release 2016.09.03

It's the same video player, the only difference is that the video player
is loaded differently, and certain metadata (title and description) is
not available under dr.dk/mu, so make it by default get that from some
of the html meta tags.

Skip the dr.dk/tv test

dr.dk/tv videos are only available for between 7 and 90 days due to
Danish law, and in certain cases may be readded. Skip this test as it is
no longer available.

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.08.24.1*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.08.24.1**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.09.03*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.09.03**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2016.08.24.1
+[debug] youtube-dl version 2016.09.03
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

AUTHORS

@@ -181,3 +181,5 @@ Nehal Patel
 Rob van Bekkum
 Petr Zvoníček
 Pratyush Singh
+Aleksander Nitecki
+Sebastian Blunt

ChangeLog

@@ -1,3 +1,79 @@
+version 2016.09.03
+
+Core
+* Restore usage of NAME attribute from EXT-X-MEDIA tag for formats codes in
+  _extract_m3u8_formats (#10522)
+* Handle semicolon in mimetype2ext
+
+
+Extractors
++ [youtube] Add support for rental videos' previews (#10532)
+* [youtube:playlist] Fallback to video extraction for video/playlist URLs when
+  no playlist is actually served (#10537)
++ [drtv] Add support for dr.dk/nyheder (#10536)
++ [facebook:plugins:video] Add extractor (#10530)
++ [go] Add extractor for *.go.com sites
+* [adobepass] Check for authz_token expiration (#10527)
+* [nytimes] improve extraction
+* [thestar] Fix extraction (#10465)
+* [glide] Fix extraction (#10478)
+- [exfm] Remove extractor (#10482)
+* [youporn] Fix categories and tags extraction (#10521)
++ [curiositystream] Add extractor for app.curiositystream.com
+- [thvideo] Remove extractor (#10464)
+* [movingimage] Fix for the new site name (#10466)
++ [cbs] Add support for once formats (#10515)
+* [limelight] Skip ism snd duplicate manifests
++ [porncom] Extract categories and tags (#10510)
++ [facebook] Extract timestamp (#10508)
++ [yahoo] Extract more formats
+
+
+version 2016.08.31
+
+Extractors
+* [soundcloud] Fix URL regular expression to avoid clashes with sets (#10505)
+* [bandcamp:album] Fix title extraction (#10455)
+* [pyvideo] Fix extraction (#10468)
++ [ctv] Add support for tsn.ca, bnn.ca and thecomedynetwork.ca (#10016)
+* [9c9media] Extract more metadata
+* [9c9media] Fix multiple stacks extraction (#10016)
+* [adultswim] Improve video info extraction (#10492)
+* [vodplatform] Improve embed regular expression
+- [played] Remove extractor (#10470)
++ [tbs] Add extractor for tbs.com and tntdrama.com (#10222)
++ [cartoonnetwork] Add extractor for cartoonnetwork.com (#10110)
+* [adultswim] Rework in terms of turner extractor
+* [cnn] Rework in terms of turner extractor
+* [nba] Rework in terms of turner extractor
++ [turner] Add base extractor for Turner Broadcasting System based sites
+* [bilibili] Fix extraction (#10375)
+* [openload] Fix extraction (#10408)
+
+
+version 2016.08.28
+
+Core
++ Add warning message that ffmpeg doesn't support SOCKS
+* Improve thumbnail sorting
++ Extract formats from #EXT-X-MEDIA tags in _extract_m3u8_formats
+* Fill IV with leading zeros for IVs shorter than 16 octets in hlsnative
++ Add ac-3 to the list of audio codecs in parse_codecs
+
+Extractors
+* [periscope:user] Fix extraction (#10453)
+* [douyutv] Fix extraction (#10153, #10318, #10444)
++ [nhk:vod] Add extractor for www3.nhk.or.jp on demand (#4437, #10424)
+- [trutube] Remove extractor (#10438)
++ [usanetwork] Add extractor for usanetwork.com
+* [crackle] Fix extraction (#10333)
+* [spankbang] Fix description and uploader extraction (#10339)
+* [discoverygo] Detect cable provider restricted videos (#10425)
++ [cbc] Add support for watch.cbc.ca
+* [kickstarter] Silent the warning for og:description (#10415)
+* [mtvservices:embedded] Fix extraction for the new 'edge' player (#10363)
+
+
 version 2016.08.24.1
 
 Extractors

README.md

@@ -412,11 +412,19 @@ You can configure youtube-dl by placing any supported command line option to a c
 
 For example, with the following configuration file youtube-dl will always extract the audio, not copy the mtime, use a proxy and save all videos under `Movies` directory in your home directory:
 ```
--x
---no-mtime
---proxy 127.0.0.1:3128
--o ~/Movies/%(title)s.%(ext)s
 # Lines starting with # are comments
+
+# Always extract audio
+-x
+
+# Do not copy the mtime
+--no-mtime
+
+# Use this proxy
+--proxy 127.0.0.1:3128
+
+# Save all videos under Movies directory in your home directory
+-o ~/Movies/%(title)s.%(ext)s
 ```
 
 Note that options in configuration file are just the same options aka switches used in regular command line calls thus there **must be no whitespace** after `-` or `--`, e.g. `-o` or `--proxy` but not `- o` or `-- proxy`.
@@ -730,7 +738,7 @@ Videos or video formats streamed via RTMP protocol can only be downloaded when [
 
 ### I have downloaded a video but how can I play it?
 
-Once the video is fully downloaded, use any video player, such as [mpv](https://mpv.io/), [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/).
+Once the video is fully downloaded, use any video player, such as [mpv](https://mpv.io/), [vlc](http://www.videolan.org/) or [mplayer](http://www.mplayerhq.hu/).
 
 ### I extracted a video URL with `-g`, but it does not play on another machine / in my webbrowser.
 
@@ -812,10 +820,32 @@ Either prepend `http://www.youtube.com/watch?v=` or separate the ID from the opt
 
 ### How do I pass cookies to youtube-dl?
 
-Use the `--cookies` option, for example `--cookies /path/to/cookies/file.txt`. Note that the cookies file must be in Mozilla/Netscape format and the first line of the cookies file must be either `# HTTP Cookie File` or `# Netscape HTTP Cookie File`. Make sure you have correct [newline format](https://en.wikipedia.org/wiki/Newline) in the cookies file and convert newlines if necessary to correspond with your OS, namely `CRLF` (`\r\n`) for Windows, `LF` (`\n`) for Linux and `CR` (`\r`) for Mac OS. `HTTP Error 400: Bad Request` when using `--cookies` is a good sign of invalid newline format.
+Use the `--cookies` option, for example `--cookies /path/to/cookies/file.txt`.
+
+In order to extract cookies from browser use any conforming browser extension for exporting cookies. For example, [cookies.txt](https://chrome.google.com/webstore/detail/cookiestxt/njabckikapfpffapmjgojcnbfjonfjfg) (for Chrome) or [Export Cookies](https://addons.mozilla.org/en-US/firefox/addon/export-cookies/) (for Firefox).
+
+Note that the cookies file must be in Mozilla/Netscape format and the first line of the cookies file must be either `# HTTP Cookie File` or `# Netscape HTTP Cookie File`. Make sure you have correct [newline format](https://en.wikipedia.org/wiki/Newline) in the cookies file and convert newlines if necessary to correspond with your OS, namely `CRLF` (`\r\n`) for Windows, `LF` (`\n`) for Linux and `CR` (`\r`) for Mac OS. `HTTP Error 400: Bad Request` when using `--cookies` is a good sign of invalid newline format.
 
