rtmp.py

@@ -94,15 +94,15 @@ class RtmpFD(FileDownloader):
             return proc.returncode
 
         url = info_dict['url']
-        player_url = info_dict.get('player_url', None)
-        page_url = info_dict.get('page_url', None)
-        app = info_dict.get('app', None)
-        play_path = info_dict.get('play_path', None)
-        tc_url = info_dict.get('tc_url', None)
-        flash_version = info_dict.get('flash_version', None)
+        player_url = info_dict.get('player_url')
+        page_url = info_dict.get('page_url')
+        app = info_dict.get('app')
+        play_path = info_dict.get('play_path')
+        tc_url = info_dict.get('tc_url')
+        flash_version = info_dict.get('flash_version')
         live = info_dict.get('rtmp_live', False)
-        conn = info_dict.get('rtmp_conn', None)
-        protocol = info_dict.get('rtmp_protocol', None)
+        conn = info_dict.get('rtmp_conn')
+        protocol = info_dict.get('rtmp_protocol')
         real_time = info_dict.get('rtmp_real_time', False)
         no_resume = info_dict.get('no_resume', False)
         continue_dl = self.params.get('continuedl', True)

youtube_dl/extractor/__init__.py

@@ -20,6 +20,7 @@ from .aftonbladet import AftonbladetIE
 from .airmozilla import AirMozillaIE
 from .aljazeera import AlJazeeraIE
 from .alphaporno import AlphaPornoIE
+from .animeondemand import AnimeOnDemandIE
 from .anitube import AnitubeIE
 from .anysex import AnySexIE
 from .aol import AolIE
@@ -44,13 +45,14 @@ from .arte import (
     ArteTVFutureIE,
     ArteTVCinemaIE,
     ArteTVDDCIE,
+    ArteTVMagazineIE,
     ArteTVEmbedIE,
 )
 from .atresplayer import AtresPlayerIE
 from .atttechchannel import ATTTechChannelIE
 from .audimedia import AudiMediaIE
 from .audiomack import AudiomackIE, AudiomackAlbumIE
-from .azubu import AzubuIE
+from .azubu import AzubuIE, AzubuLiveIE
 from .baidu import BaiduVideoIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
@@ -89,8 +91,15 @@ from .camdemy import (
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .canvas import CanvasIE
+from .cbc import (
+    CBCIE,
+    CBCPlayerIE,
+)
 from .cbs import CBSIE
-from .cbsnews import CBSNewsIE
+from .cbsnews import (
+    CBSNewsIE,
+    CBSNewsLiveVideoIE,
+)
 from .cbssports import CBSSportsIE
 from .ccc import CCCIE
 from .ceskatelevize import CeskaTelevizeIE
@@ -123,6 +132,7 @@ from .comcarcoff import ComCarCoffIE
 from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
 from .condenast import CondeNastIE
 from .cracked import CrackedIE
+from .crackle import CrackleIE
 from .criterion import CriterionIE
 from .crooksandliars import CrooksAndLiarsIE
 from .crunchyroll import (
@@ -142,6 +152,8 @@ from .dailymotion import (
 from .daum import (
     DaumIE,
     DaumClipIE,
+    DaumPlaylistIE,
+    DaumUserIE,
 )
 from .dbtv import DBTVIE
 from .dcn import (
@@ -196,7 +208,10 @@ from .everyonesmixtape import EveryonesMixtapeIE
 from .exfm import ExfmIE
 from .expotv import ExpoTVIE
 from .extremetube import ExtremeTubeIE
-from .facebook import FacebookIE
+from .facebook import (
+    FacebookIE,
+    FacebookPostIE,
+)
 from .faz import FazIE
 from .fc2 import FC2IE
 from .fczenit import FczenitIE
@@ -320,6 +335,7 @@ from .keezmovies import KeezMoviesIE
 from .khanacademy import KhanAcademyIE
 from .kickstarter import KickStarterIE
 from .keek import KeekIE
+from .konserthusetplay import KonserthusetPlayIE
 from .kontrtube import KontrTubeIE
 from .krasview import KrasViewIE
 from .ku6 import Ku6IE
@@ -369,6 +385,7 @@ from .macgamestore import MacGameStoreIE
 from .mailru import MailRuIE
 from .makertv import MakerTVIE
 from .malemotion import MalemotionIE
+from .matchtv import MatchTVIE
 from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
@@ -474,11 +491,13 @@ from .nowtv import (
     NowTVIE,
     NowTVListIE,
 )
+from .noz import NozIE
 from .npo import (
     NPOIE,
     NPOLiveIE,
     NPORadioIE,
     NPORadioFragmentIE,
+    SchoolTVIE,
     VPROIE,
     WNLIE
 )
@@ -522,6 +541,7 @@ from .planetaplay import PlanetaPlayIE
 from .pladform import PladformIE
 from .played import PlayedIE
 from .playfm import PlayFMIE
+from .plays import PlaysTVIE
 from .playtvak import PlaytvakIE
 from .playvid import PlayvidIE
 from .playwire import PlaywireIE
@@ -535,6 +555,7 @@ from .pornhd import PornHdIE
 from .pornhub import (
     PornHubIE,
     PornHubPlaylistIE,
+    PornHubUserVideosIE,
 )
 from .pornotube import PornotubeIE
 from .pornovoisines import PornoVoisinesIE
@@ -602,6 +623,7 @@ from .sbs import SBSIE
 from .scivee import SciVeeIE
 from .screencast import ScreencastIE
 from .screencastomatic import ScreencastOMaticIE
+from .screenjunkies import ScreenJunkiesIE
 from .screenwavemedia import ScreenwaveMediaIE, TeamFourIE
 from .senateisvp import SenateISVPIE
 from .servingsys import ServingSysIE
@@ -776,7 +798,11 @@ from .twitch import (
     TwitchBookmarksIE,
     TwitchStreamIE,
 )
-from .twitter import TwitterCardIE, TwitterIE
+from .twitter import (
+    TwitterCardIE,
+    TwitterIE,
+    TwitterAmplifyIE,
+)
 from .ubu import UbuIE
 from .udemy import (
     UdemyIE,
@@ -812,7 +838,11 @@ from .videomore import (
 )
 from .videopremium import VideoPremiumIE
 from .videott import VideoTtIE
-from .vidme import VidmeIE
+from .vidme import (
+    VidmeIE,
+    VidmeUserIE,
+    VidmeUserLikesIE,
+)
 from .vidzi import VidziIE
 from .vier import VierIE, VierVideosIE
 from .viewster import ViewsterIE

youtube_dl/extractor/acast.py

@@ -8,11 +8,7 @@ from ..compat import compat_str
 from ..utils import int_or_none
 
 
-class ACastBaseIE(InfoExtractor):
-    _API_BASE_URL = 'https://www.acast.com/api/'
-
-
-class ACastIE(ACastBaseIE):
+class ACastIE(InfoExtractor):
     IE_NAME = 'acast'
     _VALID_URL = r'https?://(?:www\.)?acast\.com/(?P<channel>[^/]+)/(?P<id>[^/#?]+)'
     _TEST = {
@@ -23,14 +19,19 @@ class ACastIE(ACastBaseIE):
             'ext': 'mp3',
             'title': '"Where Are You?": Taipei 101, Taiwan',
             'timestamp': 1196172000000,
-            'description': 'md5:0c5d8201dfea2b93218ea986c91eee6e',
+            'description': 'md5:a0b4ef3634e63866b542e5b1199a1a0e',
             'duration': 211,
         }
     }
 
     def _real_extract(self, url):
         channel, display_id = re.match(self._VALID_URL, url).groups()
-        cast_data = self._download_json(self._API_BASE_URL + 'channels/%s/acasts/%s/playback' % (channel, display_id), display_id)
+
+        embed_page = self._download_webpage(
+            re.sub('(?:www\.)?acast\.com', 'embedcdn.acast.com', url), display_id)
+        cast_data = self._parse_json(self._search_regex(
+            r'window\[\'acast/queries\'\]\s*=\s*([^;]+);', embed_page, 'acast data'),
+            display_id)['GetAcast/%s/%s' % (channel, display_id)]
 
         return {
             'id': compat_str(cast_data['id']),
@@ -44,7 +45,7 @@ class ACastIE(ACastBaseIE):
         }
 
 
-class ACastChannelIE(ACastBaseIE):
+class ACastChannelIE(InfoExtractor):
     IE_NAME = 'acast:channel'
     _VALID_URL = r'https?://(?:www\.)?acast\.com/(?P<id>[^/#?]+)'
     _TEST = {
@@ -56,6 +57,7 @@ class ACastChannelIE(ACastBaseIE):
         },
         'playlist_mincount': 20,
     }
+    _API_BASE_URL = 'https://www.acast.com/api/'
 
     @classmethod
     def suitable(cls, url):

youtube_dl/extractor/aenetworks.py

@@ -28,7 +28,7 @@ class AENetworksIE(InfoExtractor):
         'info_dict': {
             'id': 'eg47EERs_JsZ',
             'ext': 'mp4',
-            'title': "Winter Is Coming",
+            'title': 'Winter Is Coming',
             'description': 'md5:641f424b7a19d8e24f26dea22cf59d74',
         },
         'params': {

youtube_dl/extractor/allocine.py

@@ -8,6 +8,8 @@ from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
     qualities,
+    unescapeHTML,
+    xpath_element,
 )
 
 
@@ -31,7 +33,7 @@ class AllocineIE(InfoExtractor):
             'id': '19540403',
             'ext': 'mp4',
             'title': 'Planes 2 Bande-annonce VF',
-            'description': 'md5:eeaffe7c2d634525e21159b93acf3b1e',
+            'description': 'Regardez la bande annonce du film Planes 2 (Planes 2 Bande-annonce VF). Planes 2, un film de Roberts Gannaway',
             'thumbnail': 're:http://.*\.jpg',
         },
     }, {
@@ -41,7 +43,7 @@ class AllocineIE(InfoExtractor):
             'id': '19544709',
             'ext': 'mp4',
             'title': 'Dragons 2 - Bande annonce finale VF',
-            'description': 'md5:71742e3a74b0d692c7fce0dd2017a4ac',
+            'description': 'md5:601d15393ac40f249648ef000720e7e3',
             'thumbnail': 're:http://.*\.jpg',
         },
     }, {
@@ -59,14 +61,18 @@ class AllocineIE(InfoExtractor):
         if typ == 'film':
             video_id = self._search_regex(r'href="/video/player_gen_cmedia=([0-9]+).+"', webpage, 'video id')
         else:
-            player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player')
-
-            player_data = json.loads(player)
-            video_id = compat_str(player_data['refMedia'])
+            player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player', default=None)
+            if player:
+                player_data = json.loads(player)
+                video_id = compat_str(player_data['refMedia'])
+            else:
+                model = self._search_regex(r'data-model="([^"]+)">', webpage, 'data model')
+                model_data = self._parse_json(unescapeHTML(model), display_id)
+                video_id = compat_str(model_data['id'])
 
         xml = self._download_xml('http://www.allocine.fr/ws/AcVisiondataV4.ashx?media=%s' % video_id, display_id)
 
-        video = xml.find('.//AcVisionVideo').attrib
+        video = xpath_element(xml, './/AcVisionVideo').attrib
         quality = qualities(['ld', 'md', 'hd'])
 
         formats = []

youtube_dl/extractor/animeondemand.py

@@ -0,0 +1,160 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_urlparse
+from ..utils import (
+    determine_ext,
+    encode_dict,
+    ExtractorError,
+    sanitized_Request,
+    urlencode_postdata,
+)
+
+
+class AnimeOnDemandIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?anime-on-demand\.de/anime/(?P<id>\d+)'
+    _LOGIN_URL = 'https://www.anime-on-demand.de/users/sign_in'
+    _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
+    _NETRC_MACHINE = 'animeondemand'
+    _TEST = {
+        'url': 'https://www.anime-on-demand.de/anime/161',
+        'info_dict': {
+            'id': '161',
+            'title': 'Grimgar, Ashes and Illusions (OmU)',
+            'description': 'md5:6681ce3c07c7189d255ac6ab23812d31',
+        },
+        'playlist_mincount': 4,
+    }
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+
+        login_page = self._download_webpage(
+            self._LOGIN_URL, None, 'Downloading login page')
+
+        login_form = self._form_hidden_inputs('new_user', login_page)
+
+        login_form.update({
+            'user[login]': username,
+            'user[password]': password,
+        })
+
+        post_url = self._search_regex(
+            r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
+            'post url', default=self._LOGIN_URL, group='url')
+
+        if not post_url.startswith('http'):
+            post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
+
+        request = sanitized_Request(
+            post_url, urlencode_postdata(encode_dict(login_form)))
+        request.add_header('Referer', self._LOGIN_URL)
+
+        response = self._download_webpage(
+            request, None, 'Logging in as %s' % username)
+
+        if all(p not in response for p in ('>Logout<', 'href="/users/sign_out"')):
+            error = self._search_regex(
+                r'<p class="alert alert-danger">(.+?)</p>',
+                response, 'error', default=None)
+            if error:
+                raise ExtractorError('Unable to login: %s' % error, expected=True)
+            raise ExtractorError('Unable to log in')
+
+    def _real_initialize(self):
+        self._login()
+
+    def _real_extract(self, url):
+        anime_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, anime_id)
+
+        if 'data-playlist=' not in webpage:
+            self._download_webpage(
+                self._APPLY_HTML5_URL, anime_id,
+                'Activating HTML5 beta', 'Unable to apply HTML5 beta')
+            webpage = self._download_webpage(url, anime_id)
+
+        csrf_token = self._html_search_meta(
+            'csrf-token', webpage, 'csrf token', fatal=True)
+
+        anime_title = self._html_search_regex(
+            r'(?s)<h1[^>]+itemprop="name"[^>]*>(.+?)</h1>',
+            webpage, 'anime name')
+        anime_description = self._html_search_regex(
+            r'(?s)<div[^>]+itemprop="description"[^>]*>(.+?)</div>',
+            webpage, 'anime description', default=None)
+
+        entries = []
+
+        for episode_html in re.findall(r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', webpage):
+            m = re.search(
+                r'class="episodebox-title"[^>]+title="Episode (?P<number>\d+) - (?P<title>.+?)"', episode_html)
+            if not m:
+                continue
+
+            episode_number = int(m.group('number'))
+            episode_title = m.group('title')
+            video_id = 'episode-%d' % episode_number
+
+            common_info = {
+                'id': video_id,
+                'series': anime_title,
+                'episode': episode_title,
+                'episode_number': episode_number,
+            }
+
+            formats = []
+
+            playlist_url = self._search_regex(
+                r'data-playlist=(["\'])(?P<url>.+?)\1',
+                episode_html, 'data playlist', default=None, group='url')
+            if playlist_url:
+                request = sanitized_Request(
+                    compat_urlparse.urljoin(url, playlist_url),
+                    headers={
+                        'X-Requested-With': 'XMLHttpRequest',
+                        'X-CSRF-Token': csrf_token,
+                        'Referer': url,
+                        'Accept': 'application/json, text/javascript, */*; q=0.01',
+                    })
+
+                playlist = self._download_json(
+                    request, video_id, 'Downloading playlist JSON', fatal=False)
+                if playlist:
+                    playlist = playlist['playlist'][0]
+                    title = playlist['title']
+                    description = playlist.get('description')
+                    for source in playlist.get('sources', []):
+                        file_ = source.get('file')
+                        if file_ and determine_ext(file_) == 'm3u8':
+                            formats = self._extract_m3u8_formats(
+                                file_, video_id, 'mp4',
+                                entry_protocol='m3u8_native', m3u8_id='hls')
+
+            if formats:
+                f = common_info.copy()
+                f.update({
+                    'title': title,
+                    'description': description,
+                    'formats': formats,
+                })
+                entries.append(f)
+
+            m = re.search(
+                r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>Teaser<',
+                episode_html)
+            if m:
+                f = common_info.copy()
+                f.update({
+                    'id': '%s-teaser' % f['id'],
+                    'title': m.group('title'),
+                    'url': compat_urlparse.urljoin(url, m.group('href')),
+                })
+                entries.append(f)
+
+        return self.playlist_result(entries, anime_id, anime_title, anime_description)

youtube_dl/extractor/appletrailers.py

@@ -12,7 +12,7 @@ from ..utils import (
 
 class AppleTrailersIE(InfoExtractor):
     IE_NAME = 'appletrailers'
-    _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/(?:trailers|ca)/(?P<company>[^/]+)/(?P<movie>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.|movie)?trailers\.apple\.com/(?:trailers|ca)/(?P<company>[^/]+)/(?P<movie>[^/]+)'
     _TESTS = [{
         'url': 'http://trailers.apple.com/trailers/wb/manofsteel/',
         'info_dict': {
@@ -73,6 +73,9 @@ class AppleTrailersIE(InfoExtractor):
     }, {
         'url': 'http://trailers.apple.com/ca/metropole/autrui/',
         'only_matching': True,
+    }, {
+        'url': 'http://movietrailers.apple.com/trailers/focus_features/kuboandthetwostrings/',
+        'only_matching': True,
     }]
 
     _JSON_RE = r'iTunes.playURL\((.*?)\);'

youtube_dl/extractor/arte.py

@@ -13,6 +13,7 @@ from ..utils import (
     unified_strdate,
     get_element_by_attribute,
     int_or_none,
+    NO_DEFAULT,
     qualities,
 )
 
@@ -22,7 +23,7 @@ from ..utils import (
 
 
 class ArteTvIE(InfoExtractor):
-    _VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html'
+    _VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de|en|es)/.*-(?P<id>.*?)\.html'
     IE_NAME = 'arte.tv'
 
     def _real_extract(self, url):
@@ -62,7 +63,7 @@ class ArteTvIE(InfoExtractor):
 
 class ArteTVPlus7IE(InfoExtractor):
     IE_NAME = 'arte.tv:+7'
-    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&+])'
 
     @classmethod
     def _extract_url_info(cls, url):
@@ -93,12 +94,40 @@ class ArteTVPlus7IE(InfoExtractor):
         json_url = self._html_search_regex(
             patterns, webpage, 'json vp url', default=None)
         if not json_url:
-            iframe_url = self._html_search_regex(
-                r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
-                webpage, 'iframe url', group='url')
-            json_url = compat_parse_qs(
-                compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
-        return self._extract_from_json_url(json_url, video_id, lang)
+            def find_iframe_url(webpage, default=NO_DEFAULT):
+                return self._html_search_regex(
+                    r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
+                    webpage, 'iframe url', group='url', default=default)
+
+            iframe_url = find_iframe_url(webpage, None)
+            if not iframe_url:
+                embed_url = self._html_search_regex(
+                    r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None)
+                if embed_url:
+                    player = self._download_json(
+                        embed_url, video_id, 'Downloading player page')
+                    iframe_url = find_iframe_url(player['html'])
+            # en and es URLs produce react-based pages with different layout (e.g.
+            # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world)
+            if not iframe_url:
+                program = self._search_regex(
+                    r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n',
+                    webpage, 'program', default=None)
+                if program:
+                    embed_html = self._parse_json(program, video_id)
+                    if embed_html:
+                        iframe_url = find_iframe_url(embed_html['embed_html'])
+            if iframe_url:
+                json_url = compat_parse_qs(
+                    compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
+        if json_url:
+            return self._extract_from_json_url(json_url, video_id, lang)
+        # Differend kind of embed URL (e.g.
+        # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
+        embed_url = self._search_regex(
+            r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1',
+            webpage, 'embed url', group='url')
+        return self.url_result(embed_url)
 
     def _extract_from_json_url(self, json_url, video_id, lang):
         info = self._download_json(json_url, video_id)
@@ -106,7 +135,7 @@ class ArteTVPlus7IE(InfoExtractor):
 
         upload_date_str = player_info.get('shootingDate')
         if not upload_date_str:
-            upload_date_str = player_info.get('VDA', '').split(' ')[0]
+            upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
 
         title = player_info['VTI'].strip()
         subtitle = player_info.get('VSU', '').strip()
@@ -122,27 +151,30 @@ class ArteTVPlus7IE(InfoExtractor):
         }
         qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
 
+        LANGS = {
+            'fr': 'F',
+            'de': 'A',
+            'en': 'E[ANG]',
+            'es': 'E[ESP]',
+        }
+
         formats = []
         for format_id, format_dict in player_info['VSR'].items():
             f = dict(format_dict)
             versionCode = f.get('versionCode')
-
-            langcode = {
-                'fr': 'F',
-                'de': 'A',
-            }.get(lang, lang)
-            lang_rexs = [r'VO?%s' % langcode, r'VO?.-ST%s' % langcode]
-            lang_pref = (
-                None if versionCode is None else (
-                    10 if any(re.match(r, versionCode) for r in lang_rexs)
-                    else -10))
+            langcode = LANGS.get(lang, lang)
+            lang_rexs = [r'VO?%s-' % re.escape(langcode), r'VO?.-ST%s$' % re.escape(langcode)]
+            lang_pref = None
+            if versionCode:
+                matched_lang_rexs = [r for r in lang_rexs if re.match(r, versionCode)]
+                lang_pref = -10 if not matched_lang_rexs else 10 * len(matched_lang_rexs)
             source_pref = 0
             if versionCode is not None:
                 # The original version with subtitles has lower relevance
-                if re.match(r'VO-ST(F|A)', versionCode):
+                if re.match(r'VO-ST(F|A|E)', versionCode):
                     source_pref -= 10
                 # The version with sourds/mal subtitles has also lower relevance
-                elif re.match(r'VO?(F|A)-STM\1', versionCode):
+                elif re.match(r'VO?(F|A|E)-STM\1', versionCode):
                     source_pref -= 9
             format = {
                 'format_id': format_id,
@@ -175,7 +207,7 @@ class ArteTVPlus7IE(InfoExtractor):
 # It also uses the arte_vp_url url from the webpage to extract the information
 class ArteTVCreativeIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:creative'
-    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/(?:magazine?/)?(?P<id>[^?#]+)'
+    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:magazine?/)?(?P<id>[^/?#&]+)'
 
     _TESTS = [{
         'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
@@ -199,7 +231,7 @@ class ArteTVCreativeIE(ArteTVPlus7IE):
 
 class ArteTVFutureIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:future'
-    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(?P<id>.+)'
+    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
 
     _TESTS = [{
         'url': 'http://future.arte.tv/fr/info-sciences/les-ecrevisses-aussi-sont-anxieuses',
@@ -207,6 +239,7 @@ class ArteTVFutureIE(ArteTVPlus7IE):
             'id': '050940-028-A',
             'ext': 'mp4',
             'title': 'Les écrevisses aussi peuvent être anxieuses',
+            'upload_date': '20140902',
         },
     }, {
         'url': 'http://future.arte.tv/fr/la-science-est-elle-responsable',
@@ -216,7 +249,7 @@ class ArteTVFutureIE(ArteTVPlus7IE):
 
 class ArteTVDDCIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:ddc'
-    _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
+    _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>[^/?#&]+)'
 
     def _real_extract(self, url):
         video_id, lang = self._extract_url_info(url)
@@ -234,7 +267,7 @@ class ArteTVDDCIE(ArteTVPlus7IE):
 
 class ArteTVConcertIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:concert'
-    _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'
+    _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
 
     _TEST = {
         'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
@@ -251,7 +284,7 @@ class ArteTVConcertIE(ArteTVPlus7IE):
 
 class ArteTVCinemaIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:cinema'
-    _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'
+    _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
 
     _TEST = {
         'url': 'http://cinema.arte.tv/de/node/38291',
@@ -266,6 +299,37 @@ class ArteTVCinemaIE(ArteTVPlus7IE):
     }
 
 
+class ArteTVMagazineIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:magazine'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/magazine/[^/]+/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        # Embedded via <iframe src="http://www.arte.tv/arte_vp/index.php?json_url=..."
+        'url': 'http://www.arte.tv/magazine/trepalium/fr/entretien-avec-le-realisateur-vincent-lannoo-trepalium',
+        'md5': '2a9369bcccf847d1c741e51416299f25',
+        'info_dict': {
+            'id': '065965-000-A',
+            'ext': 'mp4',
+            'title': 'Trepalium - Extrait Ep.01',
+            'upload_date': '20160121',
+        },
+    }, {
+        # Embedded via <iframe src="http://www.arte.tv/guide/fr/embed/054813-004-A/medium"
+        'url': 'http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium',
+        'md5': 'fedc64fc7a946110fe311634e79782ca',
+        'info_dict': {
+            'id': '054813-004_PLUS7-F',
+            'ext': 'mp4',
+            'title': 'Trepalium (4/6)',
+            'description': 'md5:10057003c34d54e95350be4f9b05cb40',
+            'upload_date': '20160218',
+        },
+    }, {
+        'url': 'http://www.arte.tv/magazine/metropolis/de/frank-woeste-german-paris-metropolis',
+        'only_matching': True,
+    }]
+
+
 class ArteTVEmbedIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:embed'
     _VALID_URL = r'''(?x)

youtube_dl/extractor/azubu.py

@@ -3,7 +3,11 @@ from __future__ import unicode_literals
 import json
 
 from .common import InfoExtractor
-from ..utils import float_or_none
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    sanitized_Request,
+)
 
 
 class AzubuIE(InfoExtractor):
@@ -91,3 +95,37 @@ class AzubuIE(InfoExtractor):
             'view_count': view_count,
             'formats': formats,
         }
+
+
+class AzubuLiveIE(InfoExtractor):
+    _VALID_URL = r'http://www.azubu.tv/(?P<id>[^/]+)$'
+
+    _TEST = {
+        'url': 'http://www.azubu.tv/MarsTVMDLen',
+        'only_matching': True,
+    }
+
+    def _real_extract(self, url):
+        user = self._match_id(url)
+
+        info = self._download_json(
+            'http://api.azubu.tv/public/modules/last-video/{0}/info'.format(user),
+            user)['data']
+        if info['type'] != 'STREAM':
+            raise ExtractorError('{0} is not streaming live'.format(user), expected=True)
+
+        req = sanitized_Request(
+            'https://edge-elb.api.brightcove.com/playback/v1/accounts/3361910549001/videos/ref:' + info['reference_id'])
+        req.add_header('Accept', 'application/json;pk=BCpkADawqM1gvI0oGWg8dxQHlgT8HkdE2LnAlWAZkOlznO39bSZX726u4JqnDsK3MDXcO01JxXK2tZtJbgQChxgaFzEVdHRjaDoxaOu8hHOO8NYhwdxw9BzvgkvLUlpbDNUuDoc4E4wxDToV')
+        bc_info = self._download_json(req, user)
+        m3u8_url = next(source['src'] for source in bc_info['sources'] if source['container'] == 'M2TS')
+        formats = self._extract_m3u8_formats(m3u8_url, user, ext='mp4')
+
+        return {
+            'id': info['id'],
+            'title': self._live_title(info['title']),
+            'uploader_id': user,
+            'formats': formats,
+            'is_live': True,
+            'thumbnail': bc_info['poster'],
+        }

youtube_dl/extractor/bbc.py

@@ -86,7 +86,7 @@ class BBCCoUkIE(InfoExtractor):
                 '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.",
+                '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': {
@@ -193,6 +193,19 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+        }, {
+            # compact player (https://github.com/rg3/youtube-dl/issues/8147)
+            'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
+            'info_dict': {
+                'id': 'p028bfkj',
+                'ext': 'flv',
+                'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
+                'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
         }, {
             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
             'only_matching': True,
@@ -482,9 +495,11 @@ class BBCCoUkIE(InfoExtractor):
         if programme_id:
             formats, subtitles = self._download_media_selector(programme_id)
             title = self._og_search_title(webpage, default=None) or self._html_search_regex(
-                r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>', webpage, 'title')
+                (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
+                 r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
             description = self._search_regex(
-                r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
+                (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
+                 r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
                 webpage, 'description', default=None)
             if not description:
                 description = self._html_search_meta('description', webpage)

youtube_dl/extractor/bpb.py

@@ -1,7 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..utils import (
+    js_to_json,
+    determine_ext,
+)
 
 
 class BpbIE(InfoExtractor):
@@ -10,7 +16,8 @@ class BpbIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.bpb.de/mediathek/297/joachim-gauck-zu-1989-und-die-erinnerung-an-die-ddr',
-        'md5': '0792086e8e2bfbac9cdf27835d5f2093',
+        # md5 fails in Python 2.6 due to buggy server response and wrong handling of urllib2
+        'md5': 'c4f84c8a8044ca9ff68bb8441d300b3f',
         'info_dict': {
             'id': '297',
             'ext': 'mp4',
@@ -25,13 +32,26 @@ class BpbIE(InfoExtractor):
 
         title = self._html_search_regex(
             r'<h2 class="white">(.*?)</h2>', webpage, 'title')
-        video_url = self._html_search_regex(
-            r'(http://film\.bpb\.de/player/dokument_[0-9]+\.mp4)',
-            webpage, 'video URL')
+        video_info_dicts = re.findall(
+            r"({\s*src:\s*'http://film\.bpb\.de/[^}]+})", webpage)
+
+        formats = []
+        for video_info in video_info_dicts:
+            video_info = self._parse_json(video_info, video_id, transform_source=js_to_json)
+            quality = video_info['quality']
+            video_url = video_info['src']
+            formats.append({
+                'url': video_url,
+                'preference': 10 if quality == 'high' else 0,
+                'format_note': quality,
+                'format_id': '%s-%s' % (quality, determine_ext(video_url)),
+            })
+
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'url': video_url,
+            'formats': formats,
             'title': title,
             'description': self._og_search_description(webpage),
         }

youtube_dl/extractor/canvas.py

@@ -6,7 +6,7 @@ from ..utils import float_or_none
 
 class CanvasIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?canvas\.be/video/(?:[^/]+/)*(?P<id>[^/?#&]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
         'md5': 'ea838375a547ac787d4064d8c7860a6c',
         'info_dict': {
@@ -18,7 +18,27 @@ class CanvasIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg$',
             'duration': 49.02,
         }
-    }
+    }, {
+        # with subtitles
+        'url': 'http://www.canvas.be/video/panorama/2016/pieter-0167',
+        'info_dict': {
+            'id': 'mz-ast-5240ff21-2d30-4101-bba6-92b5ec67c625',
+            'display_id': 'pieter-0167',
+            'ext': 'mp4',
+            'title': 'Pieter 0167',
+            'description': 'md5:943cd30f48a5d29ba02c3a104dc4ec4e',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 2553.08,
+            'subtitles': {
+                'nl': [{
+                    'ext': 'vtt',
+                }],
+            },
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
@@ -54,6 +74,14 @@ class CanvasIE(InfoExtractor):
                 })
         self._sort_formats(formats)
 
+        subtitles = {}
+        subtitle_urls = data.get('subtitleUrls')
+        if isinstance(subtitle_urls, list):
+            for subtitle in subtitle_urls:
+                subtitle_url = subtitle.get('url')
+                if subtitle_url and subtitle.get('type') == 'CLOSED':
+                    subtitles.setdefault('nl', []).append({'url': subtitle_url})
+
         return {
             'id': video_id,
             'display_id': display_id,
@@ -62,4 +90,5 @@ class CanvasIE(InfoExtractor):
             'formats': formats,
             'duration': float_or_none(data.get('duration'), 1000),
             'thumbnail': data.get('posterImageUrl'),
+            'subtitles': subtitles,
         }

youtube_dl/extractor/cbc.py

@@ -0,0 +1,113 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import js_to_json
+
+
+class CBCIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?:[^/]+/)+(?P<id>[^/?#]+)'
+    _TESTS = [{
+        # with mediaId
+        'url': 'http://www.cbc.ca/22minutes/videos/clips-season-23/don-cherry-play-offs',
+        'info_dict': {
+            'id': '2682904050',
+            'ext': 'flv',
+            'title': 'Don Cherry – All-Stars',
+            'description': 'Don Cherry has a bee in his bonnet about AHL player John Scott because that guy’s got heart.',
+            'timestamp': 1454475540,
+            'upload_date': '20160203',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        # with clipId
+        'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live',
+        'info_dict': {
+            'id': '2487345465',
+            'ext': 'flv',
+            'title': 'Robin Williams freestyles on 90 Minutes Live',
+            'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.',
+            'upload_date': '19700101',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        # multiple iframes
+        'url': 'http://www.cbc.ca/natureofthings/blog/birds-eye-view-from-vancouvers-burrard-street-bridge-how-we-got-the-shot',
+        'playlist': [{
+            'info_dict': {
+                'id': '2680832926',
+                'ext': 'flv',
+                'title': 'An Eagle\'s-Eye View Off Burrard Bridge',
+                'description': 'Hercules the eagle flies from Vancouver\'s Burrard Bridge down to a nearby park with a mini-camera strapped to his back.',
+                'upload_date': '19700101',
+            },
+        }, {
+            'info_dict': {
+                'id': '2658915080',
+                'ext': 'flv',
+                'title': 'Fly like an eagle!',
+                'description': 'Eagle equipped with a mini camera flies from the world\'s tallest tower',
+                'upload_date': '19700101',
+            },
+        }],
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if CBCPlayerIE.suitable(url) else super(CBCIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        player_init = self._search_regex(
+            r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage, 'player init',
+            default=None)
+        if player_init:
+            player_info = self._parse_json(player_init, display_id, js_to_json)
+            media_id = player_info.get('mediaId')
+            if not media_id:
+                clip_id = player_info['clipId']
+                media_id = self._download_json(
+                    'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
+                    clip_id)['entries'][0]['id'].split('/')[-1]
+            return self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
+        else:
+            entries = [self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id) for media_id in re.findall(r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"', webpage)]
+            return self.playlist_result(entries)
+
+
+class CBCPlayerIE(InfoExtractor):
+    _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.cbc.ca/player/play/2683190193',
+        'info_dict': {
+            'id': '2683190193',
+            'ext': 'flv',
+            'title': 'Gerry Runs a Sweat Shop',
+            'description': 'md5:b457e1c01e8ff408d9d801c1c2cd29b0',
+            'timestamp': 1455067800,
+            'upload_date': '20160210',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        return self.url_result(
+            'http://feed.theplatform.com/f/ExhSPC/vms_5akSXx4Ng_Zn?byGuid=%s' % video_id,
+            'ThePlatformFeed', video_id)

youtube_dl/extractor/cbsnews.py

@@ -1,16 +1,17 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-import json
-
 from .common import InfoExtractor
-from ..utils import remove_start
+from .theplatform import ThePlatformIE
+from ..utils import (
+    parse_duration,
+    find_xpath_attr,
+)
 
 
-class CBSNewsIE(InfoExtractor):
+class CBSNewsIE(ThePlatformIE):
     IE_DESC = 'CBS News'
-    _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:[^/]+/)+(?P<id>[\da-z_-]+)'
+    _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
 
     _TESTS = [
         {
@@ -31,7 +32,7 @@ class CBSNewsIE(InfoExtractor):
             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
             'info_dict': {
                 'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'duration': 205,
@@ -42,60 +43,46 @@ class CBSNewsIE(InfoExtractor):
                 },
             },
             'params': {
-                # rtmp download
+                # m3u8 download
                 'skip_download': True,
             },
         },
     ]
 
+    def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
+        closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL')
+        return {
+            'en': [{
+                'ext': 'ttml',
+                'url': closed_caption_e.attrib['value'],
+            }]
+        } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
+
     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)
 
-        video_info = json.loads(self._html_search_regex(
+        video_info = self._parse_json(self._html_search_regex(
             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
-            webpage, 'video JSON info'))
+            webpage, 'video JSON info'), video_id)
 
         item = video_info['item'] if 'item' in video_info else video_info
         title = item.get('articleTitle') or item.get('hed')
         duration = item.get('duration')
         thumbnail = item.get('mediaImage') or item.get('thumbnail')
 
+        subtitles = {}
         formats = []
         for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
-            uri = item.get('media' + format_id + 'URI')
-            if not uri:
+            pid = item.get('media' + format_id)
+            if not pid:
                 continue
-            uri = remove_start(uri, '{manifest:none}')
-            fmt = {
-                'url': uri,
-                'format_id': format_id,
-            }
-            if uri.startswith('rtmp'):
-                play_path = re.sub(
-                    r'{slistFilePath}', '',
-                    uri.split('<break>')[-1].split('{break}')[-1])
-                play_path = re.sub(
-                    r'{manifest:.+}.*$', '', play_path)
-                fmt.update({
-                    'app': 'ondemand?auth=cbs',
-                    'play_path': 'mp4:' + play_path,
-                    'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf',
-                    'page_url': 'http://www.cbsnews.com',
-                    'ext': 'flv',
-                })
-            elif uri.endswith('.m3u8'):
-                fmt['ext'] = 'mp4'
-            formats.append(fmt)
-
-        subtitles = {}
-        if 'mpxRefId' in video_info:
-            subtitles['en'] = [{
-                'ext': 'ttml',
-                'url': 'http://www.cbsnews.com/videos/captions/%s.adb_xml' % video_info['mpxRefId'],
-            }]
+            release_url = 'http://link.theplatform.com/s/dJ5BDC/%s?format=SMIL&mbr=true' % pid
+            tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % pid)
+            formats.extend(tp_formats)
+            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
@@ -105,3 +92,41 @@ class CBSNewsIE(InfoExtractor):
             'formats': formats,
             'subtitles': subtitles,
         }
+
+
+class CBSNewsLiveVideoIE(InfoExtractor):
+    IE_DESC = 'CBS News Live Videos'
+    _VALID_URL = r'http://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
+
+    _TEST = {
+        'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
+        'info_dict': {
+            'id': 'clinton-sanders-prepare-to-face-off-in-nh',
+            'ext': 'flv',
+            'title': 'Clinton, Sanders Prepare To Face Off In NH',
+            'duration': 334,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_info = self._parse_json(self._html_search_regex(
+            r'data-story-obj=\'({.+?})\'', webpage, 'video JSON info'), video_id)['story']
+
+        hdcore_sign = 'hdcore=3.3.1'
+        f4m_formats = self._extract_f4m_formats(video_info['url'] + '&' + hdcore_sign, video_id)
+        if f4m_formats:
+            for entry in f4m_formats:
+                # URLs without the extra param induce an 404 error
+                entry.update({'extra_param_to_segment_url': hdcore_sign})
+
+        return {
+            'id': video_id,
+            'title': video_info['headline'],
+            'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'),
+            'duration': parse_duration(video_info.get('segmentDur')),
+            'formats': f4m_formats,
+        }

youtube_dl/extractor/ccc.py

@@ -45,7 +45,7 @@ class CCCIE(InfoExtractor):
         title = self._html_search_regex(
             r'(?s)<h1>(.*?)</h1>', webpage, 'title')
         description = self._html_search_regex(
-            r"(?s)<h3>About</h3>(.+?)<h3>",
+            r'(?s)<h3>About</h3>(.+?)<h3>',
             webpage, 'description', fatal=False)
         upload_date = unified_strdate(self._html_search_regex(
             r"(?s)<span[^>]+class='[^']*fa-calendar-o'[^>]*>(.+?)</span>",

youtube_dl/extractor/ceskatelevize.py

@@ -177,16 +177,16 @@ class CeskaTelevizeIE(InfoExtractor):
             for divider in [1000, 60, 60, 100]:
                 components.append(msec % divider)
                 msec //= divider
-            return "{3:02}:{2:02}:{1:02},{0:03}".format(*components)
+            return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
 
         def _fix_subtitle(subtitle):
             for line in subtitle.splitlines():
-                m = re.match(r"^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$", line)
+                m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
                 if m:
                     yield m.group(1)
                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
-                    yield "{0} --> {1}".format(start, stop)
+                    yield '{0} --> {1}'.format(start, stop)
                 else:
                     yield line
 
-        return "\r\n".join(_fix_subtitle(subtitles))
+        return '\r\n'.join(_fix_subtitle(subtitles))

youtube_dl/extractor/cnn.py

@@ -26,14 +26,14 @@ class CNNIE(InfoExtractor):
             'upload_date': '20130609',
         },
     }, {
-        "url": "http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
-        "md5": "b5cc60c60a3477d185af8f19a2a26f4e",
-        "info_dict": {
+        'url': 'http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29',
+        'md5': 'b5cc60c60a3477d185af8f19a2a26f4e',
+        'info_dict': {
             'id': 'us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology',
             'ext': 'mp4',
-            "title": "Student's epic speech stuns new freshmen",
-            "description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
-            "upload_date": "20130821",
+            'title': "Student's epic speech stuns new freshmen",
+            'description': "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
+            'upload_date': '20130821',
         }
     }, {
         'url': 'http://www.cnn.com/video/data/2.0/video/living/2014/12/22/growing-america-nashville-salemtown-board-episode-1.hln.html',

youtube_dl/extractor/collegerama.py

@@ -46,9 +46,9 @@ class CollegeRamaIE(InfoExtractor):
         video_id = self._match_id(url)
 
         player_options_request = {
-            "getPlayerOptionsRequest": {
-                "ResourceId": video_id,
-                "QueryString": "",
+            'getPlayerOptionsRequest': {
+                'ResourceId': video_id,
+                'QueryString': '',
             }
         }
 

youtube_dl/extractor/comcarcoff.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     int_or_none,
     parse_duration,
@@ -14,14 +15,13 @@ class ComCarCoffIE(InfoExtractor):
     _TESTS = [{
         'url': 'http://comediansincarsgettingcoffee.com/miranda-sings-happy-thanksgiving-miranda/',
         'info_dict': {
-            'id': 'miranda-sings-happy-thanksgiving-miranda',
+            'id': '2494164',
             'ext': 'mp4',
             'upload_date': '20141127',
             'timestamp': 1417107600,
             'duration': 1232,
             'title': 'Happy Thanksgiving Miranda',
             'description': 'Jerry Seinfeld and his special guest Miranda Sings cruise around town in search of coffee, complaining and apologizing along the way.',
-            'thumbnail': 'http://ccc.crackle.com/images/s5e4_thumb.jpg',
         },
         'params': {
             'skip_download': 'requires ffmpeg',
@@ -39,15 +39,14 @@ class ComCarCoffIE(InfoExtractor):
                 r'window\.app\s*=\s*({.+?});\n', webpage, 'full data json'),
             display_id)['videoData']
 
-        video_id = full_data['activeVideo']['video']
-        video_data = full_data.get('videos', {}).get(video_id) or full_data['singleshots'][video_id]
+        display_id = full_data['activeVideo']['video']
+        video_data = full_data.get('videos', {}).get(display_id) or full_data['singleshots'][display_id]
+        video_id = compat_str(video_data['mediaId'])
         thumbnails = [{
             'url': video_data['images']['thumb'],
         }, {
             'url': video_data['images']['poster'],
         }]
-        formats = self._extract_m3u8_formats(
-            video_data['mediaUrl'], video_id, ext='mp4')
 
         timestamp = int_or_none(video_data.get('pubDateTime')) or parse_iso8601(
             video_data.get('pubDate'))
@@ -55,6 +54,8 @@ class ComCarCoffIE(InfoExtractor):
             video_data.get('duration'))
 
         return {
+            '_type': 'url_transparent',
+            'url': 'crackle:%s' % video_id,
             'id': video_id,
             'display_id': display_id,
             'title': video_data['title'],
@@ -62,6 +63,7 @@ class ComCarCoffIE(InfoExtractor):
             'timestamp': timestamp,
             'duration': duration,
             'thumbnails': thumbnails,
-            'formats': formats,
+            'season_number': int_or_none(video_data.get('season')),
+            'episode_number': int_or_none(video_data.get('episode')),
             'webpage_url': 'http://comediansincarsgettingcoffee.com/%s' % (video_data.get('urlSlug', video_data.get('slug'))),
         }

youtube_dl/extractor/comedycentral.py

@@ -16,11 +16,11 @@ from ..utils import (
 
 class ComedyCentralIE(MTVServicesInfoExtractor):
     _VALID_URL = r'''(?x)https?://(?:www\.)?cc\.com/
-        (video-clips|episodes|cc-studios|video-collections|full-episodes)
+        (video-clips|episodes|cc-studios|video-collections|full-episodes|shows)
         /(?P<title>.*)'''
     _FEED_URL = 'http://comedycentral.com/feeds/mrss/'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.cc.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
         'md5': 'c4f48e9eda1b16dd10add0744344b6d8',
         'info_dict': {
@@ -29,7 +29,10 @@ class ComedyCentralIE(MTVServicesInfoExtractor):
             'title': 'CC:Stand-Up|Greg Fitzsimmons: Life on Stage|Uncensored - Too Good of a Mother',
             'description': 'After a certain point, breastfeeding becomes c**kblocking.',
         },
-    }
+    }, {
+        'url': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/interviews/6yx39d/exclusive-rand-paul-extended-interview',
+        'only_matching': True,
+    }]
 
 
 class ComedyCentralShowsIE(MTVServicesInfoExtractor):
@@ -192,7 +195,7 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
             if len(altMovieParams) == 0:
                 raise ExtractorError('unable to find Flash URL in webpage ' + url)
             else:
-                mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
+                mMovieParams = [('http://media.mtvnservices.com/' + altMovieParams[0], altMovieParams[0])]
 
         uri = mMovieParams[0][1]
         # Correct cc.com in uri

youtube_dl/extractor/common.py

@@ -10,6 +10,7 @@ import re
 import socket
 import sys
 import time
+import math
 
 from ..compat import (
     compat_cookiejar,
@@ -44,6 +45,8 @@ from ..utils import (
     xpath_text,
     xpath_with_ns,
     determine_protocol,
+    parse_duration,
+    mimetype2ext,
 )
 