 Passing cookies to youtube-dl is a good way to workaround login when a particular extractor does not implement it explicitly. Another use case is working around [CAPTCHA](https://en.wikipedia.org/wiki/CAPTCHA) some websites require you to solve in particular cases in order to get access (e.g. YouTube, CloudFlare).
 
+### How do I stream directly to media player?
+
+You will first need to tell youtube-dl to stream media to stdout with `-o -`, and also tell your media player to read from stdin (it must be capable of this for streaming) and then pipe former to latter. For example, streaming to [vlc](http://www.videolan.org/) can be achieved with:
+
+    youtube-dl -o - "http://www.youtube.com/watch?v=BaW_jenozKcj" | vlc -
+
+### How do I download only new videos from a playlist?
+
+Use download-archive feature. With this feature you should initially download the complete playlist with `--download-archive /path/to/download/archive/file.txt` that will record identifiers of all the videos in a special file. Each subsequent run with the same `--download-archive` will download only new videos and skip all videos that have been downloaded before. Note that only successful downloads are recorded in the file.
+
+For example, at first,
+
+    youtube-dl --download-archive archive.txt "https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re"
+
+will download the complete `PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re` playlist and create a file `archive.txt`. Each subsequent run will only download new videos if any:
+
+    youtube-dl --download-archive archive.txt "https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re"
+
 ### Can you add support for this anime video site, or site which shows current movies for free?
 
 As a matter of policy (as well as legality), youtube-dl does not include support for services that specialize in infringing copyright. As a rule of thumb, if you cannot easily find a video that the service is quite obviously allowed to distribute (i.e. that has been uploaded by the creator, the creator's distributor, or is published under a free license), the service is probably unfit for inclusion to youtube-dl.

docs/supportedsites.md

@@ -13,6 +13,8 @@
  - **5min**
  - **8tracks**
  - **91porn**
+ - **9c9media**
+ - **9c9media:stack**
  - **9gag**
  - **9now.com.au**
  - **abc.net.au**
@@ -115,8 +117,11 @@
  - **Canvas**
  - **CarambaTV**
  - **CarambaTVPage**
- - **CBC**
- - **CBCPlayer**
+ - **CartoonNetwork**
+ - **cbc.ca**
+ - **cbc.ca:player**
+ - **cbc.ca:watch**
+ - **cbc.ca:watch:video**
  - **CBS**
  - **CBSInteractive**
  - **CBSLocal**
@@ -166,6 +171,8 @@
  - **CTVNews**
  - **culturebox.francetvinfo.fr**
  - **CultureUnplugged**
+ - **curiositystream**
+ - **curiositystream:collection**
  - **CWTV**
  - **DailyMail**
  - **dailymotion**
@@ -218,11 +225,11 @@
  - **EsriVideo**
  - **Europa**
  - **EveryonesMixtape**
- - **exfm**: ex.fm
  - **ExpoTV**
  - **ExtremeTube**
  - **EyedoTV**
  - **facebook**
+ - **FacebookPluginsVideo**
  - **faz.net**
  - **fc2**
  - **Fczenit**
@@ -266,6 +273,7 @@
  - **Glide**: Glide mobile video messages (glide.me)
  - **Globo**
  - **GloboArticle**
+ - **Go**
  - **GodTube**
  - **GodTV**
  - **Golem**
@@ -401,6 +409,7 @@
  - **MovieClips**
  - **MovieFap**
  - **Moviezine**
+ - **MovingImage**
  - **MPORA**
  - **MSN**
  - **mtg**: MTG services
@@ -448,6 +457,7 @@
  - **NextMediaActionNews**: 蘋果日報 - 動新聞
  - **nfb**: National Film Board of Canada
  - **nfl.com**
+ - **NhkVod**
  - **nhl.com**
  - **nhl.com:news**: NHL news
  - **nhl.com:videocenter**
@@ -456,7 +466,6 @@
  - **nick.de**
  - **niconico**: ニコニコ動画
  - **NiconicoPlaylist**
- - **NineCNineMedia**
  - **Nintendo**
  - **njoy**: N-JOY
  - **njoy:embed**
@@ -514,7 +523,6 @@
  - **Pinkbike**
  - **Pladform**
  - **play.fm**
- - **played.to**
  - **PlaysTV**
  - **Playtvak**: Playtvak.cz, iDNES.cz and Lidovky.cz
  - **Playvid**
@@ -655,7 +663,6 @@
  - **sr:mediathek**: Saarländischer Rundfunk
  - **SRGSSR**
  - **SRGSSRPlay**: srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites
- - **SSA**
  - **stanfordoc**: Stanford Open ClassRoom
  - **Steam**
  - **Stitcher**
@@ -672,6 +679,7 @@
  - **Tagesschau**
  - **tagesschau:player**
  - **Tass**
+ - **TBS**
  - **TDSLifeway**
  - **teachertube**: teachertube.com videos
  - **teachertube:user:collection**: teachertube.com user and collection videos
@@ -697,8 +705,6 @@
  - **TheStar**
  - **ThisAmericanLife**
  - **ThisAV**
- - **THVideo**
- - **THVideoPlaylist**
  - **tinypic**: tinypic.com videos
  - **tlc.de**
  - **TMZ**
@@ -713,7 +719,6 @@
  - **TrailerAddict** (Currently broken)
  - **Trilulilu**
  - **trollvids**
- - **TruTube**
  - **Tube8**
  - **TubiTv**
  - **tudou**
@@ -758,6 +763,7 @@
  - **uplynk:preplay**
  - **Urort**: NRK P3 Urørt
  - **URPlay**
+ - **USANetwork**
  - **USAToday**
  - **ustream**
  - **ustream:channel**

test/test_utils.py

@@ -39,6 +39,7 @@ from youtube_dl.utils import (
     is_html,
     js_to_json,
     limit_length,
+    mimetype2ext,
     ohdave_rsa_encrypt,
     OnDemandPagedList,
     orderedSet,
@@ -625,6 +626,14 @@ class TestUtil(unittest.TestCase):
             limit_length('foo bar baz asd', 12).startswith('foo bar'))
         self.assertTrue('...' in limit_length('foo bar baz asd', 12))
 
+    def test_mimetype2ext(self):
+        self.assertEqual(mimetype2ext(None), None)
+        self.assertEqual(mimetype2ext('video/x-flv'), 'flv')
+        self.assertEqual(mimetype2ext('application/x-mpegURL'), 'm3u8')
+        self.assertEqual(mimetype2ext('text/vtt'), 'vtt')
+        self.assertEqual(mimetype2ext('text/vtt;charset=utf-8'), 'vtt')
+        self.assertEqual(mimetype2ext('text/html; charset=utf-8'), 'html')
+
     def test_parse_codecs(self):
         self.assertEqual(parse_codecs(''), {})
         self.assertEqual(parse_codecs('avc1.77.30, mp4a.40.2'), {

youtube_dl/YoutubeDL.py

@@ -1256,8 +1256,10 @@ class YoutubeDL(object):
                 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
         if thumbnails:
             thumbnails.sort(key=lambda t: (
-                t.get('preference'), t.get('width'), t.get('height'),
-                t.get('id'), t.get('url')))
+                t.get('preference') if t.get('preference') is not None else -1,
+                t.get('width') if t.get('width') is not None else -1,
+                t.get('height') if t.get('height') is not None else -1,
+                t.get('id') if t.get('id') is not None else '', t.get('url')))
             for i, t in enumerate(thumbnails):
                 t['url'] = sanitize_url(t['url'])
                 if t.get('width') and t.get('height'):

youtube_dl/downloader/external.py

@@ -220,6 +220,12 @@ class FFmpegFD(ExternalFD):
         if proxy:
             if not re.match(r'^[\da-zA-Z]+://', proxy):
                 proxy = 'http://%s' % proxy
+
+            if proxy.startswith('socks'):
+                self.report_warning(
+                    '%s does not support SOCKS proxies. Downloading is likely to fail. '
+                    'Consider adding --hls-prefer-native to your command.' % self.get_basename())
+
             # Since December 2015 ffmpeg supports -http_proxy option (see
             # http://git.videolan.org/?p=ffmpeg.git;a=commit;h=b4eb1f29ebddd60c41a2eb39f5af701e38e0d3fd)
             # We could switch to the following code if we are able to detect version properly

youtube_dl/downloader/hls.py

@@ -83,7 +83,10 @@ class HlsFD(FragmentFD):
 
         self._prepare_and_start_frag_download(ctx)
 
+        extra_query = None
         extra_param_to_segment_url = info_dict.get('extra_param_to_segment_url')
+        if extra_param_to_segment_url:
+            extra_query = compat_urlparse.parse_qs(extra_param_to_segment_url)
         i = 0
         media_sequence = 0
         decrypt_info = {'METHOD': 'NONE'}
@@ -97,8 +100,8 @@ class HlsFD(FragmentFD):
                         if re.match(r'^https?://', line)
                         else compat_urlparse.urljoin(man_url, line))
                     frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
-                    if extra_param_to_segment_url:
-                        frag_url = update_url_query(frag_url, extra_param_to_segment_url)
+                    if extra_query:
+                        frag_url = update_url_query(frag_url, extra_query)
                     success = ctx['dl'].download(frag_filename, {'url': frag_url})
                     if not success:
                         return False
@@ -120,12 +123,12 @@ class HlsFD(FragmentFD):
                     decrypt_info = parse_m3u8_attributes(line[11:])
                     if decrypt_info['METHOD'] == 'AES-128':
                         if 'IV' in decrypt_info:
-                            decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:])
+                            decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:].zfill(32))
                         if not re.match(r'^https?://', decrypt_info['URI']):
                             decrypt_info['URI'] = compat_urlparse.urljoin(
                                 man_url, decrypt_info['URI'])
-                        if extra_param_to_segment_url:
-                            decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_param_to_segment_url)
+                        if extra_query:
+                            decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_query)
                         decrypt_info['KEY'] = self.ydl.urlopen(decrypt_info['URI']).read()
                 elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):
                     media_sequence = int(line[22:])

youtube_dl/extractor/adobepass.py

@@ -37,6 +37,10 @@ class AdobePassIE(InfoExtractor):
             return self._search_regex(
                 '<%s>(.+?)</%s>' % (tag, tag), xml_str, tag)
 
+        def is_expired(token, date_ele):
+            token_expires = unified_timestamp(re.sub(r'[_ ]GMT', '', xml_text(token, date_ele)))
+            return token_expires and token_expires <= int(time.time())
+
         mvpd_headers = {
             'ap_42': 'anonymous',
             'ap_11': 'Linux i686',
@@ -47,11 +51,8 @@ class AdobePassIE(InfoExtractor):
         guid = xml_text(resource, 'guid')
         requestor_info = self._downloader.cache.load('mvpd', requestor_id) or {}
         authn_token = requestor_info.get('authn_token')
-        if authn_token:
-            token_expires = unified_timestamp(re.sub(r'[_ ]GMT', '', xml_text(authn_token, 'simpleTokenExpires')))
-            if token_expires and token_expires <= int(time.time()):
-                authn_token = None
-                requestor_info = {}
+        if authn_token and is_expired(authn_token, 'simpleTokenExpires'):
+            authn_token = None
         if not authn_token:
             # TODO add support for other TV Providers
             mso_id = 'DTV'
@@ -98,6 +99,8 @@ class AdobePassIE(InfoExtractor):
             self._downloader.cache.store('mvpd', requestor_id, requestor_info)
 
         authz_token = requestor_info.get(guid)
+        if authz_token and is_expired(authz_token, 'simpleTokenTTL'):
+            authz_token = None
         if not authz_token:
             authorize = self._download_webpage(
                 self._SERVICE_PROVIDER_TEMPLATE % 'authorize', video_id,

youtube_dl/extractor/adultswim.py

@@ -3,16 +3,14 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .turner import TurnerBaseIE
 from ..utils import (
-    determine_ext,
     ExtractorError,
-    float_or_none,
-    xpath_text,
+    int_or_none,
 )
 
 
-class AdultSwimIE(InfoExtractor):
+class AdultSwimIE(TurnerBaseIE):
     _VALID_URL = r'https?://(?:www\.)?adultswim\.com/videos/(?P<is_playlist>playlists/)?(?P<show_path>[^/]+)/(?P<episode_path>[^/?#]+)/?'
 
     _TESTS = [{
@@ -96,7 +94,8 @@ class AdultSwimIE(InfoExtractor):
         'params': {
             # m3u8 download
             'skip_download': True,
-        }
+        },
+        'expected_warnings': ['Unable to download f4m manifest'],
     }]
 
     @staticmethod
@@ -148,7 +147,10 @@ class AdultSwimIE(InfoExtractor):
                 if bootstrapped_data.get('slugged_video', {}).get('slug') == episode_path:
                     video_info = bootstrapped_data['slugged_video']
             if not video_info:
-                video_info = bootstrapped_data.get('heroMetadata', {}).get('trailer').get('video')
+                video_info = bootstrapped_data.get(
+                    'heroMetadata', {}).get('trailer', {}).get('video')
+            if not video_info:
+                video_info = bootstrapped_data.get('onlineOriginals', [None])[0]
             if not video_info:
                 raise ExtractorError('Unable to find video info')
 
@@ -162,70 +164,38 @@ class AdultSwimIE(InfoExtractor):
             elif video_info.get('videoPlaybackID'):
                 segment_ids = [video_info['videoPlaybackID']]
             else:
-                raise ExtractorError(
-                    'This video is only available via cable service provider subscription that'
-                    ' is not currently supported. You may want to use --cookies.'
-                    if video_info.get('auth') is True else 'Unable to find stream or clips',
-                    expected=True)
+                if video_info.get('auth') is True:
+                    raise ExtractorError(
+                        'This video is only available via cable service provider subscription that'
+                        ' is not currently supported. You may want to use --cookies.', expected=True)
+                else:
+                    raise ExtractorError('Unable to find stream or clips')
 
         episode_id = video_info['id']
         episode_title = video_info['title']
-        episode_description = video_info['description']
-        episode_duration = video_info.get('duration')
+        episode_description = video_info.get('description')
+        episode_duration = int_or_none(video_info.get('duration'))
+        view_count = int_or_none(video_info.get('views'))
 