 
@@ -634,7 +637,7 @@ class InfoExtractor(object):
         downloader_params = self._downloader.params
 
         # Attempt to use provided username and password or .netrc data
-        if downloader_params.get('username', None) is not None:
+        if downloader_params.get('username') is not None:
             username = downloader_params['username']
             password = downloader_params['password']
         elif downloader_params.get('usenetrc', False):
@@ -661,7 +664,7 @@ class InfoExtractor(object):
             return None
         downloader_params = self._downloader.params
 
-        if downloader_params.get('twofactor', None) is not None:
+        if downloader_params.get('twofactor') is not None:
             return downloader_params['twofactor']
 
         return compat_getpass('Type %s and press [Return]: ' % note)
@@ -742,7 +745,7 @@ class InfoExtractor(object):
             'mature': 17,
             'restricted': 19,
         }
-        return RATING_TABLE.get(rating.lower(), None)
+        return RATING_TABLE.get(rating.lower())
 
     def _family_friendly_search(self, html):
         # See http://schema.org/VideoObject
@@ -757,7 +760,7 @@ class InfoExtractor(object):
             '0': 18,
             'false': 18,
         }
-        return RATING_TABLE.get(family_friendly.lower(), None)
+        return RATING_TABLE.get(family_friendly.lower())
 
     def _twitter_search_player(self, html):
         return self._html_search_meta('twitter:player', html,
@@ -825,6 +828,12 @@ class InfoExtractor(object):
         if not formats:
             raise ExtractorError('No video formats found')
 
+        for f in formats:
+            # Automatically determine tbr when missing based on abr and vbr (improves
+            # formats sorting in some cases)
+            if 'tbr' not in f and f.get('abr') is not None and f.get('vbr') is not None:
+                f['tbr'] = f['abr'] + f['vbr']
+
         def _formats_key(f):
             # TODO remove the following workaround
             from ..utils import determine_ext
@@ -891,6 +900,16 @@ class InfoExtractor(object):
                     item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'),
                 formats)
 
+    @staticmethod
+    def _remove_duplicate_formats(formats):
+        format_urls = set()
+        unique_formats = []
+        for f in formats:
+            if f['url'] not in format_urls:
+                format_urls.add(f['url'])
+                unique_formats.append(f)
+        formats[:] = unique_formats
+
     def _is_valid_url(self, url, video_id, item='video'):
         url = self._proto_relative_url(url, scheme='http:')
         # For now assume non HTTP(S) URLs always valid
@@ -1014,6 +1033,18 @@ class InfoExtractor(object):
             return []
         m3u8_doc, urlh = res
         m3u8_url = urlh.geturl()
+        # A Media Playlist Tag MUST NOT appear in a Master Playlist
+        # https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3
+        # The EXT-X-TARGETDURATION tag is REQUIRED for every M3U8 Media Playlists
+        # https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.1
+        if '#EXT-X-TARGETDURATION' in m3u8_doc:
+            return [{
+                'url': m3u8_url,
+                'format_id': m3u8_id,
+                'ext': ext,
+                'protocol': entry_protocol,
+                'preference': preference,
+            }]
         last_info = None
         last_media = None
         kv_rex = re.compile(
@@ -1058,9 +1089,9 @@ class InfoExtractor(object):
                     # TODO: looks like video codec is not always necessarily goes first
                     va_codecs = codecs.split(',')
                     if va_codecs[0]:
-                        f['vcodec'] = va_codecs[0].partition('.')[0]
+                        f['vcodec'] = va_codecs[0]
                     if len(va_codecs) > 1 and va_codecs[1]:
-                        f['acodec'] = va_codecs[1].partition('.')[0]
+                        f['acodec'] = va_codecs[1]
                 resolution = last_info.get('RESOLUTION')
                 if resolution:
                     width_str, height_str = resolution.split('x')
@@ -1164,12 +1195,15 @@ class InfoExtractor(object):
         formats = []
         rtmp_count = 0
         http_count = 0
+        m3u8_count = 0
 
+        srcs = []
         videos = smil.findall(self._xpath_ns('.//video', namespace))
         for video in videos:
             src = video.get('src')
-            if not src:
+            if not src or src in srcs:
                 continue
+            srcs.append(src)
 
             bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
             filesize = int_or_none(video.get('size') or video.get('fileSize'))
@@ -1201,10 +1235,20 @@ class InfoExtractor(object):
                 continue
 
             src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src)
+            src_url = src_url.strip()
 
             if proto == 'm3u8' or src_ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False))
+                m3u8_formats = self._extract_m3u8_formats(
+                    src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False)
+                if len(m3u8_formats) == 1:
+                    m3u8_count += 1
+                    m3u8_formats[0].update({
+                        'format_id': 'hls-%d' % (m3u8_count if bitrate is None else bitrate),
+                        'tbr': bitrate,
+                        'width': width,
+                        'height': height,
+                    })
+                formats.extend(m3u8_formats)
                 continue
 
             if src_ext == 'f4m':
@@ -1237,21 +1281,14 @@ class InfoExtractor(object):
         return formats
 
     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
+        urls = []
         subtitles = {}
         for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))):
             src = textstream.get('src')
-            if not src:
+            if not src or src in urls:
                 continue
-            ext = textstream.get('ext') or determine_ext(src)
-            if not ext:
-                type_ = textstream.get('type')
-                SUBTITLES_TYPES = {
-                    'text/vtt': 'vtt',
-                    'text/srt': 'srt',
-                    'application/smptett+xml': 'tt',
-                }
-                if type_ in SUBTITLES_TYPES:
-                    ext = SUBTITLES_TYPES[type_]
+            urls.append(src)
+            ext = textstream.get('ext') or determine_ext(src) or mimetype2ext(textstream.get('type'))
             lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang
             subtitles.setdefault(lang, []).append({
                 'url': src,
@@ -1302,10 +1339,167 @@ class InfoExtractor(object):
             })
         return entries
 
+    def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}):
+        res = self._download_webpage_handle(
+            mpd_url, video_id,
+            note=note or 'Downloading MPD manifest',
+            errnote=errnote or 'Failed to download MPD manifest',
+            fatal=fatal)
+        if res is False:
+            return []
+        mpd, urlh = res
+        mpd_base_url = re.match(r'https?://.+/', urlh.geturl()).group()
+
+        return self._parse_mpd_formats(
+            compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url, formats_dict=formats_dict)
+
+    def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}):
+        if mpd_doc.get('type') == 'dynamic':
+            return []
+
+        namespace = self._search_regex(r'(?i)^{([^}]+)?}MPD$', mpd_doc.tag, 'namespace', default=None)
+
+        def _add_ns(path):
+            return self._xpath_ns(path, namespace)
+
+        def is_drm_protected(element):
+            return element.find(_add_ns('ContentProtection')) is not None
+
+        def extract_multisegment_info(element, ms_parent_info):
+            ms_info = ms_parent_info.copy()
+            segment_list = element.find(_add_ns('SegmentList'))
+            if segment_list is not None:
+                segment_urls_e = segment_list.findall(_add_ns('SegmentURL'))
+                if segment_urls_e:
+                    ms_info['segment_urls'] = [segment.attrib['media'] for segment in segment_urls_e]
+                initialization = segment_list.find(_add_ns('Initialization'))
+                if initialization is not None:
+                    ms_info['initialization_url'] = initialization.attrib['sourceURL']
+            else:
+                segment_template = element.find(_add_ns('SegmentTemplate'))
+                if segment_template is not None:
+                    start_number = segment_template.get('startNumber')
+                    if start_number:
+                        ms_info['start_number'] = int(start_number)
+                    segment_timeline = segment_template.find(_add_ns('SegmentTimeline'))
+                    if segment_timeline is not None:
+                        s_e = segment_timeline.findall(_add_ns('S'))
+                        if s_e:
+                            ms_info['total_number'] = 0
+                            for s in s_e:
+                                ms_info['total_number'] += 1 + int(s.get('r', '0'))
+                    else:
+                        timescale = segment_template.get('timescale')
+                        if timescale:
+                            ms_info['timescale'] = int(timescale)
+                        segment_duration = segment_template.get('duration')
+                        if segment_duration:
+                            ms_info['segment_duration'] = int(segment_duration)
+                    media_template = segment_template.get('media')
+                    if media_template:
+                        ms_info['media_template'] = media_template
+                    initialization = segment_template.get('initialization')
+                    if initialization:
+                        ms_info['initialization_url'] = initialization
+                    else:
+                        initialization = segment_template.find(_add_ns('Initialization'))
+                        if initialization is not None:
+                            ms_info['initialization_url'] = initialization.attrib['sourceURL']
+            return ms_info
+
+        mpd_duration = parse_duration(mpd_doc.get('mediaPresentationDuration'))
+        formats = []
+        for period in mpd_doc.findall(_add_ns('Period')):
+            period_duration = parse_duration(period.get('duration')) or mpd_duration
+            period_ms_info = extract_multisegment_info(period, {
+                'start_number': 1,
+                'timescale': 1,
+            })
+            for adaptation_set in period.findall(_add_ns('AdaptationSet')):
+                if is_drm_protected(adaptation_set):
+                    continue
+                adaption_set_ms_info = extract_multisegment_info(adaptation_set, period_ms_info)
+                for representation in adaptation_set.findall(_add_ns('Representation')):
+                    if is_drm_protected(representation):
+                        continue
+                    representation_attrib = adaptation_set.attrib.copy()
+                    representation_attrib.update(representation.attrib)
+                    mime_type = representation_attrib.get('mimeType')
+                    content_type = mime_type.split('/')[0] if mime_type else representation_attrib.get('contentType')
+                    if content_type == 'text':
+                        # TODO implement WebVTT downloading
+                        pass
+                    elif content_type == 'video' or content_type == 'audio':
+                        base_url = ''
+                        for element in (representation, adaptation_set, period, mpd_doc):
+                            base_url_e = element.find(_add_ns('BaseURL'))
+                            if base_url_e is not None:
+                                base_url = base_url_e.text + base_url
+                                if re.match(r'^https?://', base_url):
+                                    break
+                        if mpd_base_url and not re.match(r'^https?://', base_url):
+                            if not mpd_base_url.endswith('/') and not base_url.startswith('/'):
+                                mpd_base_url += '/'
+                            base_url = mpd_base_url + base_url
+                        representation_id = representation_attrib.get('id')
+                        lang = representation_attrib.get('lang')
+                        url_el = representation.find(_add_ns('BaseURL'))
+                        filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
+                        f = {
+                            'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id,
+                            'url': base_url,
+                            'width': int_or_none(representation_attrib.get('width')),
+                            'height': int_or_none(representation_attrib.get('height')),
+                            'tbr': int_or_none(representation_attrib.get('bandwidth'), 1000),
+                            'asr': int_or_none(representation_attrib.get('audioSamplingRate')),
+                            'fps': int_or_none(representation_attrib.get('frameRate')),
+                            'vcodec': 'none' if content_type == 'audio' else representation_attrib.get('codecs'),
+                            'acodec': 'none' if content_type == 'video' else representation_attrib.get('codecs'),
+                            'language': lang if lang not in ('mul', 'und', 'zxx', 'mis') else None,
+                            'format_note': 'DASH %s' % content_type,
+                            'filesize': filesize,
+                        }
+                        representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)
+                        if 'segment_urls' not in representation_ms_info and 'media_template' in representation_ms_info:
+                            if 'total_number' not in representation_ms_info and 'segment_duration':
+                                segment_duration = float(representation_ms_info['segment_duration']) / float(representation_ms_info['timescale'])
+                                representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration))
+                            media_template = representation_ms_info['media_template']
+                            media_template = media_template.replace('$RepresentationID$', representation_id)
+                            media_template = re.sub(r'\$(Number|Bandwidth)(?:%(0\d+)d)?\$', r'%(\1)\2d', media_template)
+                            media_template.replace('$$', '$')
+                            representation_ms_info['segment_urls'] = [media_template % {'Number': segment_number, 'Bandwidth': representation_attrib.get('bandwidth')} for segment_number in range(representation_ms_info['start_number'], representation_ms_info['total_number'] + representation_ms_info['start_number'])]
+                        if 'segment_urls' in representation_ms_info:
+                            f.update({
+                                'segment_urls': representation_ms_info['segment_urls'],
+                                'protocol': 'http_dash_segments',
+                            })
+                            if 'initialization_url' in representation_ms_info:
+                                initialization_url = representation_ms_info['initialization_url'].replace('$RepresentationID$', representation_id)
+                                f.update({
+                                    'initialization_url': initialization_url,
+                                })
+                                if not f.get('url'):
+                                    f['url'] = initialization_url
+                        try:
+                            existing_format = next(
+                                fo for fo in formats
+                                if fo['format_id'] == representation_id)
+                        except StopIteration:
+                            full_info = formats_dict.get(representation_id, {}).copy()
+                            full_info.update(f)
+                            formats.append(full_info)
+                        else:
+                            existing_format.update(f)
+                    else:
+                        self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
+        self._sort_formats(formats)
+        return formats
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()
-        now_str = now.strftime("%Y-%m-%d %H:%M")
+        now_str = now.strftime('%Y-%m-%d %H:%M')
         return name + ' ' + now_str
 
     def _int(self, v, name, fatal=False, **kwargs):
@@ -1378,7 +1572,7 @@ class InfoExtractor(object):
         return {}
 
     def _get_subtitles(self, *args, **kwargs):
-        raise NotImplementedError("This method must be implemented by subclasses")
+        raise NotImplementedError('This method must be implemented by subclasses')
 
     @staticmethod
     def _merge_subtitle_items(subtitle_list1, subtitle_list2):
@@ -1404,7 +1598,7 @@ class InfoExtractor(object):
         return {}
 
     def _get_automatic_captions(self, *args, **kwargs):
-        raise NotImplementedError("This method must be implemented by subclasses")
+        raise NotImplementedError('This method must be implemented by subclasses')
 
 
 class SearchInfoExtractor(InfoExtractor):
@@ -1444,7 +1638,7 @@ class SearchInfoExtractor(InfoExtractor):
 
     def _get_n_results(self, query, n):
         """Get a specified number of results for a query"""
-        raise NotImplementedError("This method must be implemented by subclasses")
+        raise NotImplementedError('This method must be implemented by subclasses')
 
     @property
     def SEARCH_KEY(self):

youtube_dl/extractor/crackle.py

@@ -0,0 +1,95 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import int_or_none
+
+
+class CrackleIE(InfoExtractor):
+    _VALID_URL = r'(?:crackle:|https?://(?:www\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.crackle.com/the-art-of-more/2496419',
+        'info_dict': {
+            'id': '2496419',
+            'ext': 'mp4',
+            'title': 'Heavy Lies the Head',
+            'description': 'md5:bb56aa0708fe7b9a4861535f15c3abca',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
+    }
+
+    # extracted from http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx
+    _SUBTITLE_SERVER = 'http://web-us-az.crackle.com'
+    _UPLYNK_OWNER_ID = 'e8773f7770a44dbd886eee4fca16a66b'
+    _THUMBNAIL_TEMPLATE = 'http://images-us-am.crackle.com/%stnl_1920x1080.jpg?ts=20140107233116?c=635333335057637614'
+
+    # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
+    _MEDIA_FILE_SLOTS = {
+        'c544.flv': {
+            'width': 544,
+            'height': 306,
+        },
+        '360p.mp4': {
+            'width': 640,
+            'height': 360,
+        },
+        '480p.mp4': {
+            'width': 852,
+            'height': 478,
+        },
+        '480p_1mbps.mp4': {
+            'width': 852,
+            'height': 478,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        item = self._download_xml(
+            'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id,
+            video_id).find('i')
+        title = item.attrib['t']
+
+        thumbnail = None
+        subtitles = {}
+        formats = self._extract_m3u8_formats(
+            'http://content.uplynk.com/ext/%s/%s.m3u8' % (self._UPLYNK_OWNER_ID, video_id),
+            video_id, 'mp4', m3u8_id='hls', fatal=None)
+        path = item.attrib.get('p')
+        if path:
+            thumbnail = self._THUMBNAIL_TEMPLATE % path
+            http_base_url = 'http://ahttp.crackle.com/' + path
+            for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items():
+                formats.append({
+                    'url': http_base_url + mfs_path,
+                    'format_id': 'http-' + mfs_path.split('.')[0],
+                    'width': mfs_info['width'],
+                    'height': mfs_info['height'],
+                })
+            for cc in item.findall('cc'):
+                locale = cc.attrib.get('l')
+                v = cc.attrib.get('v')
+                if locale and v:
+                    if locale not in subtitles:
+                        subtitles[locale] = []
+                    subtitles[locale] = [{
+                        'url': '%s/%s%s_%s.xml' % (self._SUBTITLE_SERVER, path, locale, v),
+                        'ext': 'ttml',
+                    }]
+        self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': item.attrib.get('d'),
+            'duration': int(item.attrib.get('r'), 16) if item.attrib.get('r') else None,
+            'series': item.attrib.get('sn'),
+            'season_number': int_or_none(item.attrib.get('se')),
+            'episode_number': int_or_none(item.attrib.get('ep')),
+            'thumbnail': thumbnail,
+            'subtitles': subtitles,
+            'formats': formats,
+        }

youtube_dl/extractor/crunchyroll.py

@@ -180,40 +180,40 @@ class CrunchyrollIE(CrunchyrollBaseIE):
             return assvalue
 
         output = '[Script Info]\n'
-        output += 'Title: %s\n' % sub_root.attrib["title"]
+        output += 'Title: %s\n' % sub_root.attrib['title']
         output += 'ScriptType: v4.00+\n'
-        output += 'WrapStyle: %s\n' % sub_root.attrib["wrap_style"]
-        output += 'PlayResX: %s\n' % sub_root.attrib["play_res_x"]
-        output += 'PlayResY: %s\n' % sub_root.attrib["play_res_y"]
+        output += 'WrapStyle: %s\n' % sub_root.attrib['wrap_style']
+        output += 'PlayResX: %s\n' % sub_root.attrib['play_res_x']
+        output += 'PlayResY: %s\n' % sub_root.attrib['play_res_y']
         output += """ScaledBorderAndShadow: yes
 
 [V4+ Styles]
 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
 """
         for style in sub_root.findall('./styles/style'):
-            output += 'Style: ' + style.attrib["name"]
-            output += ',' + style.attrib["font_name"]
-            output += ',' + style.attrib["font_size"]
-            output += ',' + style.attrib["primary_colour"]
-            output += ',' + style.attrib["secondary_colour"]
-            output += ',' + style.attrib["outline_colour"]
-            output += ',' + style.attrib["back_colour"]
-            output += ',' + ass_bool(style.attrib["bold"])
-            output += ',' + ass_bool(style.attrib["italic"])
-            output += ',' + ass_bool(style.attrib["underline"])
-            output += ',' + ass_bool(style.attrib["strikeout"])
-            output += ',' + style.attrib["scale_x"]
-            output += ',' + style.attrib["scale_y"]
-            output += ',' + style.attrib["spacing"]
-            output += ',' + style.attrib["angle"]
-            output += ',' + style.attrib["border_style"]
-            output += ',' + style.attrib["outline"]
-            output += ',' + style.attrib["shadow"]
-            output += ',' + style.attrib["alignment"]
-            output += ',' + style.attrib["margin_l"]
-            output += ',' + style.attrib["margin_r"]
-            output += ',' + style.attrib["margin_v"]
-            output += ',' + style.attrib["encoding"]
+            output += 'Style: ' + style.attrib['name']
+            output += ',' + style.attrib['font_name']
+            output += ',' + style.attrib['font_size']
+            output += ',' + style.attrib['primary_colour']
+            output += ',' + style.attrib['secondary_colour']
+            output += ',' + style.attrib['outline_colour']
+            output += ',' + style.attrib['back_colour']
+            output += ',' + ass_bool(style.attrib['bold'])
+            output += ',' + ass_bool(style.attrib['italic'])
+            output += ',' + ass_bool(style.attrib['underline'])
+            output += ',' + ass_bool(style.attrib['strikeout'])
+            output += ',' + style.attrib['scale_x']
+            output += ',' + style.attrib['scale_y']
+            output += ',' + style.attrib['spacing']
+            output += ',' + style.attrib['angle']
+            output += ',' + style.attrib['border_style']
+            output += ',' + style.attrib['outline']
+            output += ',' + style.attrib['shadow']
+            output += ',' + style.attrib['alignment']
+            output += ',' + style.attrib['margin_l']
+            output += ',' + style.attrib['margin_r']
+            output += ',' + style.attrib['margin_v']
+            output += ',' + style.attrib['encoding']
             output += '\n'
 
         output += """
@@ -222,15 +222,15 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 """
         for event in sub_root.findall('./events/event'):
             output += 'Dialogue: 0'
-            output += ',' + event.attrib["start"]
-            output += ',' + event.attrib["end"]
-            output += ',' + event.attrib["style"]
-            output += ',' + event.attrib["name"]
-            output += ',' + event.attrib["margin_l"]
-            output += ',' + event.attrib["margin_r"]
-            output += ',' + event.attrib["margin_v"]
-            output += ',' + event.attrib["effect"]
-            output += ',' + event.attrib["text"]
+            output += ',' + event.attrib['start']
+            output += ',' + event.attrib['end']
+            output += ',' + event.attrib['style']
+            output += ',' + event.attrib['name']
+            output += ',' + event.attrib['margin_l']
+            output += ',' + event.attrib['margin_r']
+            output += ',' + event.attrib['margin_v']
+            output += ',' + event.attrib['effect']
+            output += ',' + event.attrib['text']
             output += '\n'
 
         return output
@@ -376,7 +376,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
 
 class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
-    IE_NAME = "crunchyroll:playlist"
+    IE_NAME = 'crunchyroll:playlist'
     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?(?:\?|$)'
 
     _TESTS = [{

youtube_dl/extractor/cspan.py

@@ -68,11 +68,16 @@ class CSpanIE(InfoExtractor):
             video_type, video_id = matches.groups()
             video_type = 'clip' if video_type == 'id' else 'program'
         else:
-            senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
-            if senate_isvp_url:
-                title = self._og_search_title(webpage)
-                surl = smuggle_url(senate_isvp_url, {'force_title': title})
-                return self.url_result(surl, 'SenateISVP', video_id, title)
+            m = re.search(r'data-(?P<type>clip|prog)id=["\'](?P<id>\d+)', webpage)
+            if m:
+                video_id = m.group('id')
+                video_type = 'program' if m.group('type') == 'prog' else 'clip'
+            else:
+                senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
+                if senate_isvp_url:
+                    title = self._og_search_title(webpage)
+                    surl = smuggle_url(senate_isvp_url, {'force_title': title})
+                    return self.url_result(surl, 'SenateISVP', video_id, title)
         if video_type is None or video_id is None:
             raise ExtractorError('unable to find video id and type')
 
@@ -107,6 +112,13 @@ class CSpanIE(InfoExtractor):
                     'height': int_or_none(get_text_attr(quality, 'height')),
                     'tbr': int_or_none(get_text_attr(quality, 'bitrate')),
                 })
+            if not formats:
+                path = unescapeHTML(get_text_attr(f, 'path'))
+                if not path:
+                    continue
+                formats = self._extract_m3u8_formats(
+                    path, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls') if determine_ext(path) == 'm3u8' else [{'url': path, }]
             self._sort_formats(formats)
             entries.append({
                 'id': '%s_%d' % (video_id, partnum + 1),

youtube_dl/extractor/dailymotion.py

@@ -122,10 +122,13 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         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))
+        view_count_str = self._search_regex(
+            (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
+             r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
+            webpage, 'view count', fatal=False)
+        if view_count_str:
+            view_count_str = re.sub(r'\s', '', view_count_str)
+        view_count = str_to_int(view_count_str)
         comment_count = int_or_none(self._search_regex(
             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
             webpage, 'comment count', fatal=False))
@@ -396,13 +399,13 @@ class DailymotionCloudIE(DailymotionBaseInfoExtractor):
     }]
 
     @classmethod
-    def _extract_dmcloud_url(self, webpage):
-        mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage)
+    def _extract_dmcloud_url(cls, webpage):
+        mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL, webpage)
         if mobj:
             return mobj.group(1)
 
         mobj = re.search(
-            r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL,
+            r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL,
             webpage)
         if mobj:
             return mobj.group(1)

youtube_dl/extractor/daum.py

@@ -2,17 +2,26 @@
 
 from __future__ import unicode_literals
 
+import re
+import itertools
+
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse
+from ..compat import (
+    compat_parse_qs,
+    compat_urllib_parse,
+    compat_urllib_parse_unquote,
+    compat_urlparse,
+)
 from ..utils import (
     int_or_none,
     str_to_int,
     xpath_text,
+    unescapeHTML,
 )
 
 
 class DaumIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/v/(?P<id>[^?#&]+)'
+    _VALID_URL = r'https?://(?:(?:m\.)?tvpot\.daum\.net/v/|videofarm\.daum\.net/controller/player/VodPlayer\.swf\?vid=)(?P<id>[^?#&]+)'
     IE_NAME = 'daum.net'
 
     _TESTS = [{
@@ -23,25 +32,57 @@ class DaumIE(InfoExtractor):
             'title': '마크 헌트 vs 안토니오 실바',
             'description': 'Mark Hunt vs Antonio Silva',
             'upload_date': '20131217',
+            'thumbnail': 're:^https?://.*\.(?:jpg|png)',
             'duration': 2117,
             'view_count': int,
             'comment_count': int,
         },
+    }, {
+        'url': 'http://m.tvpot.daum.net/v/65139429',
+        'info_dict': {
+            'id': '65139429',
+            'ext': 'mp4',
+            'title': '1297회, \'아빠 아들로 태어나길 잘 했어\' 민수, 감동의 눈물[아빠 어디가] 20150118',
+            'description': 'md5:79794514261164ff27e36a21ad229fc5',
+            'upload_date': '20150604',
+            'thumbnail': 're:^https?://.*\.(?:jpg|png)',
+            'duration': 154,
+            'view_count': int,
+            'comment_count': int,
+        },
     }, {
         'url': 'http://tvpot.daum.net/v/07dXWRka62Y%24',
         'only_matching': True,
+    }, {
+        'url': 'http://videofarm.daum.net/controller/player/VodPlayer.swf?vid=vwIpVpCQsT8%24&ref=',
+        'info_dict': {
+            'id': 'vwIpVpCQsT8$',
+            'ext': 'flv',
+            'title': '01-Korean War ( Trouble on the horizon )',
+            'description': '\nKorean War 01\nTrouble on the horizon\n전쟁의 먹구름',
+            'upload_date': '20080223',
+            'thumbnail': 're:^https?://.*\.(?:jpg|png)',
+            'duration': 249,
+            'view_count': int,
+            'comment_count': int,
+        },
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        video_id = compat_urllib_parse_unquote(self._match_id(url))
         query = compat_urllib_parse.urlencode({'vid': video_id})
-        info = self._download_xml(
-            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
-            'Downloading video info')
         movie_data = self._download_json(
             'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json?' + query,
             video_id, 'Downloading video formats info')
 
+        # For urls like http://m.tvpot.daum.net/v/65139429, where the video_id is really a clipid
+        if not movie_data.get('output_list', {}).get('output_list') and re.match(r'^\d+$', video_id):
+            return self.url_result('http://tvpot.daum.net/clip/ClipView.do?clipid=%s' % video_id)
+
+        info = self._download_xml(
+            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
+            'Downloading video info')
+
         formats = []
         for format_el in movie_data['output_list']['output_list']:
             profile = format_el['profile']
@@ -76,8 +117,9 @@ class DaumIE(InfoExtractor):
 
 
 class DaumClipIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/(?:clip/ClipView.do|mypot/View.do)\?.*?clipid=(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/(?:clip/ClipView.(?:do|tv)|mypot/View.do)\?.*?clipid=(?P<id>\d+)'
     IE_NAME = 'daum.net:clip'
+    _URL_TEMPLATE = 'http://tvpot.daum.net/clip/ClipView.do?clipid=%s'
 
     _TESTS = [{
         'url': 'http://tvpot.daum.net/clip/ClipView.do?clipid=52554690',
@@ -87,11 +129,19 @@ class DaumClipIE(InfoExtractor):
             'title': 'DOTA 2GETHER 시즌2 6회 - 2부',
             'description': 'DOTA 2GETHER 시즌2 6회 - 2부',
             'upload_date': '20130831',
+            'thumbnail': 're:^https?://.*\.(?:jpg|png)',
             'duration': 3868,
             'view_count': int,
         },
+    }, {
+        'url': 'http://m.tvpot.daum.net/clip/ClipView.tv?clipid=54999425',
+        'only_matching': True,
     }]
 
+    @classmethod
+    def suitable(cls, url):
+        return False if DaumPlaylistIE.suitable(url) or DaumUserIE.suitable(url) else super(DaumClipIE, cls).suitable(url)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
         clip_info = self._download_json(
@@ -102,7 +152,7 @@ class DaumClipIE(InfoExtractor):
             '_type': 'url_transparent',
             'id': video_id,
             'url': 'http://tvpot.daum.net/v/%s' % clip_info['vid'],
-            'title': clip_info['title'],
+            'title': unescapeHTML(clip_info['title']),
             'thumbnail': clip_info.get('thumb_url'),
             'description': clip_info.get('contents'),
             'duration': int_or_none(clip_info.get('duration')),
@@ -110,3 +160,139 @@ class DaumClipIE(InfoExtractor):
             'view_count': int_or_none(clip_info.get('play_count')),
             'ie_key': 'Daum',
         }
+
+
+class DaumListIE(InfoExtractor):
+    def _get_entries(self, list_id, list_id_type):
+        name = None
+        entries = []
+        for pagenum in itertools.count(1):
+            list_info = self._download_json(
+                'http://tvpot.daum.net/mypot/json/GetClipInfo.do?size=48&init=true&order=date&page=%d&%s=%s' % (
+                    pagenum, list_id_type, list_id), list_id, 'Downloading list info - %s' % pagenum)
+
+            entries.extend([
+                self.url_result(
+                    'http://tvpot.daum.net/v/%s' % clip['vid'])
+                for clip in list_info['clip_list']
+            ])
+
+            if not name:
+                name = list_info.get('playlist_bean', {}).get('name') or \
+                    list_info.get('potInfo', {}).get('name')
+
+            if not list_info.get('has_more'):
+                break
+
+        return name, entries
+
+    def _check_clip(self, url, list_id):
+        query_dict = compat_parse_qs(compat_urlparse.urlparse(url).query)
+        if 'clipid' in query_dict:
+            clip_id = query_dict['clipid'][0]
+            if self._downloader.params.get('noplaylist'):
+                self.to_screen('Downloading just video %s because of --no-playlist' % clip_id)
+                return self.url_result(DaumClipIE._URL_TEMPLATE % clip_id, 'DaumClip')
+            else:
+                self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % list_id)
+
+
+class DaumPlaylistIE(DaumListIE):
+    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/mypot/(?:View\.do|Top\.tv)\?.*?playlistid=(?P<id>[0-9]+)'
+    IE_NAME = 'daum.net:playlist'
+    _URL_TEMPLATE = 'http://tvpot.daum.net/mypot/View.do?playlistid=%s'
+
+    _TESTS = [{
+        'note': 'Playlist url with clipid',
+        'url': 'http://tvpot.daum.net/mypot/View.do?playlistid=6213966&clipid=73806844',
+        'info_dict': {
+            'id': '6213966',
+            'title': 'Woorissica Official',
+        },
+        'playlist_mincount': 181
+    }, {
+        'note': 'Playlist url with clipid - noplaylist',
+        'url': 'http://tvpot.daum.net/mypot/View.do?playlistid=6213966&clipid=73806844',
+        'info_dict': {
+            'id': '73806844',
+            'ext': 'mp4',
+            'title': '151017 Airport',
+            'upload_date': '20160117',
+        },
+        'params': {
+            'noplaylist': True,
+            'skip_download': True,
+        }
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if DaumUserIE.suitable(url) else super(DaumPlaylistIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+
+        clip_result = self._check_clip(url, list_id)
+        if clip_result:
+            return clip_result
+
+        name, entries = self._get_entries(list_id, 'playlistid')
+
+        return self.playlist_result(entries, list_id, name)
+
+
+class DaumUserIE(DaumListIE):
+    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/mypot/(?:View|Top)\.(?:do|tv)\?.*?ownerid=(?P<id>[0-9a-zA-Z]+)'
+    IE_NAME = 'daum.net:user'
+
+    _TESTS = [{
+        'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0',
+        'info_dict': {
+            'id': 'o2scDLIVbHc0',
+            'title': '마이 리틀 텔레비전',
+        },
+        'playlist_mincount': 213
+    }, {
+        'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0&clipid=73801156',
+        'info_dict': {
+            'id': '73801156',
+            'ext': 'mp4',
+            'title': '[미공개] 김구라, 오만석이 부릅니다 \'오케피\' - 마이 리틀 텔레비전 20160116',
+            'upload_date': '20160117',
+            'description': 'md5:5e91d2d6747f53575badd24bd62b9f36'
+        },
+        'params': {
+            'noplaylist': True,
+            'skip_download': True,
+        }
+    }, {
+        'note': 'Playlist url has ownerid and playlistid, playlistid takes precedence',
+        'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0&playlistid=6196631',
+        'info_dict': {
+            'id': '6196631',
+            'title': '마이 리틀 텔레비전 - 20160109',
+        },
+        'playlist_count': 11
+    }, {
+        'url': 'http://tvpot.daum.net/mypot/Top.do?ownerid=o2scDLIVbHc0',
+        'only_matching': True,
+    }, {
+        'url': 'http://m.tvpot.daum.net/mypot/Top.tv?ownerid=45x1okb1If50&playlistid=3569733',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+
+        clip_result = self._check_clip(url, list_id)
+        if clip_result:
+            return clip_result
+
+        query_dict = compat_parse_qs(compat_urlparse.urlparse(url).query)
+        if 'playlistid' in query_dict:
+            playlist_id = query_dict['playlistid'][0]
+            return self.url_result(DaumPlaylistIE._URL_TEMPLATE % playlist_id, 'DaumPlaylist')
+
+        name, entries = self._get_entries(list_id, 'ownerid')
+
+        return self.playlist_result(entries, list_id, name)

youtube_dl/extractor/drbonanza.py

@@ -87,7 +87,7 @@ class DRBonanzaIE(InfoExtractor):
 
         formats = []
         for file in info['Files']:
-            if info['Type'] == "Video":
+            if info['Type'] == 'Video':
                 if file['Type'] in video_types:
                     format = parse_filename_info(file['Location'])
                     format.update({
@@ -101,10 +101,10 @@ class DRBonanzaIE(InfoExtractor):
                         if '/bonanza/' in rtmp_url:
                             format['play_path'] = rtmp_url.split('/bonanza/')[1]
                     formats.append(format)
-                elif file['Type'] == "Thumb":
+                elif file['Type'] == 'Thumb':
                     thumbnail = file['Location']
-            elif info['Type'] == "Audio":
-                if file['Type'] == "Audio":
+            elif info['Type'] == 'Audio':
+                if file['Type'] == 'Audio':
                     format = parse_filename_info(file['Location'])
                     format.update({
                         'url': file['Location'],
@@ -112,7 +112,7 @@ class DRBonanzaIE(InfoExtractor):
                         'vcodec': 'none',
                     })
                     formats.append(format)
-                elif file['Type'] == "Thumb":
+                elif file['Type'] == 'Thumb':
                     thumbnail = file['Location']
 
         description = '%s\n%s\n%s\n' % (

youtube_dl/extractor/eighttracks.py

@@ -17,85 +17,85 @@ class EightTracksIE(InfoExtractor):
     IE_NAME = '8tracks'
     _VALID_URL = r'https?://8tracks\.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$'
     _TEST = {
-        "name": "EightTracks",
-        "url": "http://8tracks.com/ytdl/youtube-dl-test-tracks-a",
-        "info_dict": {
+        'name': 'EightTracks',
+        'url': 'http://8tracks.com/ytdl/youtube-dl-test-tracks-a',
+        'info_dict': {
             'id': '1336550',
             'display_id': 'youtube-dl-test-tracks-a',
-            "description": "test chars:  \"'/\\ä↭",
-            "title": "youtube-dl test tracks \"'/\\ä↭<>",
+            'description': "test chars:  \"'/\\ä↭",
+            'title': "youtube-dl test tracks \"'/\\ä↭<>",
         },
-        "playlist": [
+        'playlist': [
             {
-                "md5": "96ce57f24389fc8734ce47f4c1abcc55",
-                "info_dict": {
-                    "id": "11885610",
-                    "ext": "m4a",
-                    "title": "youtue-dl project<>\"' - youtube-dl test track 1 \"'/\\\u00e4\u21ad",
-                    "uploader_id": "ytdl"
+                'md5': '96ce57f24389fc8734ce47f4c1abcc55',
+                'info_dict': {
+                    'id': '11885610',
+                    'ext': 'm4a',
+                    'title': "youtue-dl project<>\"' - youtube-dl test track 1 \"'/\\\u00e4\u21ad",
+                    'uploader_id': 'ytdl'
                 }
             },
             {
-                "md5": "4ab26f05c1f7291ea460a3920be8021f",
-                "info_dict": {
-                    "id": "11885608",
-                    "ext": "m4a",
-                    "title": "youtube-dl project - youtube-dl test track 2 \"'/\\\u00e4\u21ad",
-                    "uploader_id": "ytdl"
+                'md5': '4ab26f05c1f7291ea460a3920be8021f',
+                'info_dict': {
+                    'id': '11885608',
+                    'ext': 'm4a',
+                    'title': "youtube-dl project - youtube-dl test track 2 \"'/\\\u00e4\u21ad",
+                    'uploader_id': 'ytdl'
                 }
             },
             {
-                "md5": "d30b5b5f74217410f4689605c35d1fd7",
-                "info_dict": {
-                    "id": "11885679",
-                    "ext": "m4a",
-                    "title": "youtube-dl project as well - youtube-dl test track 3 \"'/\\\u00e4\u21ad",
-                    "uploader_id": "ytdl"
+                'md5': 'd30b5b5f74217410f4689605c35d1fd7',
+                'info_dict': {
+                    'id': '11885679',
+                    'ext': 'm4a',
+                    'title': "youtube-dl project as well - youtube-dl test track 3 \"'/\\\u00e4\u21ad",
+                    'uploader_id': 'ytdl'
                 }
             },
             {
-                "md5": "4eb0a669317cd725f6bbd336a29f923a",
-                "info_dict": {
-                    "id": "11885680",
-                    "ext": "m4a",
-                    "title": "youtube-dl project as well - youtube-dl test track 4 \"'/\\\u00e4\u21ad",
-                    "uploader_id": "ytdl"
+                'md5': '4eb0a669317cd725f6bbd336a29f923a',
+                'info_dict': {
+                    'id': '11885680',
+                    'ext': 'm4a',
+                    'title': "youtube-dl project as well - youtube-dl test track 4 \"'/\\\u00e4\u21ad",
+                    'uploader_id': 'ytdl'
                 }
             },
             {
-                "md5": "1893e872e263a2705558d1d319ad19e8",
-                "info_dict": {
-                    "id": "11885682",
-                    "ext": "m4a",
-                    "title": "PH - youtube-dl test track 5 \"'/\\\u00e4\u21ad",
-                    "uploader_id": "ytdl"
+                'md5': '1893e872e263a2705558d1d319ad19e8',
+                'info_dict': {
+                    'id': '11885682',
+                    'ext': 'm4a',
+                    'title': "PH - youtube-dl test track 5 \"'/\\\u00e4\u21ad",
+                    'uploader_id': 'ytdl'
                 }
             },
             {
-                "md5": "b673c46f47a216ab1741ae8836af5899",
-                "info_dict": {
-                    "id": "11885683",
-                    "ext": "m4a",
-                    "title": "PH - youtube-dl test track 6 \"'/\\\u00e4\u21ad",
-                    "uploader_id": "ytdl"
+                'md5': 'b673c46f47a216ab1741ae8836af5899',
+                'info_dict': {
+                    'id': '11885683',
+                    'ext': 'm4a',
+                    'title': "PH - youtube-dl test track 6 \"'/\\\u00e4\u21ad",
+                    'uploader_id': 'ytdl'
                 }
             },
             {
-                "md5": "1d74534e95df54986da7f5abf7d842b7",
-                "info_dict": {
-                    "id": "11885684",
-                    "ext": "m4a",
-                    "title": "phihag - youtube-dl test track 7 \"'/\\\u00e4\u21ad",
-                    "uploader_id": "ytdl"
+                'md5': '1d74534e95df54986da7f5abf7d842b7',
+                'info_dict': {
+                    'id': '11885684',
+                    'ext': 'm4a',
+                    'title': "phihag - youtube-dl test track 7 \"'/\\\u00e4\u21ad",
+                    'uploader_id': 'ytdl'
                 }
             },
             {
-                "md5": "f081f47af8f6ae782ed131d38b9cd1c0",
-                "info_dict": {
-                    "id": "11885685",
-                    "ext": "m4a",
-                    "title": "phihag - youtube-dl test track 8 \"'/\\\u00e4\u21ad",
-                    "uploader_id": "ytdl"
+                'md5': 'f081f47af8f6ae782ed131d38b9cd1c0',
+                'info_dict': {
+                    'id': '11885685',
+                    'ext': 'm4a',
+                    'title': "phihag - youtube-dl test track 8 \"'/\\\u00e4\u21ad",
+                    'uploader_id': 'ytdl'
                 }
             }
         ]

youtube_dl/extractor/ellentv.py

@@ -72,7 +72,7 @@ class EllenTVClipsIE(InfoExtractor):
     def _extract_playlist(self, webpage):
         json_string = self._search_regex(r'playerView.addClips\(\[\{(.*?)\}\]\);', webpage, 'json')
         try:
-            return json.loads("[{" + json_string + "}]")
+            return json.loads('[{' + json_string + '}]')
         except ValueError as ve:
             raise ExtractorError('Failed to download JSON', cause=ve)
 

youtube_dl/extractor/espn.py

@@ -53,8 +53,8 @@ class ESPNIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         video_id = self._search_regex(
-            r'class="video-play-button"[^>]+data-id="(\d+)',
-            webpage, 'video id')
+            r'class=(["\']).*?video-play-button.*?\1[^>]+data-id=["\'](?P<id>\d+)',
+            webpage, 'video id', group='id')
 
         cms = 'espn'
         if 'data-source="intl"' in webpage:

youtube_dl/extractor/everyonesmixtape.py

@@ -14,14 +14,14 @@ class EveryonesMixtapeIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://everyonesmixtape.com/#/mix/m7m0jJAbMQi/5',
-        "info_dict": {
+        'info_dict': {
             'id': '5bfseWNmlds',
             'ext': 'mp4',
-            "title": "Passion Pit - \"Sleepyhead\" (Official Music Video)",
-            "uploader": "FKR.TV",
-            "uploader_id": "frenchkissrecords",
-            "description": "Music video for \"Sleepyhead\" from Passion Pit's debut EP Chunk Of Change.\nBuy on iTunes: https://itunes.apple.com/us/album/chunk-of-change-ep/id300087641\n\nDirected by The Wilderness.\n\nhttp://www.passionpitmusic.com\nhttp://www.frenchkissrecords.com",
-            "upload_date": "20081015"
+            'title': "Passion Pit - \"Sleepyhead\" (Official Music Video)",
+            'uploader': 'FKR.TV',
+            'uploader_id': 'frenchkissrecords',
+            'description': "Music video for \"Sleepyhead\" from Passion Pit's debut EP Chunk Of Change.\nBuy on iTunes: https://itunes.apple.com/us/album/chunk-of-change-ep/id300087641\n\nDirected by The Wilderness.\n\nhttp://www.passionpitmusic.com\nhttp://www.frenchkissrecords.com",
+            'upload_date': '20081015'
         },
         'params': {
             'skip_download': True,  # This is simply YouTube

youtube_dl/extractor/exfm.py

@@ -41,7 +41,7 @@ class ExfmIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         song_id = mobj.group('id')
-        info_url = "http://ex.fm/api/v3/song/%s" % song_id
+        info_url = 'http://ex.fm/api/v3/song/%s' % song_id
         info = self._download_json(info_url, song_id)['song']
         song_url = info['url']
         if re.match(self._SOUNDCLOUD_URL, song_url) is not None:

youtube_dl/extractor/facebook.py

@@ -6,9 +6,11 @@ import socket
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_etree_fromstring,
     compat_http_client,
     compat_urllib_error,
     compat_urllib_parse_unquote,
+    compat_urllib_parse_unquote_plus,
 )
 from ..utils import (
     error_to_compat_str,
@@ -23,19 +25,30 @@ from ..utils import (
 
 class FacebookIE(InfoExtractor):
     _VALID_URL = r'''(?x)
-        https?://(?:\w+\.)?facebook\.com/
-        (?:[^#]*?\#!/)?
-        (?:
-            (?:video/video\.php|photo\.php|video\.php|video/embed)\?(?:.*?)
-            (?:v|video_id)=|
-            [^/]+/videos/(?:[^/]+/)?
-        )
-        (?P<id>[0-9]+)
-        (?:.*)'''
+                (?:
+                    https?://
+                        (?:\w+\.)?facebook\.com/
+                        (?:[^#]*?\#!/)?
+                        (?:
+                            (?:
+                                video/video\.php|
+                                photo\.php|
+                                video\.php|
+                                video/embed
+                            )\?(?:.*?)(?:v|video_id)=|
+                            [^/]+/videos/(?:[^/]+/)?
+                        )|
+                    facebook:
+                )
+                (?P<id>[0-9]+)
+                '''
     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
     _NETRC_MACHINE = 'facebook'
     IE_NAME = 'facebook'
+
+    _CHROME_USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.97 Safari/537.36'
+
     _TESTS = [{
         'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
         'md5': '6a40d33c0eccbb1af76cf0485a052659',
@@ -57,6 +70,16 @@ class FacebookIE(InfoExtractor):
         'expected_warnings': [
             'title'
         ]
+    }, {
+        'note': 'Video with DASH manifest',
+        'url': 'https://www.facebook.com/video.php?v=957955867617029',
+        'md5': '54706e4db4f5ad58fbad82dde1f1213f',
+        'info_dict': {
+            'id': '957955867617029',
+            'ext': 'mp4',
+            'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...',
+            'uploader': 'Demy de Zeeuw',
+        },
     }, {
         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
         'only_matching': True,
@@ -66,6 +89,9 @@ class FacebookIE(InfoExtractor):
     }, {
         'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
         'only_matching': True,
+    }, {
+        'url': 'facebook:544765982287235',
+        'only_matching': True,
     }]
 
     def _login(self):
@@ -136,13 +162,36 @@ class FacebookIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
-        webpage = self._download_webpage(url, video_id)
+        req = sanitized_Request('https://www.facebook.com/video/video.php?v=%s' % video_id)
+        req.add_header('User-Agent', self._CHROME_USER_AGENT)
+        webpage = self._download_webpage(req, video_id)
+
+        video_data = None
 