         entries = []
         for part_num, segment_id in enumerate(segment_ids):
-            segment_url = 'http://www.adultswim.com/videos/api/v0/assets?id=%s&platform=desktop' % segment_id
-
+            segement_info = self._extract_cvp_info(
+                'http://www.adultswim.com/videos/api/v0/assets?id=%s&platform=desktop' % segment_id,
+                segment_id, {
+                    'secure': {
+                        'media_src': 'http://androidhls-secure.cdn.turner.com/adultswim/big',
+                        'tokenizer_src': 'http://www.adultswim.com/astv/mvpd/processors/services/token_ipadAdobe.do',
+                    },
+                })
             segment_title = '%s - %s' % (show_title, episode_title)
             if len(segment_ids) > 1:
                 segment_title += ' Part %d' % (part_num + 1)
-
-            idoc = self._download_xml(
-                segment_url, segment_title,
-                'Downloading segment information', 'Unable to download segment information')
-
-            segment_duration = float_or_none(
-                xpath_text(idoc, './/trt', 'segment duration').strip())
-
-            formats = []
-            file_els = idoc.findall('.//files/file') or idoc.findall('./files/file')
-
-            unique_urls = []
-            unique_file_els = []
-            for file_el in file_els:
-                media_url = file_el.text
-                if not media_url or determine_ext(media_url) == 'f4m':
-                    continue
-                if file_el.text not in unique_urls:
-                    unique_urls.append(file_el.text)
-                    unique_file_els.append(file_el)
-
-            for file_el in unique_file_els:
-                bitrate = file_el.attrib.get('bitrate')
-                ftype = file_el.attrib.get('type')
-                media_url = file_el.text
-                if determine_ext(media_url) == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
-                        media_url, segment_title, 'mp4', preference=0,
-                        m3u8_id='hls', fatal=False))
-                else:
-                    formats.append({
-                        'format_id': '%s_%s' % (bitrate, ftype),
-                        'url': file_el.text.strip(),
-                        # The bitrate may not be a number (for example: 'iphone')
-                        'tbr': int(bitrate) if bitrate.isdigit() else None,
-                    })
-
-            self._sort_formats(formats)
-
-            entries.append({
+            segement_info.update({
                 'id': segment_id,
                 'title': segment_title,
-                'formats': formats,
-                'duration': segment_duration,
-                'description': episode_description
+                'description': episode_description,
             })
+            entries.append(segement_info)
 
         return {
             '_type': 'playlist',
@@ -234,5 +204,6 @@ class AdultSwimIE(InfoExtractor):
             'entries': entries,
             'title': '%s - %s' % (show_title, episode_title),
             'description': episode_description,
-            'duration': episode_duration
+            'duration': episode_duration,
+            'view_count': view_count,
         }

youtube_dl/extractor/bandcamp.py

@@ -162,6 +162,15 @@ class BandcampAlbumIE(InfoExtractor):
             'uploader_id': 'dotscale',
         },
         'playlist_mincount': 7,
+    }, {
+        # with escaped quote in title
+        'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
+        'info_dict': {
+            'title': '"Entropy" EP',
+            'uploader_id': 'jstrecords',
+            'id': 'entropy-ep',
+        },
+        'playlist_mincount': 3,
     }]
 
     def _real_extract(self, url):
@@ -176,8 +185,11 @@ class BandcampAlbumIE(InfoExtractor):
         entries = [
             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
             for t_path in tracks_paths]
-        title = self._search_regex(
-            r'album_title\s*:\s*"(.*?)"', webpage, 'title', fatal=False)
+        title = self._html_search_regex(
+            r'album_title\s*:\s*"((?:\\.|[^"\\])+?)"',
+            webpage, 'title', fatal=False)
+        if title:
+            title = title.replace(r'\"', '"')
         return {
             '_type': 'playlist',
             'uploader_id': uploader_id,

youtube_dl/extractor/bet.py

@@ -2,7 +2,6 @@ from __future__ import unicode_literals
 
 from .mtv import MTVServicesInfoExtractor
 from ..utils import unified_strdate
-from ..compat import compat_urllib_parse_urlencode
 
 
 class BetIE(MTVServicesInfoExtractor):
@@ -53,9 +52,9 @@ class BetIE(MTVServicesInfoExtractor):
     _FEED_URL = "http://feeds.mtvnservices.com/od/feed/bet-mrss-player"
 
     def _get_feed_query(self, uri):
-        return compat_urllib_parse_urlencode({
+        return {
             'uuid': uri,
-        })
+        }
 
     def _extract_mgid(self, webpage):
         return self._search_regex(r'data-uri="([^"]+)', webpage, 'mgid')

youtube_dl/extractor/bilibili.py

@@ -1,22 +1,15 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import calendar
-import datetime
+import hashlib
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_etree_fromstring,
-    compat_str,
-    compat_parse_qs,
-    compat_xml_parse_error,
-)
+from ..compat import compat_parse_qs
 from ..utils import (
-    ExtractorError,
     int_or_none,
     float_or_none,
-    xpath_text,
+    unified_timestamp,
 )
 
 
@@ -27,7 +20,7 @@ class BiliBiliIE(InfoExtractor):
         'url': 'http://www.bilibili.tv/video/av1074402/',
         'md5': '9fa226fe2b8a9a4d5a69b4c6a183417e',
         'info_dict': {
-            'id': '1554319',
+            'id': '1074402',
             'ext': 'mp4',
             'title': '【金坷垃】金泡沫',
             'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
@@ -41,24 +34,28 @@ class BiliBiliIE(InfoExtractor):
     }, {
         'url': 'http://www.bilibili.com/video/av1041170/',
         'info_dict': {
-            'id': '1507019',
+            'id': '1041170',
             'ext': 'mp4',
             'title': '【BD1080P】刀语【诸神&异域】',
             'description': '这是个神奇的故事~每个人不留弹幕不给走哦~切利哦！~',
+            'duration': 3382.259,
             'timestamp': 1396530060,
             'upload_date': '20140403',
+            'thumbnail': 're:^https?://.+\.jpg',
             'uploader': '枫叶逝去',
             'uploader_id': '520116',
         },
     }, {
         'url': 'http://www.bilibili.com/video/av4808130/',
         'info_dict': {
-            'id': '7802182',
+            'id': '4808130',
             'ext': 'mp4',
             'title': '【长篇】哆啦A梦443【钉铛】',
             'description': '(2016.05.27)来组合客人的脸吧&寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&illust_id=56912929',
+            'duration': 1493.995,
             'timestamp': 1464564180,
             'upload_date': '20160529',
+            'thumbnail': 're:^https?://.+\.jpg',
             'uploader': '喜欢拉面',
             'uploader_id': '151066',
         },
@@ -66,12 +63,14 @@ class BiliBiliIE(InfoExtractor):
         # Missing upload time
         'url': 'http://www.bilibili.com/video/av1867637/',
         'info_dict': {
-            'id': '2880301',
+            'id': '1867637',
             'ext': 'mp4',
             'title': '【HDTV】【喜剧】岳父岳母真难当 （2014）【法国票房冠军】',
             'description': '一个信奉天主教的法国旧式传统资产阶级家庭中有四个女儿。三个女儿却分别找了阿拉伯、犹太、中国丈夫，老夫老妻唯独期盼剩下未嫁的小女儿能找一个信奉天主教的法国白人，结果没想到小女儿找了一位非裔黑人……【这次应该不会跳帧了】',
+            'duration': 5760.0,
             'uploader': '黑夜为猫',
             'uploader_id': '610729',
+            'thumbnail': 're:^https?://.+\.jpg',
         },
         'params': {
             # Just to test metadata extraction
@@ -80,86 +79,61 @@ class BiliBiliIE(InfoExtractor):
         'expected_warnings': ['upload time'],
     }]
 
-    # BiliBili blocks keys from time to time. The current key is extracted from
-    # the Android client
-    # TODO: find the sign algorithm used in the flash player
-    _APP_KEY = '86385cdc024c0f6c'
+    _APP_KEY = '6f90a59ac58a4123'
+    _BILIBILI_KEY = '0bfd84cc3940035173f35e6777508326'
 
     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)
 
-        params = compat_parse_qs(self._search_regex(
+        cid = compat_parse_qs(self._search_regex(
             [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)',
              r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'],
-            webpage, 'player parameters'))
-        cid = params['cid'][0]
+            webpage, 'player parameters'))['cid'][0]
 
-        info_xml_str = self._download_webpage(
-            'http://interface.bilibili.com/v_cdn_play',
-            cid, query={'appkey': self._APP_KEY, 'cid': cid},
-            note='Downloading video info page')
+        payload = 'appkey=%s&cid=%s&otype=json&quality=2&type=mp4' % (self._APP_KEY, cid)
+        sign = hashlib.md5((payload + self._BILIBILI_KEY).encode('utf-8')).hexdigest()
 