         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
-        if not m:
+        if m:
+            data = dict(json.loads(m.group(1)))
+            params_raw = compat_urllib_parse_unquote(data['params'])
+            video_data = json.loads(params_raw)['video_data']
+
+        def video_data_list2dict(video_data):
+            ret = {}
+            for item in video_data:
+                format_id = item['stream_type']
+                ret.setdefault(format_id, []).append(item)
+            return ret
+
+        if not video_data:
+            server_js_data = self._parse_json(self._search_regex(
+                r'handleServerJS\(({.+})\);', webpage, 'server js data'), video_id)
+            for item in server_js_data.get('instances', []):
+                if item[1][0] == 'VideoConfig':
+                    video_data = video_data_list2dict(item[2][0]['videoData'])
+                    break
+
+        if not video_data:
             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
             if m_msg is not None:
                 raise ExtractorError(
@@ -150,12 +199,9 @@ class FacebookIE(InfoExtractor):
                     expected=True)
             else:
                 raise ExtractorError('Cannot parse data')
-        data = dict(json.loads(m.group(1)))
-        params_raw = compat_urllib_parse_unquote(data['params'])
-        params = json.loads(params_raw)
 
         formats = []
-        for format_id, f in params['video_data'].items():
+        for format_id, f in video_data.items():
             if not f or not isinstance(f, list):
                 continue
             for quality in ('sd', 'hd'):
@@ -167,9 +213,15 @@ class FacebookIE(InfoExtractor):
                             'url': src,
                             'preference': -10 if format_id == 'progressive' else 0,
                         })
+            dash_manifest = f[0].get('dash_manifest')
+            if dash_manifest:
+                formats.extend(self._parse_mpd_formats(
+                    compat_etree_fromstring(compat_urllib_parse_unquote_plus(dash_manifest))))
         if not formats:
             raise ExtractorError('Cannot find video formats')
 
+        self._sort_formats(formats)
+
         video_title = self._html_search_regex(
             r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title',
             default=None)
@@ -188,3 +240,33 @@ class FacebookIE(InfoExtractor):
             'formats': formats,
             'uploader': uploader,
         }
+
+
+class FacebookPostIE(InfoExtractor):
+    IE_NAME = 'facebook:post'
+    _VALID_URL = r'https?://(?:\w+\.)?facebook\.com/[^/]+/posts/(?P<id>\d+)'
+    _TEST = {
+        'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
+        'md5': '037b1fa7f3c2d02b7a0d7bc16031ecc6',
+        'info_dict': {
+            'id': '544765982287235',
+            'ext': 'mp4',
+            'title': '"What are you doing running in the snow?"',
+            'uploader': 'FailArmy',
+        }
+    }
+
+    def _real_extract(self, url):
+        post_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, post_id)
+
+        entries = [
+            self.url_result('facebook:%s' % video_id, FacebookIE.ie_key())
+            for video_id in self._parse_json(
+                self._search_regex(
+                    r'(["\'])video_ids\1\s*:\s*(?P<ids>\[.+?\])',
+                    webpage, 'video ids', group='ids'),
+                post_id)]
+
+        return self.playlist_result(entries, post_id)

youtube_dl/extractor/faz.py

@@ -52,7 +52,7 @@ class FazIE(InfoExtractor):
         formats = []
         for pref, code in enumerate(['LOW', 'HIGH', 'HQ']):
             encoding = xpath_element(encodings, code)
-            if encoding:
+            if encoding is not None:
                 encoding_url = xpath_text(encoding, 'FILENAME')
                 if encoding_url:
                     formats.append({

youtube_dl/extractor/fc2.py

@@ -87,7 +87,7 @@ class FC2IE(InfoExtractor):
         mimi = hashlib.md5((video_id + '_gGddgPfeaf_gzyr').encode('utf-8')).hexdigest()
 
         info_url = (
-            "http://video.fc2.com/ginfo.php?mimi={1:s}&href={2:s}&v={0:s}&fversion=WIN%2011%2C6%2C602%2C180&from=2&otag=0&upid={0:s}&tk=null&".
+            'http://video.fc2.com/ginfo.php?mimi={1:s}&href={2:s}&v={0:s}&fversion=WIN%2011%2C6%2C602%2C180&from=2&otag=0&upid={0:s}&tk=null&'.
             format(video_id, mimi, compat_urllib_request.quote(refer, safe=b'').replace('.', '%2E')))
 
         info_webpage = self._download_webpage(

youtube_dl/extractor/fox.py

@@ -9,6 +9,7 @@ class FOXIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?fox\.com/watch/(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.fox.com/watch/255180355939/7684182528',
+        'md5': 'ebd296fcc41dd4b19f8115d8461a3165',
         'info_dict': {
             'id': '255180355939',
             'ext': 'mp4',
@@ -17,10 +18,6 @@ class FOXIE(InfoExtractor):
             'duration': 129,
         },
         'add_ie': ['ThePlatform'],
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
     }
 
     def _real_extract(self, url):
@@ -29,7 +26,7 @@ class FOXIE(InfoExtractor):
 
         release_url = self._parse_json(self._search_regex(
             r'"fox_pdk_player"\s*:\s*({[^}]+?})', webpage, 'fox_pdk_player'),
-            video_id)['release_url'] + '&manifest=m3u'
+            video_id)['release_url'] + '&switch=http'
 
         return {
             '_type': 'url_transparent',

youtube_dl/extractor/franceinter.py

@@ -10,7 +10,7 @@ class FranceInterIE(InfoExtractor):
     _TEST = {
         'url': 'http://www.franceinter.fr/player/reecouter?play=793962',
         'md5': '4764932e466e6f6c79c317d2e74f6884',
-        "info_dict": {
+        'info_dict': {
             'id': '793962',
             'ext': 'mp3',
             'title': 'L’Histoire dans les jeux vidéo',

youtube_dl/extractor/francetv.py

@@ -289,7 +289,7 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
         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=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
             webpage, 'video ID').split('@')
         return self._extract_video(video_id, catalogue)
 

youtube_dl/extractor/freevideo.py

@@ -12,8 +12,8 @@ class FreeVideoIE(InfoExtractor):
         'info_dict': {
             'id': 'vysukany-zadecek-22033',
             'ext': 'mp4',
-            "title": "vysukany-zadecek-22033",
-            "age_limit": 18,
+            'title': 'vysukany-zadecek-22033',
+            'age_limit': 18,
         },
         'skip': 'Blocked outside .cz',
     }

youtube_dl/extractor/gamekings.py

@@ -6,24 +6,29 @@ from ..utils import (
     xpath_text,
     xpath_with_ns,
 )
+from .youtube import YoutubeIE
 
 
 class GamekingsIE(InfoExtractor):
-    _VALID_URL = r'http://www\.gamekings\.tv/(?:videos|nieuws)/(?P<id>[^/]+)'
+    _VALID_URL = r'http://www\.gamekings\.nl/(?:videos|nieuws)/(?P<id>[^/]+)'
     _TESTS = [{
-        'url': 'http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/',
-        # MD5 is flaky, seems to change regularly
-        # 'md5': '2f32b1f7b80fdc5cb616efb4f387f8a3',
+        # YouTube embed video
+        'url': 'http://www.gamekings.nl/videos/phoenix-wright-ace-attorney-dual-destinies-review/',
+        'md5': '5208d3a17adeaef829a7861887cb9029',
         'info_dict': {
-            'id': 'phoenix-wright-ace-attorney-dual-destinies-review',
+            'id': 'HkSQKetlGOU',
             'ext': 'mp4',
-            'title': 'Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review',
-            'description': 'md5:36fd701e57e8c15ac8682a2374c99731',
+            'title': 'Phoenix Wright: Ace Attorney - Dual Destinies Review',
+            'description': 'md5:db88c0e7f47e9ea50df3271b9dc72e1d',
             'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader_id': 'UCJugRGo4STYMeFr5RoOShtQ',
+            'uploader': 'Gamekings Vault',
+            'upload_date': '20151123',
         },
+        'add_ie': ['Youtube'],
     }, {
         # vimeo video
-        'url': 'http://www.gamekings.tv/videos/the-legend-of-zelda-majoras-mask/',
+        'url': 'http://www.gamekings.nl/videos/the-legend-of-zelda-majoras-mask/',
         'md5': '12bf04dfd238e70058046937657ea68d',
         'info_dict': {
             'id': 'the-legend-of-zelda-majoras-mask',
@@ -33,7 +38,7 @@ class GamekingsIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg$',
         },
     }, {
-        'url': 'http://www.gamekings.tv/nieuws/gamekings-extra-shelly-en-david-bereiden-zich-voor-op-de-livestream/',
+        'url': 'http://www.gamekings.nl/nieuws/gamekings-extra-shelly-en-david-bereiden-zich-voor-op-de-livestream/',
         'only_matching': True,
     }]
 
@@ -43,7 +48,11 @@ class GamekingsIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         playlist_id = self._search_regex(
-            r'gogoVideo\(\s*\d+\s*,\s*"([^"]+)', webpage, 'playlist id')
+            r'gogoVideo\([^,]+,\s*"([^"]+)', webpage, 'playlist id')
+
+        # Check if a YouTube embed is used
+        if YoutubeIE.suitable(playlist_id):
+            return self.url_result(playlist_id, ie='Youtube')
 
         playlist = self._download_xml(
             'http://www.gamekings.tv/wp-content/themes/gk2010/rss_playlist.php?id=%s' % playlist_id,

youtube_dl/extractor/generic.py

@@ -224,6 +224,20 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        # MPD from http://dash-mse-test.appspot.com/media.html
+        {
+            'url': 'http://yt-dash-mse-test.commondatastorage.googleapis.com/media/car-20120827-manifest.mpd',
+            'md5': '4b57baab2e30d6eb3a6a09f0ba57ef53',
+            'info_dict': {
+                'id': 'car-20120827-manifest',
+                'ext': 'mp4',
+                'title': 'car-20120827-manifest',
+                'formats': 'mincount:9',
+            },
+            'params': {
+                'format': 'bestvideo',
+            },
+        },
         # google redirect
         {
             'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
@@ -1229,19 +1243,24 @@ class GenericIE(InfoExtractor):
 
         # Check for direct link to a video
         content_type = head_response.headers.get('Content-Type', '')
-        m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
+        m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>.+)$', content_type)
         if m:
             upload_date = unified_strdate(
                 head_response.headers.get('Last-Modified'))
+            formats = []
+            if m.group('format_id').endswith('mpegurl'):
+                formats = self._extract_m3u8_formats(url, video_id, 'mp4')
+            else:
+                formats = [{
+                    'format_id': m.group('format_id'),
+                    'url': url,
+                    'vcodec': 'none' if m.group('type') == 'audio' else None
+                }]
             return {
                 'id': video_id,
                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
                 'direct': True,
-                'formats': [{
-                    'format_id': m.group('format_id'),
-                    'url': url,
-                    'vcodec': 'none' if m.group('type') == 'audio' else None
-                }],
+                'formats': formats,
                 'upload_date': upload_date,
             }
 
@@ -1284,7 +1303,7 @@ class GenericIE(InfoExtractor):
 
         self.report_extraction(video_id)
 
-        # Is it an RSS feed, a SMIL file or a XSPF playlist?
+        # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest?
         try:
             doc = compat_etree_fromstring(webpage.encode('utf-8'))
             if doc.tag == 'rss':
@@ -1293,6 +1312,13 @@ class GenericIE(InfoExtractor):
                 return self._parse_smil(doc, url, video_id)
             elif doc.tag == '{http://xspf.org/ns/0/}playlist':
                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
+            elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
+                return {
+                    'id': video_id,
+                    'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
+                    'formats': self._parse_mpd_formats(
+                        doc, video_id, mpd_base_url=url.rpartition('/')[0]),
+                }
         except compat_xml_parse_error:
             pass
 
@@ -1402,7 +1428,7 @@ class GenericIE(InfoExtractor):
 
         # Look for embedded Dailymotion player
         matches = re.findall(
-            r'<(?:embed|iframe)[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
+            r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
         if matches:
             return _playlist_from_matches(
                 matches, lambda m: unescapeHTML(m[1]))
@@ -1547,6 +1573,11 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'VK')
 
+        # Look for embedded Odnoklassniki player
+        mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Odnoklassniki')
+
         # Look for embedded ivi player
         mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
         if mobj is not None:
@@ -1819,6 +1850,17 @@ class GenericIE(InfoExtractor):
         if digiteka_url:
             return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key())
 
+        # Look for Limelight embeds
+        mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
+        if mobj:
+            lm = {
+                'Media': 'media',
+                'Channel': 'channel',
+                'ChannelList': 'channel_list',
+            }
+            return self.url_result('limelight:%s:%s' % (
+                lm[mobj.group(1)], mobj.group(2)), 'Limelight%s' % mobj.group(1), mobj.group(2))
+
         # Look for AdobeTVVideo embeds
         mobj = re.search(
             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
@@ -1935,6 +1977,8 @@ class GenericIE(InfoExtractor):
                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
             elif ext == 'm3u8':
                 entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
+            elif ext == 'mpd':
+                entry_info_dict['formats'] = self._extract_mpd_formats(video_url, video_id)
             else:
                 entry_info_dict['url'] = video_url
 

youtube_dl/extractor/globo.py

@@ -65,7 +65,7 @@ class GloboIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    class MD5:
+    class MD5(object):
         HEX_FORMAT_LOWERCASE = 0
         HEX_FORMAT_UPPERCASE = 1
         BASE64_PAD_CHARACTER_DEFAULT_COMPLIANCE = ''

youtube_dl/extractor/googledrive.py

@@ -82,7 +82,7 @@ class GoogleDriveIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
             'duration': duration,
             'formats': formats,
         }

youtube_dl/extractor/hentaistigma.py

@@ -11,8 +11,8 @@ class HentaiStigmaIE(InfoExtractor):
         'info_dict': {
             'id': 'inyouchuu-etsu-bonus',
             'ext': 'mp4',
-            "title": "Inyouchuu Etsu Bonus",
-            "age_limit": 18,
+            'title': 'Inyouchuu Etsu Bonus',
+            'age_limit': 18,
         }
     }
 

youtube_dl/extractor/hotstar.py

@@ -10,8 +10,8 @@ from ..utils import (
 
 
 class HotStarIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?hotstar\.com/.*?[/-](?P<id>\d{10})'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
+    _TESTS = [{
         'url': 'http://www.hotstar.com/on-air-with-aib--english-1000076273',
         'info_dict': {
             'id': '1000076273',
@@ -26,7 +26,13 @@ class HotStarIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
-    }
+    }, {
+        'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.hotstar.com/1000000515',
+        'only_matching': True,
+    }]
 
     _GET_CONTENT_TEMPLATE = 'http://account.hotstar.com/AVS/besc?action=GetAggregatedContentDetails&channel=PCTV&contentId=%s'
     _GET_CDN_TEMPLATE = 'http://getcdn.hotstar.com/AVS/besc?action=GetCDN&asJson=Y&channel=%s&id=%s&type=%s'

youtube_dl/extractor/instagram.py

@@ -21,6 +21,18 @@ class InstagramIE(InfoExtractor):
             'title': 'Video by naomipq',
             'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
         }
+    }, {
+        # missing description
+        'url': 'https://www.instagram.com/p/BA-pQFBG8HZ/?taken-by=britneyspears',
+        'info_dict': {
+            'id': 'BA-pQFBG8HZ',
+            'ext': 'mp4',
+            'uploader_id': 'britneyspears',
+            'title': 'Video by britneyspears',
+        },
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'https://instagram.com/p/-Cmh1cukG2/',
         'only_matching': True,
@@ -32,8 +44,8 @@ class InstagramIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
                                          webpage, 'uploader id', fatal=False)
-        desc = self._search_regex(r'"caption":"(.*?)"', webpage, 'description',
-                                  fatal=False)
+        desc = self._search_regex(
+            r'"caption":"(.+?)"', webpage, 'description', default=None)
 
         return {
             'id': video_id,

youtube_dl/extractor/iprima.py

@@ -2,46 +2,30 @@
 from __future__ import unicode_literals
 
 import re
-from random import random
-from math import floor
+import time
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
-    remove_end,
     sanitized_Request,
 )
 
 
 class IPrimaIE(InfoExtractor):
-    _WORKING = False
-    _VALID_URL = r'https?://play\.iprima\.cz/(?:[^/]+/)*(?P<id>[^?#]+)'
+    _VALID_URL = r'https?://play\.iprima\.cz/(?:.+/)?(?P<id>[^?#]+)'
 
     _TESTS = [{
+        'url': 'http://play.iprima.cz/gondici-s-r-o-33',
+        'info_dict': {
+            'id': 'p136534',
+            'ext': 'mp4',
+            'title': 'Gondíci s. r. o. (34)',
+            'description': 'md5:16577c629d006aa91f59ca8d8e7f99bd',
+        },
+        'params': {
+            'skip_download': True,  # m3u8 download
+        },
+    }, {
         'url': 'http://play.iprima.cz/particka/particka-92',
-        'info_dict': {
-            'id': '39152',
-            'ext': 'flv',
-            'title': 'Partička (92)',
-            'description': 'md5:74e9617e51bca67c3ecfb2c6f9766f45',
-            'thumbnail': 'http://play.iprima.cz/sites/default/files/image_crops/image_620x349/3/491483_particka-92_image_620x349.jpg',
-        },
-        'params': {
-            'skip_download': True,  # requires rtmpdump
-        },
-    }, {
-        'url': 'http://play.iprima.cz/particka/tchibo-particka-jarni-moda',
-        'info_dict': {
-            'id': '9718337',
-            'ext': 'flv',
-            'title': 'Tchibo Partička - Jarní móda',
-            'thumbnail': 're:^http:.*\.jpg$',
-        },
-        'params': {
-            'skip_download': True,  # requires rtmpdump
-        },
-    }, {
-        'url': 'http://play.iprima.cz/zpravy-ftv-prima-2752015',
         'only_matching': True,
     }]
 
@@ -51,62 +35,24 @@ class IPrimaIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        if re.search(r'Nemáte oprávnění přistupovat na tuto stránku\.\s*</div>', webpage):
-            raise ExtractorError(
-                '%s said: You do not have permission to access this page' % self.IE_NAME, expected=True)
+        video_id = self._search_regex(r'data-product="([^"]+)">', webpage, 'real id')
 
-        player_url = (
-            'http://embed.livebox.cz/iprimaplay/player-embed-v2.js?__tok%s__=%s' %
-            (floor(random() * 1073741824), floor(random() * 1073741824))
-        )
-
-        req = sanitized_Request(player_url)
+        req = sanitized_Request(
+            'http://play.iprima.cz/prehravac/init?_infuse=1'
+            '&_ts=%s&productId=%s' % (round(time.time()), video_id))
         req.add_header('Referer', url)
-        playerpage = self._download_webpage(req, video_id)
+        playerpage = self._download_webpage(req, video_id, note='Downloading player')
 
-        base_url = ''.join(re.findall(r"embed\['stream'\] = '(.+?)'.+'(\?auth=)'.+'(.+?)';", playerpage)[1])
+        m3u8_url = self._search_regex(r"'src': '([^']+\.m3u8)'", playerpage, 'm3u8 url')
 
-        zoneGEO = self._html_search_regex(r'"zoneGEO":(.+?),', webpage, 'zoneGEO')
-        if zoneGEO != '0':
-            base_url = base_url.replace('token', 'token_' + zoneGEO)
-
-        formats = []
-        for format_id in ['lq', 'hq', 'hd']:
-            filename = self._html_search_regex(
-                r'"%s_id":(.+?),' % format_id, webpage, 'filename')
-
-            if filename == 'null':
-                continue
-
-            real_id = self._search_regex(
-                r'Prima-(?:[0-9]{10}|WEB)-([0-9]+)[-_]',
-                filename, 'real video id')
-
-            if format_id == 'lq':
-                quality = 0
-            elif format_id == 'hq':
-                quality = 1
-            elif format_id == 'hd':
-                quality = 2
-                filename = 'hq/' + filename
-
-            formats.append({
-                'format_id': format_id,
-                'url': base_url,
-                'quality': quality,
-                'play_path': 'mp4:' + filename.replace('"', '')[:-4],
-                'rtmp_live': True,
-                'ext': 'flv',
-            })
+        formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
 
         self._sort_formats(formats)
 
         return {
-            'id': real_id,
-            'title': remove_end(self._og_search_title(webpage), ' | Prima PLAY'),
+            'id': video_id,
+            'title': self._og_search_title(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
             'formats': formats,
-            'description': self._search_regex(
-                r'<p[^>]+itemprop="description"[^>]*>([^<]+)',
-                webpage, 'description', default=None),
+            'description': self._og_search_description(webpage),
         }

youtube_dl/extractor/iqiyi.py

@@ -2,14 +2,194 @@
 from __future__ import unicode_literals
 
 import hashlib
+import itertools
 import math
+import os
 import random
+import re
 import time
 import uuid
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse
-from ..utils import ExtractorError
+from ..compat import (
+    compat_parse_qs,
+    compat_str,
+    compat_urllib_parse,
+    compat_urllib_parse_urlparse,
+)
+from ..utils import (
+    ExtractorError,
+    ohdave_rsa_encrypt,
+    remove_start,
+    sanitized_Request,
+    urlencode_postdata,
+    url_basename,
+)
+
+
+def md5_text(text):
+    return hashlib.md5(text.encode('utf-8')).hexdigest()
+
+
+class IqiyiSDK(object):
+    def __init__(self, target, ip, timestamp):
+        self.target = target
+        self.ip = ip
+        self.timestamp = timestamp
+
+    @staticmethod
+    def split_sum(data):
+        return compat_str(sum(map(lambda p: int(p, 16), list(data))))
+
+    @staticmethod
+    def digit_sum(num):
+        if isinstance(num, int):
+            num = compat_str(num)
+        return compat_str(sum(map(int, num)))
+
+    def even_odd(self):
+        even = self.digit_sum(compat_str(self.timestamp)[::2])
+        odd = self.digit_sum(compat_str(self.timestamp)[1::2])
+        return even, odd
+
+    def preprocess(self, chunksize):
+        self.target = md5_text(self.target)
+        chunks = []
+        for i in range(32 // chunksize):
+            chunks.append(self.target[chunksize * i:chunksize * (i + 1)])
+        if 32 % chunksize:
+            chunks.append(self.target[32 - 32 % chunksize:])
+        return chunks, list(map(int, self.ip.split('.')))
+
+    def mod(self, modulus):
+        chunks, ip = self.preprocess(32)
+        self.target = chunks[0] + ''.join(map(lambda p: compat_str(p % modulus), ip))
+
+    def split(self, chunksize):
+        modulus_map = {
+            4: 256,
+            5: 10,
+            8: 100,
+        }
+
+        chunks, ip = self.preprocess(chunksize)
+        ret = ''
+        for i in range(len(chunks)):
+            ip_part = compat_str(ip[i] % modulus_map[chunksize]) if i < 4 else ''
+            if chunksize == 8:
+                ret += ip_part + chunks[i]
+            else:
+                ret += chunks[i] + ip_part
+        self.target = ret
+
+    def handle_input16(self):
+        self.target = md5_text(self.target)
+        self.target = self.split_sum(self.target[:16]) + self.target + self.split_sum(self.target[16:])
+
+    def handle_input8(self):
+        self.target = md5_text(self.target)
+        ret = ''
+        for i in range(4):
+            part = self.target[8 * i:8 * (i + 1)]
+            ret += self.split_sum(part) + part
+        self.target = ret
+
+    def handleSum(self):
+        self.target = md5_text(self.target)
+        self.target = self.split_sum(self.target) + self.target
+
+    def date(self, scheme):
+        self.target = md5_text(self.target)
+        d = time.localtime(self.timestamp)
+        strings = {
+            'y': compat_str(d.tm_year),
+            'm': '%02d' % d.tm_mon,
+            'd': '%02d' % d.tm_mday,
+        }
+        self.target += ''.join(map(lambda c: strings[c], list(scheme)))
+
+    def split_time_even_odd(self):
+        even, odd = self.even_odd()
+        self.target = odd + md5_text(self.target) + even
+
+    def split_time_odd_even(self):
+        even, odd = self.even_odd()
+        self.target = even + md5_text(self.target) + odd
+
+    def split_ip_time_sum(self):
+        chunks, ip = self.preprocess(32)
+        self.target = compat_str(sum(ip)) + chunks[0] + self.digit_sum(self.timestamp)
+
+    def split_time_ip_sum(self):
+        chunks, ip = self.preprocess(32)
+        self.target = self.digit_sum(self.timestamp) + chunks[0] + compat_str(sum(ip))
+
+
+class IqiyiSDKInterpreter(object):
+    BASE62_TABLE = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
+
+    def __init__(self, sdk_code):
+        self.sdk_code = sdk_code
+
+    @classmethod
+    def base62(cls, num):
+        if num == 0:
+            return '0'
+        ret = ''
+        while num:
+            ret = cls.BASE62_TABLE[num % 62] + ret
+            num = num // 62
+        return ret
+
+    def decode_eval_codes(self):
+        self.sdk_code = self.sdk_code[5:-3]
+
+        mobj = re.search(
+            r"'([^']+)',62,(\d+),'([^']+)'\.split\('\|'\),[^,]+,{}",
+            self.sdk_code)
+        obfucasted_code, count, symbols = mobj.groups()
+        count = int(count)
+        symbols = symbols.split('|')
+        symbol_table = {}
+
+        while count:
+            count -= 1
+            b62count = self.base62(count)
+            symbol_table[b62count] = symbols[count] or b62count
+
+        self.sdk_code = re.sub(
+            r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)],
+            obfucasted_code)
+
+    def run(self, target, ip, timestamp):
+        self.decode_eval_codes()
+
+        functions = re.findall(r'input=([a-zA-Z0-9]+)\(input', self.sdk_code)
+
+        sdk = IqiyiSDK(target, ip, timestamp)
+
+        other_functions = {
+            'handleSum': sdk.handleSum,
+            'handleInput8': sdk.handle_input8,
+            'handleInput16': sdk.handle_input16,
+            'splitTimeEvenOdd': sdk.split_time_even_odd,
+            'splitTimeOddEven': sdk.split_time_odd_even,
+            'splitIpTimeSum': sdk.split_ip_time_sum,
+            'splitTimeIpSum': sdk.split_time_ip_sum,
+        }
+        for function in functions:
+            if re.match(r'mod\d+', function):
+                sdk.mod(int(function[3:]))
+            elif re.match(r'date[ymd]{3}', function):
+                sdk.date(function[4:])
+            elif re.match(r'split\d+', function):
+                sdk.split(int(function[5:]))
+            elif function in other_functions:
+                other_functions[function]()
+            else:
+                raise ExtractorError('Unknown funcion %s' % function)
+
+        return sdk.target
 
 
 class IqiyiIE(InfoExtractor):
@@ -18,6 +198,8 @@ class IqiyiIE(InfoExtractor):
 
     _VALID_URL = r'http://(?:[^.]+\.)?iqiyi\.com/.+\.html'
 
+    _NETRC_MACHINE = 'iqiyi'
+
     _TESTS = [{
         'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
         'md5': '2cb594dc2781e6c941a110d8f358118b',
@@ -93,6 +275,35 @@ class IqiyiIE(InfoExtractor):
     }, {
         'url': 'http://yule.iqiyi.com/pcb.html',
         'only_matching': True,
+    }, {
+        # VIP-only video. The first 2 parts (6 minutes) are available without login
+        # MD5 sums omitted as values are different on Travis CI and my machine
+        'url': 'http://www.iqiyi.com/v_19rrny4w8w.html',
+        'info_dict': {
+            'id': 'f3cf468b39dddb30d676f89a91200dc1',
+            'title': '泰坦尼克号',
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': 'f3cf468b39dddb30d676f89a91200dc1_part1',
+                'ext': 'f4v',
+                'title': '泰坦尼克号',
+            },
+        }, {
+            'info_dict': {
+                'id': 'f3cf468b39dddb30d676f89a91200dc1_part2',
+                'ext': 'f4v',
+                'title': '泰坦尼克号',
+            },
+        }],
+        'expected_warnings': ['Needs a VIP account for full video'],
+    }, {
+        'url': 'http://www.iqiyi.com/a_19rrhb8ce1.html',
+        'info_dict': {
+            'id': '202918101',
+            'title': '灌篮高手 国语版',
+        },
+        'playlist_count': 101,
     }]
 
     _FORMATS_MAP = [
@@ -104,11 +315,98 @@ class IqiyiIE(InfoExtractor):
         ('10', 'h1'),
     ]
 
-    @staticmethod
-    def md5_text(text):
-        return hashlib.md5(text.encode('utf-8')).hexdigest()
+    def _real_initialize(self):
+        self._login()
 
-    def construct_video_urls(self, data, video_id, _uuid):
+    @staticmethod
+    def _rsa_fun(data):
+        # public key extracted from http://static.iqiyi.com/js/qiyiV2/20160129180840/jobs/i18n/i18nIndex.js
+        N = 0xab86b6371b5318aaa1d3c9e612a9f1264f372323c8c0f19875b5fc3b3fd3afcc1e5bec527aa94bfa85bffc157e4245aebda05389a5357b75115ac94f074aefcd
+        e = 65537
+
+        return ohdave_rsa_encrypt(data, e, N)
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+
+        # No authentication to be performed
+        if not username:
+            return True
+
+        data = self._download_json(
+            'http://kylin.iqiyi.com/get_token', None,
+            note='Get token for logging', errnote='Unable to get token for logging')
+        sdk = data['sdk']
+        timestamp = int(time.time())
+        target = '/apis/reglogin/login.action?lang=zh_TW&area_code=null&email=%s&passwd=%s&agenttype=1&from=undefined&keeplogin=0&piccode=&fromurl=&_pos=1' % (
+            username, self._rsa_fun(password.encode('utf-8')))
+
+        interp = IqiyiSDKInterpreter(sdk)
+        sign = interp.run(target, data['ip'], timestamp)
+
+        validation_params = {
+            'target': target,
+            'server': 'BEA3AA1908656AABCCFF76582C4C6660',
+            'token': data['token'],
+            'bird_src': 'f8d91d57af224da7893dd397d52d811a',
+            'sign': sign,
+            'bird_t': timestamp,
+        }
+        validation_result = self._download_json(
+            'http://kylin.iqiyi.com/validate?' + compat_urllib_parse.urlencode(validation_params), None,
+            note='Validate credentials', errnote='Unable to validate credentials')
+
+        MSG_MAP = {
+            'P00107': 'please login via the web interface and enter the CAPTCHA code',
+            'P00117': 'bad username or password',
+        }
+
+        code = validation_result['code']
+        if code != 'A00000':
+            msg = MSG_MAP.get(code)
+            if not msg:
+                msg = 'error %s' % code
+                if validation_result.get('msg'):
+                    msg += ': ' + validation_result['msg']
+            self._downloader.report_warning('unable to log in: ' + msg)
+            return False
+
+        return True
+
+    def _authenticate_vip_video(self, api_video_url, video_id, tvid, _uuid, do_report_warning):
+        auth_params = {
+            # version and platform hard-coded in com/qiyi/player/core/model/remote/AuthenticationRemote.as
+            'version': '2.0',
+            'platform': 'b6c13e26323c537d',
+            'aid': tvid,
+            'tvid': tvid,
+            'uid': '',
+            'deviceId': _uuid,
+            'playType': 'main',  # XXX: always main?
+            'filename': os.path.splitext(url_basename(api_video_url))[0],
+        }
+
+        qd_items = compat_parse_qs(compat_urllib_parse_urlparse(api_video_url).query)
+        for key, val in qd_items.items():
+            auth_params[key] = val[0]
+
+        auth_req = sanitized_Request(
+            'http://api.vip.iqiyi.com/services/ckn.action',
+            urlencode_postdata(auth_params))
+        # iQiyi server throws HTTP 405 error without the following header
+        auth_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        auth_result = self._download_json(
+            auth_req, video_id,
+            note='Downloading video authentication JSON',
+            errnote='Unable to download video authentication JSON')
+        if auth_result['code'] == 'Q00506':  # requires a VIP account
+            if do_report_warning:
+                self.report_warning('Needs a VIP account for full video')
+            return False
+
+        return auth_result
+
+    def construct_video_urls(self, data, video_id, _uuid, tvid):
         def do_xor(x, y):
             a = y % 3
             if a == 1:
@@ -134,9 +432,10 @@ class IqiyiIE(InfoExtractor):
                 note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
             )['t']
             t = str(int(math.floor(int(tm) / (600.0))))
-            return self.md5_text(t + mg + x)
+            return md5_text(t + mg + x)
 
         video_urls_dict = {}
+        need_vip_warning_report = True
         for format_item in data['vp']['tkl'][0]['vs']:
             if 0 < int(format_item['bid']) <= 10:
                 format_id = self.get_format(format_item['bid'])
@@ -155,11 +454,13 @@ class IqiyiIE(InfoExtractor):
                 vl = segment['l']
                 if not vl.startswith('/'):
                     vl = get_encode_code(vl)
-                key = get_path_key(
-                    vl.split('/')[-1].split('.')[0], format_id, segment_index)
+                is_vip_video = '/vip/' in vl
                 filesize = segment['b']
                 base_url = data['vp']['du'].split('/')
-                base_url.insert(-1, key)
+                if not is_vip_video:
+                    key = get_path_key(
+                        vl.split('/')[-1].split('.')[0], format_id, segment_index)
+                    base_url.insert(-1, key)
                 base_url = '/'.join(base_url)
                 param = {
                     'su': _uuid,
@@ -170,8 +471,23 @@ class IqiyiIE(InfoExtractor):
                     'ct': '',
                     'tn': str(int(time.time()))
                 }
-                api_video_url = base_url + vl + '?' + \
-                    compat_urllib_parse.urlencode(param)
+                api_video_url = base_url + vl
+                if is_vip_video:
+                    api_video_url = api_video_url.replace('.f4v', '.hml')
+                    auth_result = self._authenticate_vip_video(
+                        api_video_url, video_id, tvid, _uuid, need_vip_warning_report)
+                    if auth_result is False:
+                        need_vip_warning_report = False
+                        break
+                    param.update({
+                        't': auth_result['data']['t'],
+                        # cid is hard-coded in com/qiyi/player/core/player/RuntimeData.as
+                        'cid': 'afbe8fd3d73448c9',
+                        'vid': video_id,
+                        'QY00001': auth_result['data']['u'],
+                    })
+                api_video_url += '?' if '?' not in api_video_url else '&'
+                api_video_url += compat_urllib_parse.urlencode(param)
                 js = self._download_json(
                     api_video_url, video_id,
                     note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
@@ -195,16 +511,17 @@ class IqiyiIE(InfoExtractor):
         tail = tm + tvid
         param = {
             'key': 'fvip',
-            'src': self.md5_text('youtube-dl'),
+            'src': md5_text('youtube-dl'),
             'tvId': tvid,
             'vid': video_id,
             'vinfo': 1,
             'tm': tm,
-            'enc': self.md5_text(enc_key + tail),
+            'enc': md5_text(enc_key + tail),
             'qyid': _uuid,
             'tn': random.random(),
             'um': 0,
-            'authkey': self.md5_text(self.md5_text('') + tail),
+            'authkey': md5_text(md5_text('') + tail),
+            'k_tag': 1,
         }
 
         api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
@@ -218,9 +535,49 @@ class IqiyiIE(InfoExtractor):
         enc_key = '6ab6d0280511493ba85594779759d4ed'
         return enc_key
 
+    def _extract_playlist(self, webpage):
+        PAGE_SIZE = 50
+
+        links = re.findall(
+            r'<a[^>]+class="site-piclist_pic_link"[^>]+href="(http://www\.iqiyi\.com/.+\.html)"',
+            webpage)
+        if not links:
+            return
+
+        album_id = self._search_regex(
+            r'albumId\s*:\s*(\d+),', webpage, 'album ID')
+        album_title = self._search_regex(
+            r'data-share-title="([^"]+)"', webpage, 'album title', fatal=False)
+
+        entries = list(map(self.url_result, links))
+
+        # Start from 2 because links in the first page are already on webpage
+        for page_num in itertools.count(2):
+            pagelist_page = self._download_webpage(
+                'http://cache.video.qiyi.com/jp/avlist/%s/%d/%d/' % (album_id, page_num, PAGE_SIZE),
+                album_id,
+                note='Download playlist page %d' % page_num,
+                errnote='Failed to download playlist page %d' % page_num)
+            pagelist = self._parse_json(
+                remove_start(pagelist_page, 'var tvInfoJs='), album_id)
+            vlist = pagelist['data']['vlist']
+            for item in vlist:
+                entries.append(self.url_result(item['vurl']))
+            if len(vlist) < PAGE_SIZE:
+                break
+
+        return self.playlist_result(entries, album_id, album_title)
+
     def _real_extract(self, url):
         webpage = self._download_webpage(
             url, 'temp_id', note='download video page')
+
+        # There's no simple way to determine whether an URL is a playlist or not
+        # So detect it
+        playlist_result = self._extract_playlist(webpage)
+        if playlist_result:
+            return playlist_result
+
         tvid = self._search_regex(
             r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
         video_id = self._search_regex(
@@ -236,16 +593,13 @@ class IqiyiIE(InfoExtractor):
         if raw_data['code'] != 'A000000':
             raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
 
-        if not raw_data['data']['vp']['tkl']:
-            raise ExtractorError('No support iQiqy VIP video')
-
         data = raw_data['data']
 
         title = data['vi']['vn']
 
         # generate video_urls_dict
         video_urls_dict = self.construct_video_urls(
-            data, video_id, _uuid)
+            data, video_id, _uuid, tvid)
 
         # construct info
         entries = []

youtube_dl/extractor/kankan.py

@@ -28,7 +28,7 @@ class KankanIE(InfoExtractor):
 
         title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, 'video title')
         surls = re.search(r'surls:\[\'.+?\'\]|lurl:\'.+?\.flv\'', webpage).group(0)
-        gcids = re.findall(r"http://.+?/.+?/(.+?)/", surls)
+        gcids = re.findall(r'http://.+?/.+?/(.+?)/', surls)
         gcid = gcids[-1]
 
         info_url = 'http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid

youtube_dl/extractor/kickstarter.py

@@ -2,12 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import smuggle_url
 
 
 class KickStarterIE(InfoExtractor):
     _VALID_URL = r'https?://www\.kickstarter\.com/projects/(?P<id>[^/]*)/.*'
     _TESTS = [{
-        'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant?ref=home_location',
+        'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant/description',
         'md5': 'c81addca81327ffa66c642b5d8b08cab',
         'info_dict': {
             'id': '1404461844',
@@ -27,7 +28,8 @@ class KickStarterIE(InfoExtractor):
             'uploader_id': 'pebble',
             'uploader': 'Pebble Technology',
             'title': 'Pebble iOS Notifications',
-        }
+        },
+        'add_ie': ['Vimeo'],
     }, {
         'url': 'https://www.kickstarter.com/projects/1420158244/power-drive-2000/widget/video.html',
         'info_dict': {
@@ -43,7 +45,7 @@ class KickStarterIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         title = self._html_search_regex(
-            r'<title>\s*(.*?)(?:\s*&mdash; Kickstarter)?\s*</title>',
+            r'<title>\s*(.*?)(?:\s*&mdash;\s*Kickstarter)?\s*</title>',
             webpage, 'title')
         video_url = self._search_regex(
             r'data-video-url="(.*?)"',
@@ -52,7 +54,7 @@ class KickStarterIE(InfoExtractor):
             return {
                 '_type': 'url_transparent',
                 'ie_key': 'Generic',
-                'url': url,
+                'url': smuggle_url(url, {'to_generic': True}),
                 'title': title,
             }
 

youtube_dl/extractor/konserthusetplay.py

@@ -0,0 +1,107 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    int_or_none,
+)
+
+
+class KonserthusetPlayIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?konserthusetplay\.se/\?.*\bm=(?P<id>[^&]+)'
+    _TEST = {
+        'url': 'http://www.konserthusetplay.se/?m=CKDDnlCY-dhWAAqiMERd-A',
+        'info_dict': {
+            'id': 'CKDDnlCY-dhWAAqiMERd-A',
+            'ext': 'flv',
+            'title': 'Orkesterns instrument: Valthornen',
+            'description': 'md5:f10e1f0030202020396a4d712d2fa827',
+            'thumbnail': 're:^https?://.*$',
+            'duration': 398.8,
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        e = self._search_regex(
+            r'https?://csp\.picsearch\.com/rest\?.*\be=(.+?)[&"\']', webpage, 'e')
+
+        rest = self._download_json(
+            'http://csp.picsearch.com/rest?e=%s&containerId=mediaplayer&i=object' % e,
+            video_id, transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
+
+        media = rest['media']
+        player_config = media['playerconfig']
+        playlist = player_config['playlist']
+
+        source = next(f for f in playlist if f.get('bitrates'))
+
+        FORMAT_ID_REGEX = r'_([^_]+)_h264m\.mp4'
+
+        formats = []
+
+        fallback_url = source.get('fallbackUrl')
+        fallback_format_id = None
+        if fallback_url:
+            fallback_format_id = self._search_regex(
+                FORMAT_ID_REGEX, fallback_url, 'format id', default=None)
+
+        connection_url = (player_config.get('rtmp', {}).get(
+            'netConnectionUrl') or player_config.get(
+            'plugins', {}).get('bwcheck', {}).get('netConnectionUrl'))
+        if connection_url:
+            for f in source['bitrates']:
+                video_url = f.get('url')
+                if not video_url:
+                    continue
+                format_id = self._search_regex(
+                    FORMAT_ID_REGEX, video_url, 'format id', default=None)
+                f_common = {
+                    'vbr': int_or_none(f.get('bitrate')),
+                    'width': int_or_none(f.get('width')),
+                    'height': int_or_none(f.get('height')),
+                }
+                f = f_common.copy()
+                f.update({
+                    'url': connection_url,
+                    'play_path': video_url,
+                    'format_id': 'rtmp-%s' % format_id if format_id else 'rtmp',
+                    'ext': 'flv',
+                })
+                formats.append(f)
+                if format_id and format_id == fallback_format_id:
+                    f = f_common.copy()
+                    f.update({
+                        'url': fallback_url,
+                        'format_id': 'http-%s' % format_id if format_id else 'http',
+                    })
+                    formats.append(f)
+
+        if not formats and fallback_url:
+            formats.append({
+                'url': fallback_url,
+            })
+
+        self._sort_formats(formats)
+
+        title = player_config.get('title') or media['title']
+        description = player_config.get('mediaInfo', {}).get('description')
+        thumbnail = media.get('image')
+        duration = float_or_none(media.get('duration'), 1000)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/kuwo.py