-        err_msg = None
-        durls = None
-        info_xml = None
-        try:
-            info_xml = compat_etree_fromstring(info_xml_str.encode('utf-8'))
-        except compat_xml_parse_error:
-            info_json = self._parse_json(info_xml_str, video_id, fatal=False)
-            err_msg = (info_json or {}).get('error_text')
-        else:
-            err_msg = xpath_text(info_xml, './message')
-
-        if info_xml is not None:
-            durls = info_xml.findall('./durl')
-        if not durls:
-            if err_msg:
-                raise ExtractorError('%s said: %s' % (self.IE_NAME, err_msg), expected=True)
-            else:
-                raise ExtractorError('No videos found!')
+        video_info = self._download_json(
+            'http://interface.bilibili.com/playurl?%s&sign=%s' % (payload, sign),
+            video_id, note='Downloading video info page')
 
         entries = []
 
-        for durl in durls:
-            size = xpath_text(durl, ['./filesize', './size'])
+        for idx, durl in enumerate(video_info['durl']):
             formats = [{
-                'url': durl.find('./url').text,
-                'filesize': int_or_none(size),
+                'url': durl['url'],
+                'filesize': int_or_none(durl['size']),
             }]
-            for backup_url in durl.findall('./backup_url/url'):
+            for backup_url in durl['backup_url']:
                 formats.append({
-                    'url': backup_url.text,
+                    'url': backup_url,
                     # backup URLs have lower priorities
-                    'preference': -2 if 'hd.mp4' in backup_url.text else -3,
+                    'preference': -2 if 'hd.mp4' in backup_url else -3,
                 })
 
             self._sort_formats(formats)
 
             entries.append({
-                'id': '%s_part%s' % (cid, xpath_text(durl, './order')),
-                'duration': int_or_none(xpath_text(durl, './length'), 1000),
+                'id': '%s_part%s' % (video_id, idx),
+                'duration': float_or_none(durl.get('length'), 1000),
                 'formats': formats,
             })
 
         title = self._html_search_regex('<h1[^>]+title="([^"]+)">', webpage, 'title')
         description = self._html_search_meta('description', webpage)
-        datetime_str = self._html_search_regex(
-            r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', fatal=False)
-        timestamp = None
-        if datetime_str:
-            timestamp = calendar.timegm(datetime.datetime.strptime(datetime_str, '%Y-%m-%dT%H:%M').timetuple())
+        timestamp = unified_timestamp(self._html_search_regex(
+            r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', fatal=False))
 
         # TODO 'view_count' requires deobfuscating Javascript
         info = {
-            'id': compat_str(cid),
+            'id': video_id,
             'title': title,
             'description': description,
             'timestamp': timestamp,
             'thumbnail': self._html_search_meta('thumbnailUrl', webpage),
-            'duration': float_or_none(xpath_text(info_xml, './timelength'), scale=1000),
+            'duration': float_or_none(video_info.get('timelength'), scale=1000),
         }
 
         uploader_mobj = re.search(

youtube_dl/extractor/cartoonnetwork.py

@@ -0,0 +1,36 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .turner import TurnerBaseIE
+
+
+class CartoonNetworkIE(TurnerBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?cartoonnetwork\.com/video/(?:[^/]+/)+(?P<id>[^/?#]+)-(?:clip|episode)\.html'
+    _TEST = {
+        'url': 'http://www.cartoonnetwork.com/video/teen-titans-go/starfire-the-cat-lady-clip.html',
+        'info_dict': {
+            'id': '8a250ab04ed07e6c014ef3f1e2f9016c',
+            'ext': 'mp4',
+            'title': 'Starfire the Cat Lady',
+            'description': 'Robin decides to become a cat so that Starfire will finally love him.',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        id_type, video_id = re.search(r"_cnglobal\.cvp(Video|Title)Id\s*=\s*'([^']+)';", webpage).groups()
+        query = ('id' if id_type == 'Video' else 'titleId') + '=' + video_id
+        return self._extract_cvp_info(
+            'http://www.cartoonnetwork.com/video-seo-svc/episodeservices/getCvpPlaylist?networkName=CN2&' + query, video_id, {
+                'secure': {
+                    'media_src': 'http://apple-secure.cdn.turner.com/toon/big',
+                    'tokenizer_src': 'http://www.cartoonnetwork.com/cntv/mvpd/processors/services/token_ipadAdobe.do',
+                },
+            })

youtube_dl/extractor/cbc.py

@@ -9,10 +9,19 @@ from ..utils import (
     js_to_json,
     smuggle_url,
     try_get,
+    xpath_text,
+    xpath_element,
+    xpath_with_ns,
+    find_xpath_attr,
+    parse_iso8601,
+    parse_age_limit,
+    int_or_none,
+    ExtractorError,
 )
 
 
 class CBCIE(InfoExtractor):
+    IE_NAME = 'cbc.ca'
     _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?!player/)(?:[^/]+/)+(?P<id>[^/?#]+)'
     _TESTS = [{
         # with mediaId
@@ -114,6 +123,7 @@ class CBCIE(InfoExtractor):
 