@@ -31,6 +31,10 @@ class KuwoBaseIE(InfoExtractor):
                 (file_format['ext'], file_format.get('br', ''), song_id),
                 song_id, note='Download %s url info' % file_format['format'],
             )
+
+            if song_url == 'IPDeny':
+                raise ExtractorError('This song is blocked in this region', expected=True)
+
             if song_url.startswith('http://') or song_url.startswith('https://'):
                 formats.append({
                     'url': song_url,

youtube_dl/extractor/laola1tv.py

@@ -1,86 +1,125 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
-import random
 import re
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse,
+    compat_urlparse,
+)
 from ..utils import (
     ExtractorError,
+    sanitized_Request,
+    unified_strdate,
+    urlencode_postdata,
+    xpath_element,
     xpath_text,
 )
 
 
 class Laola1TvIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?laola1\.tv/(?P<lang>[a-z]+)-(?P<portal>[a-z]+)/.*?/(?P<id>[0-9]+)\.html'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?laola1\.tv/(?P<lang>[a-z]+)-(?P<portal>[a-z]+)/[^/]+/(?P<slug>[^/?#&]+)'
+    _TESTS = [{
         'url': 'http://www.laola1.tv/de-de/video/straubing-tigers-koelner-haie/227883.html',
         'info_dict': {
             'id': '227883',
-            'ext': 'mp4',
+            'display_id': 'straubing-tigers-koelner-haie',
+            'ext': 'flv',
             'title': 'Straubing Tigers - Kölner Haie',
-            'categories': ['Eishockey'],
+            'upload_date': '20140912',
             'is_live': False,
+            'categories': ['Eishockey'],
         },
         'params': {
             'skip_download': True,
         }
-    }
+    }, {
+        'url': 'http://www.laola1.tv/de-de/video/straubing-tigers-koelner-haie',
+        'info_dict': {
+            'id': '464602',
+            'display_id': 'straubing-tigers-koelner-haie',
+            'ext': 'flv',
+            'title': 'Straubing Tigers - Kölner Haie',
+            'upload_date': '20160129',
+            'is_live': False,
+            'categories': ['Eishockey'],
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        display_id = mobj.group('slug')
         lang = mobj.group('lang')
         portal = mobj.group('portal')
 
-        webpage = self._download_webpage(url, video_id)
-        iframe_url = self._search_regex(
-            r'<iframe[^>]*?class="main_tv_player"[^>]*?src="([^"]+)"',
-            webpage, 'iframe URL')
+        webpage = self._download_webpage(url, display_id)
 
-        iframe = self._download_webpage(
-            iframe_url, video_id, note='Downloading iframe')
-        flashvars_m = re.findall(
-            r'flashvars\.([_a-zA-Z0-9]+)\s*=\s*"([^"]*)";', iframe)
-        flashvars = dict((m[0], m[1]) for m in flashvars_m)
+        iframe_url = self._search_regex(
+            r'<iframe[^>]*?id="videoplayer"[^>]*?src="([^"]+)"',
+            webpage, 'iframe url')
+
+        video_id = self._search_regex(
+            r'videoid=(\d+)', iframe_url, 'video id')
+
+        iframe = self._download_webpage(compat_urlparse.urljoin(
+            url, iframe_url), display_id, 'Downloading iframe')
 
         partner_id = self._search_regex(
-            r'partnerid\s*:\s*"([^"]+)"', iframe, 'partner id')
+            r'partnerid\s*:\s*(["\'])(?P<partner_id>.+?)\1',
+            iframe, 'partner id', group='partner_id')
 
-        xml_url = ('http://www.laola1.tv/server/hd_video.php?' +
-                   'play=%s&partner=%s&portal=%s&v5ident=&lang=%s' % (
-                       video_id, partner_id, portal, lang))
-        hd_doc = self._download_xml(xml_url, video_id)
+        hd_doc = self._download_xml(
+            'http://www.laola1.tv/server/hd_video.php?%s'
+            % compat_urllib_parse.urlencode({
+                'play': video_id,
+                'partner': partner_id,
+                'portal': portal,
+                'lang': lang,
+                'v5ident': '',
+            }), display_id)
 
-        title = xpath_text(hd_doc, './/video/title', fatal=True)
-        flash_url = xpath_text(hd_doc, './/video/url', fatal=True)
-        uploader = xpath_text(hd_doc, './/video/meta_organistation')
-        is_live = xpath_text(hd_doc, './/video/islive') == 'true'
+        _v = lambda x, **k: xpath_text(hd_doc, './/video/' + x, **k)
+        title = _v('title', fatal=True)
 
-        categories = xpath_text(hd_doc, './/video/meta_sports')
-        if categories:
-            categories = categories.split(',')
+        req = sanitized_Request(
+            'https://club.laola1.tv/sp/laola1/api/v3/user/session/premium/player/stream-access?%s' %
+            compat_urllib_parse.urlencode({
+                'videoId': video_id,
+                'target': '2',
+                'label': 'laola1tv',
+                'area': _v('area'),
+            }),
+            urlencode_postdata(
+                dict((i, v) for i, v in enumerate(_v('req_liga_abos').split(',')))))
 
-        ident = random.randint(10000000, 99999999)
-        token_url = '%s&ident=%s&klub=0&unikey=0&timestamp=%s&auth=%s' % (
-            flash_url, ident, flashvars['timestamp'], flashvars['auth'])
+        token_url = self._download_json(req, display_id)['data']['stream-access'][0]
+        token_doc = self._download_xml(token_url, display_id, 'Downloading token')
 
-        token_doc = self._download_xml(
-            token_url, video_id, note='Downloading token')
-        token_attrib = token_doc.find('.//token').attrib
-        if token_attrib.get('auth') in ('blocked', 'restricted'):
+        token_attrib = xpath_element(token_doc, './/token').attrib
+        token_auth = token_attrib['auth']
+
+        if token_auth in ('blocked', 'restricted', 'error'):
             raise ExtractorError(
-                'Token error: %s' % token_attrib.get('comment'), expected=True)
+                'Token error: %s' % token_attrib['comment'], expected=True)
 
-        video_url = '%s?hdnea=%s&hdcore=3.2.0' % (
-            token_attrib['url'], token_attrib['auth'])
+        formats = self._extract_f4m_formats(
+            '%s?hdnea=%s&hdcore=3.2.0' % (token_attrib['url'], token_auth),
+            video_id, f4m_id='hds')
+
+        categories_str = _v('meta_sports')
+        categories = categories_str.split(',') if categories_str else []
 
         return {
             'id': video_id,
-            'is_live': is_live,
+            'display_id': display_id,
             'title': title,
-            'url': video_url,
-            'uploader': uploader,
+            'upload_date': unified_strdate(_v('time_date')),
+            'uploader': _v('meta_organisation'),
             'categories': categories,
-            'ext': 'mp4',
+            'is_live': _v('islive') == 'true',
+            'formats': formats,
         }

youtube_dl/extractor/letv.py

@@ -5,11 +5,13 @@ import datetime
 import re
 import time
 import base64
+import hashlib
 
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
     compat_ord,
+    compat_str,
 )
 from ..utils import (
     determine_ext,
@@ -258,6 +260,7 @@ class LetvCloudIE(InfoExtractor):
         },
     }, {
         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
+        'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
         'info_dict': {
             'id': 'p7jnfw5hw9_ec93197892',
             'ext': 'mp4',
@@ -265,6 +268,7 @@ class LetvCloudIE(InfoExtractor):
         },
     }, {
         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
+        'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
         'info_dict': {
             'id': 'p7jnfw5hw9_187060b6fd',
             'ext': 'mp4',
@@ -272,21 +276,37 @@ class LetvCloudIE(InfoExtractor):
         },
     }]
 
-    def _real_extract(self, url):
-        uu_mobj = re.search('uu=([\w]+)', url)
-        vu_mobj = re.search('vu=([\w]+)', url)
+    @staticmethod
+    def sign_data(obj):
+        if obj['cf'] == 'flash':
+            salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
+            items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
+        elif obj['cf'] == 'html5':
+            salt = 'fbeh5player12c43eccf2bec3300344'
+            items = ['cf', 'ran', 'uu', 'bver', 'vu']
+        input_data = ''.join([item + obj[item] for item in items]) + salt
+        obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
 
-        if not uu_mobj or not vu_mobj:
-            raise ExtractorError('Invalid URL: %s' % url, expected=True)
+    def _get_formats(self, cf, uu, vu, media_id):
+        def get_play_json(cf, timestamp):
+            data = {
+                'cf': cf,
+                'ver': '2.2',
+                'bver': 'firefox44.0',
+                'format': 'json',
+                'uu': uu,
+                'vu': vu,
+                'ran': compat_str(timestamp),
+            }
+            self.sign_data(data)
+            return self._download_json(
+                'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse.urlencode(data),
+                media_id, 'Downloading playJson data for type %s' % cf)
 
-        uu = uu_mobj.group(1)
-        vu = vu_mobj.group(1)
-        media_id = uu + '_' + vu
-
-        play_json_req = sanitized_Request(
-            'http://api.letvcloud.com/gpc.php?cf=html5&sign=signxxxxx&ver=2.2&format=json&' +
-            'uu=' + uu + '&vu=' + vu)
-        play_json = self._download_json(play_json_req, media_id, 'Downloading playJson data')
+        play_json = get_play_json(cf, time.time())
+        # The server time may be different from local time
+        if play_json.get('code') == 10071:
+            play_json = get_play_json(cf, play_json['timestamp'])
 
         if not play_json.get('data'):
             if play_json.get('message'):
@@ -312,6 +332,21 @@ class LetvCloudIE(InfoExtractor):
                 'width': int_or_none(play_url.get('vwidth')),
                 'height': int_or_none(play_url.get('vheight')),
             })
+
+        return formats
+
+    def _real_extract(self, url):
+        uu_mobj = re.search('uu=([\w]+)', url)
+        vu_mobj = re.search('vu=([\w]+)', url)
+
+        if not uu_mobj or not vu_mobj:
+            raise ExtractorError('Invalid URL: %s' % url, expected=True)
+
+        uu = uu_mobj.group(1)
+        vu = vu_mobj.group(1)
+        media_id = uu + '_' + vu
+
+        formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/limelight.py

@@ -40,7 +40,8 @@ class LimelightBaseIE(InfoExtractor):
             if not stream_url:
                 continue
             if '.f4m' in stream_url:
-                formats.extend(self._extract_f4m_formats(stream_url, video_id))
+                formats.extend(self._extract_f4m_formats(
+                    stream_url, video_id, fatal=False))
             else:
                 fmt = {
                     'url': stream_url,
@@ -72,8 +73,8 @@ class LimelightBaseIE(InfoExtractor):
             format_id = mobile_url.get('targetMediaPlatform')
             if determine_ext(media_url) == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    media_url, video_id, 'mp4', entry_protocol='m3u8_native',
-                    preference=-1, m3u8_id=format_id))
+                    media_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id=format_id, fatal=False))
             else:
                 formats.append({
                     'url': media_url,

youtube_dl/extractor/liveleak.py

@@ -47,7 +47,7 @@ class LiveLeakIE(InfoExtractor):
         'info_dict': {
             'id': '801_1409392012',
             'ext': 'mp4',
-            'description': "Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.",
+            'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
             'uploader': 'bony333',
             'title': 'Crazy Hungarian tourist films close call waterspout in Croatia'
         }

youtube_dl/extractor/mailru.py

@@ -4,6 +4,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    remove_end,
+)
 
 
 class MailRuIE(InfoExtractor):
@@ -34,14 +38,30 @@ class MailRuIE(InfoExtractor):
                 'id': '46843144_1263',
                 'ext': 'mp4',
                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
-                'timestamp': 1397217632,
-                'upload_date': '20140411',
-                'uploader': 'hitech',
+                'timestamp': 1397039888,
+                'upload_date': '20140409',
+                'uploader': 'hitech@corp.mail.ru',
                 'uploader_id': 'hitech@corp.mail.ru',
                 'duration': 245,
             },
             'skip': 'Not accessible from Travis CI server',
         },
+        {
+            # only available via metaUrl API
+            'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html',
+            'md5': '3b26d2491c6949d031a32b96bd97c096',
+            'info_dict': {
+                'id': '56664382_502',
+                'ext': 'mp4',
+                'title': ':8336',
+                'timestamp': 1449094163,
+                'upload_date': '20151202',
+                'uploader': '720pizle@mail.ru',
+                'uploader_id': '720pizle@mail.ru',
+                'duration': 6001,
+            },
+            'skip': 'Not accessible from Travis CI server',
+        }
     ]
 
     def _real_extract(self, url):
@@ -51,32 +71,55 @@ class MailRuIE(InfoExtractor):
         if not video_id:
             video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
 
-        video_data = self._download_json(
-            'http://api.video.mail.ru/videos/%s.json?new=1' % video_id, video_id, 'Downloading video JSON')
+        webpage = self._download_webpage(url, video_id)
 
-        author = video_data['author']
-        uploader = author['name']
-        uploader_id = author.get('id') or author.get('email')
-        view_count = video_data.get('views_count')
+        video_data = None
+
+        page_config = self._parse_json(self._search_regex(
+            r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
+            webpage, 'page config', default='{}'), video_id, fatal=False)
+        if page_config:
+            meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl')
+            if meta_url:
+                video_data = self._download_json(
+                    meta_url, video_id, 'Downloading video meta JSON', fatal=False)
+
+        # Fallback old approach
+        if not video_data:
+            video_data = self._download_json(
+                'http://api.video.mail.ru/videos/%s.json?new=1' % video_id,
+                video_id, 'Downloading video JSON')
+
+        formats = []
+        for f in video_data['videos']:
+            video_url = f.get('url')
+            if not video_url:
+                continue
+            format_id = f.get('key')
+            height = int_or_none(self._search_regex(
+                r'^(\d+)[pP]$', format_id, 'height', default=None)) if format_id else None
+            formats.append({
+                'url': video_url,
+                'format_id': format_id,
+                'height': height,
+            })
+        self._sort_formats(formats)
 
         meta_data = video_data['meta']
-        content_id = '%s_%s' % (
-            meta_data.get('accId', ''), meta_data['itemId'])
-        title = meta_data['title']
-        if title.endswith('.mp4'):
-            title = title[:-4]
-        thumbnail = meta_data['poster']
-        duration = meta_data['duration']
-        timestamp = meta_data['timestamp']
+        title = remove_end(meta_data['title'], '.mp4')
 
-        formats = [
-            {
-                'url': video['url'],
-                'format_id': video['key'],
-                'height': int(video['key'].rstrip('p'))
-            } for video in video_data['videos']
-        ]
-        self._sort_formats(formats)
+        author = video_data.get('author')
+        uploader = author.get('name')
+        uploader_id = author.get('id') or author.get('email')
+        view_count = int_or_none(video_data.get('viewsCount') or video_data.get('views_count'))
+
+        acc_id = meta_data.get('accId')
+        item_id = meta_data.get('itemId')
+        content_id = '%s_%s' % (acc_id, item_id) if acc_id and item_id else video_id
+
+        thumbnail = meta_data.get('poster')
+        duration = int_or_none(meta_data.get('duration'))
+        timestamp = int_or_none(meta_data.get('timestamp'))
 
         return {
             'id': content_id,

youtube_dl/extractor/matchtv.py

@@ -0,0 +1,55 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import random
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_parse
+from ..utils import (
+    sanitized_Request,
+    xpath_text,
+)
+
+
+class MatchTVIE(InfoExtractor):
+    _VALID_URL = r'https?://matchtv\.ru/?#live-player'
+    _TEST = {
+        'url': 'http://matchtv.ru/#live-player',
+        'info_dict': {
+            'id': 'matchtv-live',
+            'ext': 'flv',
+            'title': 're:^Матч ТВ - Прямой эфир \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = 'matchtv-live'
+        request = sanitized_Request(
+            'http://player.matchtv.ntvplus.tv/player/smil?%s' % compat_urllib_parse.urlencode({
+                'ts': '',
+                'quality': 'SD',
+                'contentId': '561d2c0df7159b37178b4567',
+                'sign': '',
+                'includeHighlights': '0',
+                'userId': '',
+                'sessionId': random.randint(1, 1000000000),
+                'contentType': 'channel',
+                'timeShift': '0',
+                'platform': 'portal',
+            }),
+            headers={
+                'Referer': 'http://player.matchtv.ntvplus.tv/embed-player/NTVEmbedPlayer.swf',
+            })
+        video_url = self._download_json(request, video_id)['data']['videoUrl']
+        f4m_url = xpath_text(self._download_xml(video_url, video_id), './to')
+        formats = self._extract_f4m_formats(f4m_url, video_id)
+        return {
+            'id': video_id,
+            'title': self._live_title('Матч ТВ - Прямой эфир'),
+            'is_live': True,
+            'formats': formats,
+        }

youtube_dl/extractor/mofosex.py

@@ -38,7 +38,7 @@ class MofosexIE(InfoExtractor):
         path = compat_urllib_parse_urlparse(video_url).path
         extension = os.path.splitext(path)[1][1:]
         format = path.split('/')[5].split('_')[:2]
-        format = "-".join(format)
+        format = '-'.join(format)
 
         age_limit = self._rta_search(webpage)
 

youtube_dl/extractor/mtv.py

@@ -11,6 +11,7 @@ from ..utils import (
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
+    float_or_none,
     HEADRequest,
     sanitized_Request,
     unescapeHTML,
@@ -110,7 +111,8 @@ class MTVServicesInfoExtractor(InfoExtractor):
         uri = itemdoc.find('guid').text
         video_id = self._id_from_uri(uri)
         self.report_extraction(video_id)
-        mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
+        content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
+        mediagen_url = content_el.attrib['url']
         # Remove the templates, like &device={device}
         mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
         if 'acceptMethods' not in mediagen_url:
@@ -165,6 +167,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,
+            'duration': float_or_none(content_el.attrib.get('duration')),
         }
 
     def _get_feed_query(self, uri):

youtube_dl/extractor/myspass.py

@@ -18,8 +18,8 @@ class MySpassIE(InfoExtractor):
         'info_dict': {
             'id': '11741',
             'ext': 'mp4',
-            "description": "Wer kann in die Fu\u00dfstapfen von Wolfgang Kubicki treten und die Mehrheit der Zuschauer hinter sich versammeln? Wird vielleicht sogar die Absolute Mehrheit geknackt und der Jackpot von 200.000 Euro mit nach Hause genommen?",
-            "title": "Absolute Mehrheit vom 17.02.2013 - Die Highlights, Teil 2",
+            'description': 'Wer kann in die Fu\u00dfstapfen von Wolfgang Kubicki treten und die Mehrheit der Zuschauer hinter sich versammeln? Wird vielleicht sogar die Absolute Mehrheit geknackt und der Jackpot von 200.000 Euro mit nach Hause genommen?',
+            'title': 'Absolute Mehrheit vom 17.02.2013 - Die Highlights, Teil 2',
         },
     }
 

youtube_dl/extractor/myvideo.py

@@ -19,6 +19,7 @@ from ..utils import (
 
 
 class MyVideoIE(InfoExtractor):
+    _WORKING = False
     _VALID_URL = r'http://(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/(?P<id>[0-9]+)/[^?/]+.*'
     IE_NAME = 'myvideo'
     _TEST = {

youtube_dl/extractor/nba.py

@@ -18,13 +18,17 @@ class NBAIE(InfoExtractor):
         'md5': '9e7729d3010a9c71506fd1248f74e4f4',
         'info_dict': {
             'id': '0021200253-okc-bkn-recap',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Thunder vs. Nets',
             'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
             'duration': 181,
             'timestamp': 1354638466,
             'upload_date': '20121204',
         },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }, {
         'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
         'only_matching': True,
@@ -68,7 +72,7 @@ class NBAIE(InfoExtractor):
             if video_url.startswith('/'):
                 continue
             if video_url.endswith('.m3u8'):
-                formats.extend(self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls', fatal=False))
+                formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls', fatal=False))
             elif video_url.endswith('.f4m'):
                 formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds', fatal=False))
             else:

youtube_dl/extractor/nbc.py

@@ -19,38 +19,45 @@ class NBCIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
-            # md5 checksum is not stable
             'info_dict': {
-                'id': 'c9xnCo0YPOPH',
-                'ext': 'flv',
+                'id': '112966',
+                'ext': 'mp4',
                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
             },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
         },
         {
             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
             'info_dict': {
-                'id': 'XwU9KZkp98TH',
+                'id': '176',
                 'ext': 'flv',
                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
             },
-            'skip': 'Only works from US',
+            'skip': '404 Not Found',
         },
         {
             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
             'info_dict': {
-                'id': '8iUuyzWDdYUZ',
-                'ext': 'flv',
+                'id': '2832821',
+                'ext': 'mp4',
                 'title': 'Star Wars Teaser',
                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
             },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
             'skip': 'Only works from US',
         },
         {
             # This video has expired but with an escaped embedURL
             'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
-            'skip': 'Expired'
+            'only_matching': True,
         }
     ]
 
@@ -66,7 +73,11 @@ class NBCIE(InfoExtractor):
             webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
         if theplatform_url.startswith('//'):
             theplatform_url = 'http:' + theplatform_url
-        return self.url_result(smuggle_url(theplatform_url, {'source_url': url}))
+        return {
+            '_type': 'url_transparent',
+            'url': smuggle_url(theplatform_url, {'source_url': url}),
+            'id': video_id,
+        }
 
 
 class NBCSportsVPlayerIE(InfoExtractor):

youtube_dl/extractor/ndr.py

@@ -193,7 +193,7 @@ class NDREmbedBaseIE(InfoExtractor):
                     src + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, f4m_id='hds'))
             elif ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    src, video_id, m3u8_id='hls', entry_protocol='m3u8_native'))
+                    src, video_id, 'mp4', m3u8_id='hls', entry_protocol='m3u8_native'))
             else:
                 quality = f.get('quality')
                 ff = {

youtube_dl/extractor/nerdcubed.py

@@ -18,14 +18,14 @@ class NerdCubedFeedIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        feed = self._download_json(url, url, "Downloading NerdCubed JSON feed")
+        feed = self._download_json(url, url, 'Downloading NerdCubed JSON feed')
 
         entries = [{
             '_type': 'url',
             'title': feed_entry['title'],
             'uploader': feed_entry['source']['name'] if feed_entry['source'] else None,
             'upload_date': datetime.datetime.strptime(feed_entry['date'], '%Y-%m-%d').strftime('%Y%m%d'),
-            'url': "http://www.youtube.com/watch?v=" + feed_entry['youtube_id'],
+            'url': 'http://www.youtube.com/watch?v=' + feed_entry['youtube_id'],
         } for feed_entry in feed]
 
         return {

youtube_dl/extractor/noz.py

@@ -0,0 +1,69 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
+from ..utils import (
+    int_or_none,
+    xpath_text,
+)
+
+
+class NozIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?noz\.de/video/(?P<id>[0-9]+)/'
+    _TESTS = [{
+        'url': 'http://www.noz.de/video/25151/32-Deutschland-gewinnt-Badminton-Lnderspiel-in-Melle',
+        'info_dict': {
+            'id': '25151',
+            'ext': 'mp4',
+            'duration': 215,
+            'title': '3:2 - Deutschland gewinnt Badminton-Länderspiel in Melle',
+            'description': 'Vor rund 370 Zuschauern gewinnt die deutsche Badminton-Nationalmannschaft am Donnerstag ein EM-Vorbereitungsspiel gegen Frankreich in Melle. Video Moritz Frankenberg.',
+            'thumbnail': 're:^http://.*\.jpg',
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        description = self._og_search_description(webpage)
+
+        edge_url = self._html_search_regex(
+            r'<script\s+(?:type="text/javascript"\s+)?src="(.*?/videojs_.*?)"',
+            webpage, 'edge URL')
+        edge_content = self._download_webpage(edge_url, 'meta configuration')
+
+        config_url_encoded = self._search_regex(
+            r'so\.addVariable\("config_url","[^,]*,(.*?)"',
+            edge_content, 'config URL'
+        )
+        config_url = compat_urllib_parse_unquote(config_url_encoded)
+
+        doc = self._download_xml(config_url, 'video configuration')
+        title = xpath_text(doc, './/title')
+        thumbnail = xpath_text(doc, './/article/thumbnail/url')
+        duration = int_or_none(xpath_text(
+            doc, './/article/movie/file/duration'))
+        formats = []
+        for qnode in doc.findall('.//article/movie/file/qualities/qual'):
+            video_node = qnode.find('./html_urls/video_url[@format="video/mp4"]')
+            if video_node is None:
+                continue  # auto
+            formats.append({
+                'url': video_node.text,
+                'format_name': xpath_text(qnode, './name'),
+                'format_id': xpath_text(qnode, './id'),
+                'height': int_or_none(xpath_text(qnode, './height')),
+                'width': int_or_none(xpath_text(qnode, './width')),
+                'tbr': int_or_none(xpath_text(qnode, './bitrate'), scale=1000),
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': title,
+            'duration': duration,
+            'description': description,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/npo.py