 
 class CBCPlayerIE(InfoExtractor):
+    IE_NAME = 'cbc.ca:player'
     _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P<id>\d+)'
     _TESTS = [{
         'url': 'http://www.cbc.ca/player/play/2683190193',
@@ -167,3 +177,165 @@ class CBCPlayerIE(InfoExtractor):
                 }),
             'id': video_id,
         }
+
+
+class CBCWatchBaseIE(InfoExtractor):
+    _device_id = None
+    _device_token = None
+    _API_BASE_URL = 'https://api-cbc.cloud.clearleap.com/cloffice/client/'
+    _NS_MAP = {
+        'media': 'http://search.yahoo.com/mrss/',
+        'clearleap': 'http://www.clearleap.com/namespace/clearleap/1.0/',
+    }
+
+    def _call_api(self, path, video_id):
+        url = path if path.startswith('http') else self._API_BASE_URL + path
+        result = self._download_xml(url, video_id, headers={
+            'X-Clearleap-DeviceId': self._device_id,
+            'X-Clearleap-DeviceToken': self._device_token,
+        })
+        error_message = xpath_text(result, 'userMessage') or xpath_text(result, 'systemMessage')
+        if error_message:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message))
+        return result
+
+    def _real_initialize(self):
+        if not self._device_id or not self._device_token:
+            device = self._downloader.cache.load('cbcwatch', 'device') or {}
+            self._device_id, self._device_token = device.get('id'), device.get('token')
+            if not self._device_id or not self._device_token:
+                result = self._download_xml(
+                    self._API_BASE_URL + 'device/register',
+                    None, data=b'<device><type>web</type></device>')
+                self._device_id = xpath_text(result, 'deviceId', fatal=True)
+                self._device_token = xpath_text(result, 'deviceToken', fatal=True)
+                self._downloader.cache.store(
+                    'cbcwatch', 'device', {
+                        'id': self._device_id,
+                        'token': self._device_token,
+                    })
+
+    def _parse_rss_feed(self, rss):
+        channel = xpath_element(rss, 'channel', fatal=True)
+
+        def _add_ns(path):
+            return xpath_with_ns(path, self._NS_MAP)
+
+        entries = []
+        for item in channel.findall('item'):
+            guid = xpath_text(item, 'guid', fatal=True)
+            title = xpath_text(item, 'title', fatal=True)
+
+            media_group = xpath_element(item, _add_ns('media:group'), fatal=True)
+            content = xpath_element(media_group, _add_ns('media:content'), fatal=True)
+            content_url = content.attrib['url']
+
+            thumbnails = []
+            for thumbnail in media_group.findall(_add_ns('media:thumbnail')):
+                thumbnail_url = thumbnail.get('url')
+                if not thumbnail_url:
+                    continue
+                thumbnails.append({
+                    'id': thumbnail.get('profile'),
+                    'url': thumbnail_url,
+                    'width': int_or_none(thumbnail.get('width')),
+                    'height': int_or_none(thumbnail.get('height')),
+                })
+
+            timestamp = None
+            release_date = find_xpath_attr(
+                item, _add_ns('media:credit'), 'role', 'releaseDate')
+            if release_date is not None:
+                timestamp = parse_iso8601(release_date.text)
+
+            entries.append({
+                '_type': 'url_transparent',
+                'url': content_url,
+                'id': guid,
+                'title': title,
+                'description': xpath_text(item, 'description'),
+                'timestamp': timestamp,
+                'duration': int_or_none(content.get('duration')),
+                'age_limit': parse_age_limit(xpath_text(item, _add_ns('media:rating'))),
+                'episode': xpath_text(item, _add_ns('clearleap:episode')),
+                'episode_number': int_or_none(xpath_text(item, _add_ns('clearleap:episodeInSeason'))),
+                'series': xpath_text(item, _add_ns('clearleap:series')),
+                'season_number': int_or_none(xpath_text(item, _add_ns('clearleap:season'))),
+                'thumbnails': thumbnails,
+                'ie_key': 'CBCWatchVideo',
+            })
+
+        return self.playlist_result(
+            entries, xpath_text(channel, 'guid'),
+            xpath_text(channel, 'title'),
+            xpath_text(channel, 'description'))
+
+
+class CBCWatchVideoIE(CBCWatchBaseIE):
+    IE_NAME = 'cbc.ca:watch:video'
+    _VALID_URL = r'https?://api-cbc\.cloud\.clearleap\.com/cloffice/client/web/play/?\?.*?\bcontentId=(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        result = self._call_api(url, video_id)
+
+        m3u8_url = xpath_text(result, 'url', fatal=True)
+        formats = self._extract_m3u8_formats(re.sub(r'/([^/]+)/[^/?]+\.m3u8', r'/\1/\1.m3u8', m3u8_url), video_id, 'mp4', fatal=False)
+        if len(formats) < 2:
+            formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
+        # Despite metadata in m3u8 all video+audio formats are
+        # actually video-only (no audio)
+        for f in formats:
+            if f.get('acodec') != 'none' and f.get('vcodec') != 'none':
+                f['acodec'] = 'none'
+        self._sort_formats(formats)
+
+        info = {
+            'id': video_id,
+            'title': video_id,
+            'formats': formats,
+        }
+
+        rss = xpath_element(result, 'rss')
+        if rss:
+            info.update(self._parse_rss_feed(rss)['entries'][0])
+            del info['url']
+            del info['_type']
+            del info['ie_key']
+        return info
+
+
+class CBCWatchIE(CBCWatchBaseIE):
+    IE_NAME = 'cbc.ca:watch'
+    _VALID_URL = r'https?://watch\.cbc\.ca/(?:[^/]+/)+(?P<id>[0-9a-f-]+)'
+    _TESTS = [{
+        'url': 'http://watch.cbc.ca/doc-zone/season-6/customer-disservice/38e815a-009e3ab12e4',
+        'info_dict': {
+            'id': '38e815a-009e3ab12e4',
+            'ext': 'mp4',
+            'title': 'Customer (Dis)Service',
+            'description': 'md5:8bdd6913a0fe03d4b2a17ebe169c7c87',
+            'upload_date': '20160219',
+            'timestamp': 1455840000,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+            'format': 'bestvideo',
+        },
+        'skip': 'Geo-restricted to Canada',
+    }, {
+        'url': 'http://watch.cbc.ca/arthur/all/1ed4b385-cd84-49cf-95f0-80f004680057',
+        'info_dict': {
+            'id': '1ed4b385-cd84-49cf-95f0-80f004680057',
+            'title': 'Arthur',
+            'description': 'Arthur, the sweetest 8-year-old aardvark, and his pals solve all kinds of problems with humour, kindness and teamwork.',
+        },
+        'playlist_mincount': 30,
+        'skip': 'Geo-restricted to Canada',
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        rss = self._call_api('web/browse/' + video_id, video_id)
+        return self._parse_rss_feed(rss)

youtube_dl/extractor/cbs.py

@@ -51,7 +51,7 @@ class CBSIE(CBSBaseIE):
         path = 'dJ5BDC/media/guid/2198311517/' + guid
         smil_url = 'http://link.theplatform.com/s/%s?mbr=true' % path
         formats, subtitles = self._extract_theplatform_smil(smil_url + '&manifest=m3u', guid)
-        for r in ('HLS&formats=M3U', 'RTMP', 'WIFI', '3G'):
+        for r in ('OnceURL&formats=M3U', 'HLS&formats=M3U', 'RTMP', 'WIFI', '3G'):
             try:
                 tp_formats, _ = self._extract_theplatform_smil(smil_url + '&assetTypes=' + r, guid, 'Downloading %s SMIL data' % r.split('&')[0])
                 formats.extend(tp_formats)

youtube_dl/extractor/cnn.py

@@ -3,14 +3,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    parse_duration,
-    url_basename,
-)
+from .turner import TurnerBaseIE
+from ..utils import url_basename
 
 
-class CNNIE(InfoExtractor):
+class CNNIE(TurnerBaseIE):
     _VALID_URL = r'''(?x)https?://(?:(?P<sub_domain>edition|www|money)\.)?cnn\.com/(?:video/(?:data/.+?|\?)/)?videos?/
         (?P<path>.+?/(?P<title>[^/]+?)(?:\.(?:[a-z\-]+)|(?=&)))'''
 
@@ -25,6 +22,7 @@ class CNNIE(InfoExtractor):
             'duration': 135,
             'upload_date': '20130609',
         },
+        'expected_warnings': ['Failed to download m3u8 information'],
     }, {
         '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',
@@ -34,7 +32,8 @@ class CNNIE(InfoExtractor):
             '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',
-        }
+        },
+        'expected_warnings': ['Failed to download m3u8 information'],
     }, {
         'url': 'http://www.cnn.com/video/data/2.0/video/living/2014/12/22/growing-america-nashville-salemtown-board-episode-1.hln.html',
         'md5': 'f14d02ebd264df951feb2400e2c25a1b',
@@ -44,7 +43,8 @@ class CNNIE(InfoExtractor):
             'title': 'Nashville Ep. 1: Hand crafted skateboards',
             'description': 'md5:e7223a503315c9f150acac52e76de086',
             'upload_date': '20141222',
-        }
+        },
+        'expected_warnings': ['Failed to download m3u8 information'],
     }, {
         'url': 'http://money.cnn.com/video/news/2016/08/19/netflix-stunning-stats.cnnmoney/index.html',
         'md5': '52a515dc1b0f001cd82e4ceda32be9d1',
@@ -54,7 +54,11 @@ class CNNIE(InfoExtractor):
             'title': '5 stunning stats about Netflix',
             'description': 'Did you know that Netflix has more than 80 million members? Here are five facts about the online video distributor that you probably didn\'t know.',
             'upload_date': '20160819',
-        }
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }, {
         'url': 'http://cnn.com/video/?/video/politics/2015/03/27/pkg-arizona-senator-church-attendance-mandatory.ktvk',
         'only_matching': True,
@@ -79,72 +83,21 @@ class CNNIE(InfoExtractor):
         },
     }
 
+    def _extract_timestamp(self, video_data):
+        # TODO: fix timestamp extraction
+        return None
+
     def _real_extract(self, url):
         sub_domain, path, page_title = re.match(self._VALID_URL, url).groups()
         if sub_domain not in ('money', 'edition'):
             sub_domain = 'edition'
         config = self._CONFIG[sub_domain]
-        info_url = config['data_src'] % path
-        info = self._download_xml(info_url, page_title)
-
-        formats = []
-        rex = re.compile(r'''(?x)
-            (?P<width>[0-9]+)x(?P<height>[0-9]+)
-            (?:_(?P<bitrate>[0-9]+)k)?
-        ''')
-        for f in info.findall('files/file'):
-            video_url = config['media_src'] + f.text.strip()
-            fdct = {
-                'format_id': f.attrib['bitrate'],
-                'url': video_url,
-            }
-
-            mf = rex.match(f.attrib['bitrate'])
-            if mf:
-                fdct['width'] = int(mf.group('width'))
-                fdct['height'] = int(mf.group('height'))
-                fdct['tbr'] = int_or_none(mf.group('bitrate'))
-            else:
-                mf = rex.search(f.text)
-                if mf:
-                    fdct['width'] = int(mf.group('width'))
-                    fdct['height'] = int(mf.group('height'))
-                    fdct['tbr'] = int_or_none(mf.group('bitrate'))
-                else:
-                    mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
-                    if mi:
-                        if mi.group(1) == 'audio':
-                            fdct['vcodec'] = 'none'
-                            fdct['ext'] = 'm4a'
-                        else:
-                            fdct['tbr'] = int(mi.group(1))
-
-            formats.append(fdct)
-
-        self._sort_formats(formats)
-
-        thumbnails = [{
-            'height': int(t.attrib['height']),
-            'width': int(t.attrib['width']),
-            'url': t.text,
-        } for t in info.findall('images/image')]
-
-        metas_el = info.find('metas')
-        upload_date = (
-            metas_el.attrib.get('version') if metas_el is not None else None)
-
-        duration_el = info.find('length')
-        duration = parse_duration(duration_el.text)
-
-        return {
-            'id': info.attrib['id'],
-            'title': info.find('headline').text,
-            'formats': formats,
-            'thumbnails': thumbnails,
-            'description': info.find('description').text,
-            'duration': duration,
-            'upload_date': upload_date,
-        }
+        return self._extract_cvp_info(
+            config['data_src'] % path, page_title, {
+                'default': {
+                    'media_src': config['media_src'],
+                }
+            })
 
 
 class CNNBlogsIE(InfoExtractor):
@@ -159,6 +112,7 @@ class CNNBlogsIE(InfoExtractor):
             'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.',
             'upload_date': '20140209',
         },
+        'expected_warnings': ['Failed to download m3u8 information'],
         'add_ie': ['CNN'],
     }
 
@@ -181,9 +135,10 @@ class CNNArticleIE(InfoExtractor):
             'id': 'bestoftv/2014/12/21/ip-north-korea-obama.cnn',
             'ext': 'mp4',
             'title': 'Obama: Cyberattack not an act of war',
-            'description': 'md5:51ce6750450603795cad0cdfbd7d05c5',
+            'description': 'md5:0a802a40d2376f60e6b04c8d5bcebc4b',
             'upload_date': '20141221',
         },
+        'expected_warnings': ['Failed to download m3u8 information'],
         'add_ie': ['CNN'],
     }
 

youtube_dl/extractor/common.py

@@ -1201,27 +1201,44 @@ class InfoExtractor(object):
                 'protocol': entry_protocol,
                 'preference': preference,
             }]
-        last_info = None
-        last_media = None
+        last_info = {}
+        last_media = {}
         for line in m3u8_doc.splitlines():
             if line.startswith('#EXT-X-STREAM-INF:'):
                 last_info = parse_m3u8_attributes(line)
             elif line.startswith('#EXT-X-MEDIA:'):
-                last_media = parse_m3u8_attributes(line)
+                media = parse_m3u8_attributes(line)
+                media_type = media.get('TYPE')
+                if media_type in ('VIDEO', 'AUDIO'):
+                    media_url = media.get('URI')
+                    if media_url:
+                        format_id = []
+                        for v in (media.get('GROUP-ID'), media.get('NAME')):
+                            if v:
+                                format_id.append(v)
+                        formats.append({
+                            'format_id': '-'.join(format_id),
+                            'url': format_url(media_url),
+                            'language': media.get('LANGUAGE'),
+                            'vcodec': 'none' if media_type == 'AUDIO' else None,
+                            'ext': ext,
+                            'protocol': entry_protocol,
+                            'preference': preference,
+                        })
+                    else:
+                        # When there is no URI in EXT-X-MEDIA let this tag's
+                        # data be used by regular URI lines below
+                        last_media = media
             elif line.startswith('#') or not line.strip():
                 continue
             else:
-                if last_info is None:
-                    formats.append({'url': format_url(line)})
-                    continue
-                tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000)
+                tbr = int_or_none(last_info.get('AVERAGE-BANDWIDTH') or last_info.get('BANDWIDTH'), scale=1000)
                 format_id = []
                 if m3u8_id:
                     format_id.append(m3u8_id)
-                last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') not in ('SUBTITLES', 'CLOSED-CAPTIONS') else None
                 # Despite specification does not mention NAME attribute for
                 # EXT-X-STREAM-INF it still sometimes may be present
-                stream_name = last_info.get('NAME') or last_media_name
+                stream_name = last_info.get('NAME') or last_media.get('NAME')
                 # Bandwidth of live streams may differ over time thus making
                 # format_id unpredictable. So it's better to keep provided
                 # format_id intact.
@@ -1252,11 +1269,9 @@ class InfoExtractor(object):
                         'abr': abr,
                     })
                 f.update(parse_codecs(last_info.get('CODECS')))
-                if last_media is not None:
-                    f['m3u8_media'] = last_media
-                    last_media = None
                 formats.append(f)
                 last_info = {}
+                last_media = {}
         return formats
 
     @staticmethod

youtube_dl/extractor/crackle.py

@@ -1,5 +1,5 @@
 # coding: utf-8
-from __future__ import unicode_literals
+from __future__ import unicode_literals, division
 
 from .common import InfoExtractor
 from ..utils import int_or_none
@@ -8,12 +8,22 @@ 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',
+        'url': 'http://www.crackle.com/comedians-in-cars-getting-coffee/2498934',
         'info_dict': {
-            'id': '2496419',
+            'id': '2498934',
             'ext': 'mp4',
-            'title': 'Heavy Lies the Head',
-            'description': 'md5:bb56aa0708fe7b9a4861535f15c3abca',
+            'title': 'Everybody Respects A Bloody Nose',
+            'description': 'Jerry is kaffeeklatsching in L.A. with funnyman J.B. Smoove (Saturday Night Live, Real Husbands of Hollywood). They’re headed for brew at 10 Speed Coffee in a 1964 Studebaker Avanti.',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 906,
+            'series': 'Comedians In Cars Getting Coffee',
+            'season_number': 8,
+            'episode_number': 4,
+            'subtitles': {
+                'en-US': [{
+                    'ext': 'ttml',
+                }]
+            },
         },
         'params': {
             # m3u8 download
@@ -21,12 +31,8 @@ class CrackleIE(InfoExtractor):
         }
     }
 