@@ -189,7 +189,7 @@ class NPOIE(NPOBaseIE):
                 if not video_url:
                     continue
                 if format_id == 'adaptive':
-                    formats.extend(self._extract_m3u8_formats(video_url, video_id))
+                    formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4'))
                 else:
                     formats.append({
                         'url': video_url,
@@ -406,6 +406,38 @@ class NPORadioFragmentIE(InfoExtractor):
         }
 
 
+class SchoolTVIE(InfoExtractor):
+    IE_NAME = 'schooltv'
+    _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)'
+
+    _TEST = {
+        'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/',
+        'info_dict': {
+            'id': 'WO_NTR_429477',
+            'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam',
+            'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?',
+            'ext': 'mp4',
+            'description': 'md5:abfa0ff690adb73fd0297fd033aaa631'
+        },
+        'params': {
+            # Skip because of m3u8 download
+            'skip_download': True
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_id = self._search_regex(
+            r'data-mid=(["\'])(?P<id>.+?)\1', webpage, 'video_id', group='id')
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'NPO',
+            'url': 'npo:%s' % video_id,
+            'display_id': display_id
+        }
+
+
 class VPROIE(NPOIE):
     IE_NAME = 'vpro'
     _VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html'

youtube_dl/extractor/nrk.py

@@ -133,26 +133,32 @@ class NRKTVIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
-            'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
             'info_dict': {
                 'id': 'MUHH48000314',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': '20 spørsmål',
                 'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
                 'upload_date': '20140523',
                 'duration': 1741.52,
             },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
         },
         {
             'url': 'https://tv.nrk.no/program/mdfp15000514',
-            'md5': '383650ece2b25ecec996ad7b5bb2a384',
             'info_dict': {
                 'id': 'mdfp15000514',
-                'ext': 'flv',
-                'title': 'Kunnskapskanalen: Grunnlovsjubiléet - Stor ståhei for ingenting',
+                'ext': 'mp4',
+                'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting',
                 'description': 'md5:654c12511f035aed1e42bdf5db3b206a',
                 'upload_date': '20140524',
-                'duration': 4605.0,
+                'duration': 4605.08,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
             },
         },
         {

youtube_dl/extractor/odnoklassniki.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 
 class OdnoklassnikiIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer)/(?P<id>[\d-]+)'
+    _VALID_URL = r'https?://(?:(?:www|m|mobile)\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer)/(?P<id>[\d-]+)'
     _TESTS = [{
         # metadata in JSON
         'url': 'http://ok.ru/video/20079905452',
@@ -69,6 +69,12 @@ class OdnoklassnikiIE(InfoExtractor):
     }, {
         'url': 'http://www.ok.ru/videoembed/20648036891',
         'only_matching': True,
+    }, {
+        'url': 'http://m.ok.ru/video/20079905452',
+        'only_matching': True,
+    }, {
+        'url': 'http://mobile.ok.ru/video/20079905452',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/orf.py

@@ -112,6 +112,7 @@ class ORFTVthekIE(InfoExtractor):
                             % geo_str),
                         fatal=False)
 
+            self._check_formats(formats, video_id)
             self._sort_formats(formats)
 
             upload_date = unified_strdate(sd['created_date'])

youtube_dl/extractor/pbs.py

@@ -4,10 +4,12 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_HTTPError
 from ..utils import (
     ExtractorError,
     determine_ext,
     int_or_none,
+    js_to_json,
     strip_jsonp,
     unified_strdate,
     US_RATINGS,
@@ -199,7 +201,7 @@ class PBSIE(InfoExtractor):
                 'id': '2365006249',
                 'ext': 'mp4',
                 'title': 'Constitution USA with Peter Sagal - A More Perfect Union',
-                'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
+                'description': 'md5:36f341ae62e251b8f5bd2b754b95a071',
                 'duration': 3190,
             },
             'params': {
@@ -213,7 +215,7 @@ class PBSIE(InfoExtractor):
                 'id': '2365297690',
                 'ext': 'mp4',
                 'title': 'FRONTLINE - Losing Iraq',
-                'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
+                'description': 'md5:4d3eaa01f94e61b3e73704735f1196d9',
                 'duration': 5050,
             },
             'params': {
@@ -227,7 +229,7 @@ class PBSIE(InfoExtractor):
                 'id': '2201174722',
                 'ext': 'mp4',
                 'title': 'PBS NewsHour - Cyber Schools Gain Popularity, but Quality Questions Persist',
-                'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28',
+                'description': 'md5:95a19f568689d09a166dff9edada3301',
                 'duration': 801,
             },
         },
@@ -237,8 +239,8 @@ class PBSIE(InfoExtractor):
             'info_dict': {
                 'id': '2365297708',
                 'ext': 'mp4',
-                'description': 'md5:68d87ef760660eb564455eb30ca464fe',
                 'title': 'Great Performances - Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full',
+                'description': 'md5:657897370e09e2bc6bf0f8d2cd313c6b',
                 'duration': 6559,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
@@ -278,7 +280,7 @@ class PBSIE(InfoExtractor):
                 'display_id': 'player',
                 'ext': 'mp4',
                 'title': 'American Experience - Death and the Civil War, Chapter 1',
-                'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.',
+                'description': 'md5:1b80a74e0380ed2a4fb335026de1600d',
                 'duration': 682,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
@@ -287,20 +289,19 @@ class PBSIE(InfoExtractor):
             },
         },
         {
-            'url': 'http://video.pbs.org/video/2365367186/',
+            'url': 'http://www.pbs.org/video/2365245528/',
             'info_dict': {
-                'id': '2365367186',
-                'display_id': '2365367186',
+                'id': '2365245528',
+                'display_id': '2365245528',
                 '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,
+                'title': 'FRONTLINE - United States of Secrets (Part One)',
+                'description': 'md5:55756bd5c551519cc4b7703e373e217e',
+                'duration': 6851,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
             'params': {
                 'skip_download': True,  # requires ffmpeg
             },
-            'skip': 'Expired',
         },
         {
             # Video embedded in iframe containing angle brackets as attribute's value (e.g.
@@ -312,7 +313,7 @@ class PBSIE(InfoExtractor):
                 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business',
                 'ext': 'mp4',
                 'title': "A Chef's Life - Season 3, Ep. 5: Prickly Business",
-                'description': 'md5:61db2ddf27c9912f09c241014b118ed1',
+                'description': 'md5:54033c6baa1f9623607c6e2ed245888b',
                 'duration': 1480,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
@@ -328,7 +329,7 @@ class PBSIE(InfoExtractor):
                 'display_id': 'the-atomic-artists',
                 'ext': 'mp4',
                 'title': 'FRONTLINE - The Atomic Artists',
-                'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
+                'description': 'md5:1a2481e86b32b2e12ec1905dd473e2c1',
                 'duration': 723,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
@@ -336,6 +337,21 @@ class PBSIE(InfoExtractor):
                 'skip_download': True,  # requires ffmpeg
             },
         },
+        {
+            # Serves hd only via wigget/partnerplayer page
+            'url': 'http://www.pbs.org/video/2365641075/',
+            'info_dict': {
+                'id': '2365641075',
+                'ext': 'mp4',
+                'title': 'FRONTLINE - Netanyahu at War',
+                'duration': 6852,
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'formats': 'mincount:8',
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
+        },
         {
             'url': 'http://player.pbs.org/widget/partnerplayer/2365297708/?start=0&end=0&chapterbar=false&endscreen=false&topbar=true',
             'only_matching': True,
@@ -365,10 +381,14 @@ class PBSIE(InfoExtractor):
                 webpage, 'upload date', default=None))
 
             # tabbed frontline videos
-            tabbed_videos = re.findall(
-                r'<div[^>]+class="videotab[^"]*"[^>]+vid="(\d+)"', webpage)
-            if tabbed_videos:
-                return tabbed_videos, presumptive_id, upload_date
+            MULTI_PART_REGEXES = (
+                r'<div[^>]+class="videotab[^"]*"[^>]+vid="(\d+)"',
+                r'<a[^>]+href=["\']#video-\d+["\'][^>]+data-coveid=["\'](\d+)',
+            )
+            for p in MULTI_PART_REGEXES:
+                tabbed_videos = re.findall(p, webpage)
+                if tabbed_videos:
+                    return tabbed_videos, presumptive_id, upload_date
 
             MEDIA_ID_REGEXES = [
                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
@@ -432,22 +452,54 @@ class PBSIE(InfoExtractor):
                 for vid_id in video_id]
             return self.playlist_result(entries, display_id)
 
-        info = self._download_json(
-            'http://player.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
-            display_id)
+        info = None
+        redirects = []
+        redirect_urls = set()
+
+        def extract_redirect_urls(info):
+            for encoding_name in ('recommended_encoding', 'alternate_encoding'):
+                redirect = info.get(encoding_name)
+                if not redirect:
+                    continue
+                redirect_url = redirect.get('url')
+                if redirect_url and redirect_url not in redirect_urls:
+                    redirects.append(redirect)
+                    redirect_urls.add(redirect_url)
+
+        try:
+            video_info = self._download_json(
+                'http://player.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
+                display_id, 'Downloading video info JSON')
+            extract_redirect_urls(video_info)
+            info = video_info
+        except ExtractorError as e:
+            # videoInfo API may not work for some videos
+            if not isinstance(e.cause, compat_HTTPError) or e.cause.code != 404:
+                raise
+
+        # Player pages may also serve different qualities
+        for page in ('widget/partnerplayer', 'portalplayer'):
+            player = self._download_webpage(
+                'http://player.pbs.org/%s/%s' % (page, video_id),
+                display_id, 'Downloading %s page' % page, fatal=False)
+            if player:
+                video_info = self._parse_json(
+                    self._search_regex(
+                        r'(?s)PBS\.videoData\s*=\s*({.+?});\n',
+                        player, '%s video data' % page, default='{}'),
+                    display_id, transform_source=js_to_json, fatal=False)
+                if video_info:
+                    extract_redirect_urls(video_info)
+                    if not info:
+                        info = video_info
 
         formats = []
-        for encoding_name in ('recommended_encoding', 'alternate_encoding'):
-            redirect = info.get(encoding_name)
-            if not redirect:
-                continue
-            redirect_url = redirect.get('url')
-            if not redirect_url:
-                continue
+        for num, redirect in enumerate(redirects):
+            redirect_id = redirect.get('eeid')
 
             redirect_info = self._download_json(
-                redirect_url + '?format=json', display_id,
-                'Downloading %s video url info' % encoding_name)
+                '%s?format=json' % redirect['url'], display_id,
+                'Downloading %s video url info' % (redirect_id or num))
 
             if redirect_info['status'] == 'error':
                 raise ExtractorError(
@@ -466,8 +518,9 @@ class PBSIE(InfoExtractor):
             else:
                 formats.append({
                     'url': format_url,
-                    'format_id': redirect.get('eeid'),
+                    'format_id': redirect_id,
                 })
+        self._remove_duplicate_formats(formats)
         self._sort_formats(formats)
 
         rating_str = info.get('rating')
@@ -493,7 +546,7 @@ class PBSIE(InfoExtractor):
             'id': video_id,
             'display_id': display_id,
             'title': info['title'],
-            'description': info['program'].get('description'),
+            'description': info.get('description') or info.get('program', {}).get('description'),
             'thumbnail': info.get('image_url'),
             'duration': int_or_none(info.get('duration')),
             'age_limit': age_limit,

youtube_dl/extractor/plays.py

@@ -0,0 +1,51 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import int_or_none
+
+
+class PlaysTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?plays\.tv/video/(?P<id>[0-9a-f]{18})'
+    _TEST = {
+        'url': 'http://plays.tv/video/56af17f56c95335490/when-you-outplay-the-azir-wall',
+        'md5': 'dfeac1198506652b5257a62762cec7bc',
+        'info_dict': {
+            'id': '56af17f56c95335490',
+            'ext': 'mp4',
+            'title': 'When you outplay the Azir wall',
+            'description': 'Posted by Bjergsen',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._og_search_title(webpage)
+        content = self._parse_json(
+            self._search_regex(
+                r'R\.bindContent\(({.+?})\);', webpage,
+                'content'), video_id)['content']
+        mpd_url, sources = re.search(
+            r'(?s)<video[^>]+data-mpd="([^"]+)"[^>]*>(.+?)</video>',
+            content).groups()
+        formats = self._extract_mpd_formats(
+            self._proto_relative_url(mpd_url), video_id, mpd_id='DASH')
+        for format_id, height, format_url in re.findall(r'<source\s+res="((\d+)h?)"\s+src="([^"]+)"', sources):
+            formats.append({
+                'url': self._proto_relative_url(format_url),
+                'format_id': 'http-' + format_id,
+                'height': int_or_none(height),
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'formats': formats,
+        }

youtube_dl/extractor/pornhub.py

@@ -11,6 +11,7 @@ from ..compat import (
 )
 from ..utils import (
     ExtractorError,
+    int_or_none,
     sanitized_Request,
     str_to_int,
 )
@@ -23,13 +24,18 @@ class PornHubIE(InfoExtractor):
     _VALID_URL = r'https?://(?:[a-z]+\.)?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',
+        'md5': '1e19b41231a02eba417839222ac9d58e',
         'info_dict': {
             'id': '648719015',
             'ext': 'mp4',
-            "uploader": "Babes",
-            "title": "Seductive Indian beauty strips down and fingers her pink pussy",
-            "age_limit": 18
+            'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
+            'uploader': 'Babes',
+            'duration': 361,
+            'view_count': int,
+            'like_count': int,
+            'dislike_count': int,
+            'comment_count': int,
+            'age_limit': 18,
         }
     }, {
         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
@@ -67,13 +73,23 @@ class PornHubIE(InfoExtractor):
                 'PornHub said: %s' % error_msg,
                 expected=True, video_id=video_id)
 
-        video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title')
+        flashvars = self._parse_json(
+            self._search_regex(
+                r'var\s+flashv1ars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'),
+            video_id)
+        if flashvars:
+            video_title = flashvars.get('video_title')
+            thumbnail = flashvars.get('image_url')
+            duration = int_or_none(flashvars.get('video_duration'))
+        else:
+            video_title, thumbnail, duration = [None] * 3
+
+        if not video_title:
+            video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title')
+
         video_uploader = self._html_search_regex(
             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
             webpage, 'uploader', fatal=False)
-        thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, 'thumbnail', fatal=False)
-        if thumbnail:
-            thumbnail = compat_urllib_parse_unquote(thumbnail)
 
         view_count = self._extract_count(
             r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
@@ -95,7 +111,7 @@ class PornHubIE(InfoExtractor):
             path = compat_urllib_parse_urlparse(video_url).path
             extension = os.path.splitext(path)[1][1:]
             format = path.split('/')[5].split('_')[:2]
-            format = "-".join(format)
+            format = '-'.join(format)
 
             m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format)
             if m is None:
@@ -120,6 +136,7 @@ class PornHubIE(InfoExtractor):
             'uploader': video_uploader,
             'title': video_title,
             'thumbnail': thumbnail,
+            'duration': duration,
             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
@@ -129,7 +146,31 @@ class PornHubIE(InfoExtractor):
         }
 
 
-class PornHubPlaylistIE(InfoExtractor):
+class PornHubPlaylistBaseIE(InfoExtractor):
+    def _extract_entries(self, webpage):
+        return [
+            self.url_result('http://www.pornhub.com/%s' % video_url, PornHubIE.ie_key())
+            for video_url in set(re.findall(
+                r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"', webpage))
+        ]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        entries = self._extract_entries(webpage)
+
+        playlist = self._parse_json(
+            self._search_regex(
+                r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
+            playlist_id)
+
+        return self.playlist_result(
+            entries, playlist_id, playlist.get('title'), playlist.get('description'))
+
+
+class PornHubPlaylistIE(PornHubPlaylistBaseIE):
     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
     _TESTS = [{
         'url': 'http://www.pornhub.com/playlist/6201671',
@@ -140,21 +181,20 @@ class PornHubPlaylistIE(InfoExtractor):
         'playlist_mincount': 35,
     }]
 
+
+class PornHubUserVideosIE(PornHubPlaylistBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?pornhub\.com/users/(?P<id>[^/]+)/videos'
+    _TESTS = [{
+        'url': 'http://www.pornhub.com/users/rushandlia/videos',
+        'info_dict': {
+            'id': 'rushandlia',
+        },
+        'playlist_mincount': 13,
+    }]
+
     def _real_extract(self, url):
-        playlist_id = self._match_id(url)
+        user_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, playlist_id)
+        webpage = self._download_webpage(url, user_id)
 
-        entries = [
-            self.url_result('http://www.pornhub.com/%s' % video_url, 'PornHub')
-            for video_url in set(re.findall(
-                r'href="/?(view_video\.php\?.*\bviewkey=[\da-z]+[^"]*)"', webpage))
-        ]
-
-        playlist = self._parse_json(
-            self._search_regex(
-                r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
-            playlist_id)
-
-        return self.playlist_result(
-            entries, playlist_id, playlist.get('title'), playlist.get('description'))
+        return self.playlist_result(self._extract_entries(webpage), user_id)

youtube_dl/extractor/pornovoisines.py

@@ -56,7 +56,7 @@ class PornoVoisinesIE(InfoExtractor):
             r'<h1>(.+?)</h1>', webpage, 'title', flags=re.DOTALL)
         description = self._html_search_regex(
             r'<article id="descriptif">(.+?)</article>',
-            webpage, "description", fatal=False, flags=re.DOTALL)
+            webpage, 'description', fatal=False, flags=re.DOTALL)
 
         thumbnail = self._search_regex(
             r'<div id="mediaspace%s">\s*<img src="/?([^"]+)"' % video_id,

youtube_dl/extractor/radiobremen.py

@@ -28,16 +28,16 @@ class RadioBremenIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        meta_url = "http://www.radiobremen.de/apps/php/mediathek/metadaten.php?id=%s" % video_id
+        meta_url = 'http://www.radiobremen.de/apps/php/mediathek/metadaten.php?id=%s' % video_id
         meta_doc = self._download_webpage(
             meta_url, video_id, 'Downloading metadata')
         title = self._html_search_regex(
-            r"<h1.*>(?P<title>.+)</h1>", meta_doc, "title")
+            r'<h1.*>(?P<title>.+)</h1>', meta_doc, 'title')
         description = self._html_search_regex(
-            r"<p>(?P<description>.*)</p>", meta_doc, "description", fatal=False)
+            r'<p>(?P<description>.*)</p>', meta_doc, 'description', fatal=False)
         duration = parse_duration(self._html_search_regex(
-            r"L&auml;nge:</td>\s+<td>(?P<duration>[0-9]+:[0-9]+)</td>",
-            meta_doc, "duration", fatal=False))
+            r'L&auml;nge:</td>\s+<td>(?P<duration>[0-9]+:[0-9]+)</td>',
+            meta_doc, 'duration', fatal=False))
 
         page_doc = self._download_webpage(
             url, video_id, 'Downloading video information')
@@ -51,7 +51,7 @@ class RadioBremenIE(InfoExtractor):
         formats = [{
             'url': video_url,
             'ext': 'mp4',
-            'width': int(mobj.group("width")),
+            'width': int(mobj.group('width')),
         }]
         return {
             'id': video_id,

youtube_dl/extractor/radiofrance.py

@@ -16,9 +16,9 @@ class RadioFranceIE(InfoExtractor):
         'info_dict': {
             'id': 'one-one',
             'ext': 'ogg',
-            "title": "One to one",
-            "description": "Plutôt que d'imaginer la radio de demain comme technologie ou comme création de contenu, je veux montrer que quelles que soient ses évolutions, j'ai l'intime conviction que la radio continuera d'être un grand média de proximité pour les auditeurs.",
-            "uploader": "Thomas Hercouët",
+            'title': 'One to one',
+            'description': "Plutôt que d'imaginer la radio de demain comme technologie ou comme création de contenu, je veux montrer que quelles que soient ses évolutions, j'ai l'intime conviction que la radio continuera d'être un grand média de proximité pour les auditeurs.",
+            'uploader': 'Thomas Hercouët',
         },
     }
 

youtube_dl/extractor/rbmaradio.py

@@ -18,11 +18,11 @@ class RBMARadioIE(InfoExtractor):
         'info_dict': {
             'id': 'ford-lopatin-live-at-primavera-sound-2011',
             'ext': 'mp3',
-            "uploader_id": "ford-lopatin",
-            "location": "Spain",
-            "description": "Joel Ford and Daniel ’Oneohtrix Point Never’ Lopatin fly their midified pop extravaganza to Spain. Live at Primavera Sound 2011.",
-            "uploader": "Ford & Lopatin",
-            "title": "Live at Primavera Sound 2011",
+            'uploader_id': 'ford-lopatin',
+            'location': 'Spain',
+            'description': 'Joel Ford and Daniel ’Oneohtrix Point Never’ Lopatin fly their midified pop extravaganza to Spain. Live at Primavera Sound 2011.',
+            'uploader': 'Ford & Lopatin',
+            'title': 'Live at Primavera Sound 2011',
         },
     }