-    # 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
+    _THUMBNAIL_TEMPLATE = 'http://images-us-am.crackle.com/%stnl_1920x1080.jpg?ts=20140107233116?c=635333335057637614'
     _MEDIA_FILE_SLOTS = {
         'c544.flv': {
             'width': 544,
@@ -48,16 +54,21 @@ class CrackleIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
+        config_doc = self._download_xml(
+            'http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx?site=16',
+            video_id, 'Downloading config')
+
         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),
+            'http://content.uplynk.com/ext/%s/%s.m3u8' % (config_doc.attrib['strUplynkOwnerId'], video_id),
             video_id, 'mp4', m3u8_id='hls', fatal=None)
+        thumbnail = None
         path = item.attrib.get('p')
         if path:
             thumbnail = self._THUMBNAIL_TEMPLATE % path
@@ -76,7 +87,7 @@ class CrackleIE(InfoExtractor):
                     if locale not in subtitles:
                         subtitles[locale] = []
                     subtitles[locale] = [{
-                        'url': '%s/%s%s_%s.xml' % (self._SUBTITLE_SERVER, path, locale, v),
+                        'url': '%s/%s%s_%s.xml' % (config_doc.attrib['strSubtitleServer'], path, locale, v),
                         'ext': 'ttml',
                     }]
         self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
@@ -85,7 +96,7 @@ class CrackleIE(InfoExtractor):
             'id': video_id,
             'title': title,
             'description': item.attrib.get('d'),
-            'duration': int(item.attrib.get('r'), 16) if item.attrib.get('r') else None,
+            'duration': int(item.attrib.get('r'), 16) / 1000 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')),

youtube_dl/extractor/ctv.py

@@ -1,11 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 
 
 class CTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?ctv\.ca/video/player\?vid=(?P<id>[0-9.]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?P<domain>ctv|tsn|bnn|thecomedynetwork)\.ca/.*?(?:\bvid=|-vid|~|%7E)(?P<id>[0-9.]+)'
     _TESTS = [{
         'url': 'http://www.ctv.ca/video/player?vid=706966',
         'md5': 'ff2ebbeae0aa2dcc32a830c3fd69b7b0',
@@ -18,13 +20,27 @@ class CTVIE(InfoExtractor):
             'timestamp': 1442624700,
         },
         'expected_warnings': ['HTTP Error 404'],
+    }, {
+        'url': 'http://www.thecomedynetwork.ca/video/player?vid=923582',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.tsn.ca/video/expectations-high-for-milos-raonic-at-us-open~939549',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bnn.ca/video/berman-s-call-part-two-viewer-questions~939654',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.ctv.ca/YourMorning/Video/S1E6-Monday-August-29-2016-vid938009',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        domain, video_id = re.match(self._VALID_URL, url).groups()
+        if domain == 'thecomedynetwork':
+            domain = 'comedy'
         return {
             '_type': 'url_transparent',
             'id': video_id,
-            'url': '9c9media:ctv_web:%s' % video_id,
+            'url': '9c9media:%s_web:%s' % (domain, video_id),
             'ie_key': 'NineCNineMedia',
         }

youtube_dl/extractor/curiositystream.py

@@ -0,0 +1,120 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    urlencode_postdata,
+    compat_str,
+    ExtractorError,
+)
+
+
+class CuriosityStreamBaseIE(InfoExtractor):
+    _NETRC_MACHINE = 'curiositystream'
+    _auth_token = None
+    _API_BASE_URL = 'https://api.curiositystream.com/v1/'
+
+    def _handle_errors(self, result):
+        error = result.get('error', {}).get('message')
+        if error:
+            if isinstance(error, dict):
+                error = ', '.join(error.values())
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, error), expected=True)
+
+    def _call_api(self, path, video_id):
+        headers = {}
+        if self._auth_token:
+            headers['X-Auth-Token'] = self._auth_token
+        result = self._download_json(
+            self._API_BASE_URL + path, video_id, headers=headers)
+        self._handle_errors(result)
+        return result['data']
+
+    def _real_initialize(self):
+        (email, password) = self._get_login_info()
+        if email is None:
+            return
+        result = self._download_json(
+            self._API_BASE_URL + 'login', None, data=urlencode_postdata({
+                'email': email,
+                'password': password,
+            }))
+        self._handle_errors(result)
+        self._auth_token = result['message']['auth_token']
+
+    def _extract_media_info(self, media):
+        video_id = compat_str(media['id'])
+        limelight_media_id = media['limelight_media_id']
+        title = media['title']
+
+        subtitles = {}
+        for closed_caption in media.get('closed_captions', []):
+            sub_url = closed_caption.get('file')
+            if not sub_url:
+                continue
+            lang = closed_caption.get('code') or closed_caption.get('language') or 'en'
+            subtitles.setdefault(lang, []).append({
+                'url': sub_url,
+            })
+
+        return {
+            '_type': 'url_transparent',
+            'id': video_id,
+            'url': 'limelight:media:' + limelight_media_id,
+            'title': title,
+            'description': media.get('description'),
+            'thumbnail': media.get('image_large') or media.get('image_medium') or media.get('image_small'),
+            'duration': int_or_none(media.get('duration')),
+            'tags': media.get('tags'),
+            'subtitles': subtitles,
+            'ie_key': 'LimelightMedia',
+        }
+
+
+class CuriosityStreamIE(CuriosityStreamBaseIE):
+    IE_NAME = 'curiositystream'
+    _VALID_URL = r'https?://app\.curiositystream\.com/video/(?P<id>\d+)'
+    _TEST = {
+        'url': 'https://app.curiositystream.com/video/2',
+        'md5': 'a0074c190e6cddaf86900b28d3e9ee7a',
+        'info_dict': {
+            'id': '2',
+            'ext': 'mp4',
+            'title': 'How Did You Develop The Internet?',
+            'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.',
+            'timestamp': 1448388615,
+            'upload_date': '20151124',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        media = self._call_api('media/' + video_id, video_id)
+        return self._extract_media_info(media)
+
+
+class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
+    IE_NAME = 'curiositystream:collection'
+    _VALID_URL = r'https?://app\.curiositystream\.com/collection/(?P<id>\d+)'
+    _TEST = {
+        'url': 'https://app.curiositystream.com/collection/2',
+        'info_dict': {
+            'id': '2',
+            'title': 'Curious Minds: The Internet',
+            'description': 'How is the internet shaping our lives in the 21st Century?',
+        },
+        'playlist_mincount': 17,
+    }
+
+    def _real_extract(self, url):
+        collection_id = self._match_id(url)
+        collection = self._call_api(
+            'collections/' + collection_id, collection_id)
+        entries = []
+        for media in collection.get('media', []):
+            entries.append(self._extract_media_info(media))
+        return self.playlist_result(
+            entries, collection_id,
+            collection.get('title'), collection.get('description'))

youtube_dl/extractor/discoverygo.py

@@ -7,6 +7,7 @@ from ..utils import (
     int_or_none,
     parse_age_limit,
     unescapeHTML,
+    ExtractorError,
 )
 
 
@@ -53,7 +54,14 @@ class DiscoveryGoIE(InfoExtractor):
 
         title = video['name']
 
-        stream = video['stream']
+        stream = video.get('stream')
+        if not stream:
+            if video.get('authenticated') is True:
+                raise ExtractorError(
+                    'This video is only available via cable service provider subscription that'
+                    ' is not currently supported. You may want to use --cookies.', expected=True)
+            else:
+                raise ExtractorError('Unable to find stream')
         STREAM_URL_SUFFIX = 'streamUrl'
         formats = []
         for stream_kind in ('', 'hds'):

youtube_dl/extractor/douyutv.py

@@ -3,9 +3,17 @@ from __future__ import unicode_literals
 
 import hashlib
 import time
+import uuid
+
 from .common import InfoExtractor
-from ..utils import (ExtractorError, unescapeHTML)
-from ..compat import (compat_str, compat_basestring)
+from ..compat import (
+    compat_str,
+    compat_urllib_parse_urlencode,
+)
+from ..utils import (
+    ExtractorError,
+    unescapeHTML,
+)
 
 
 class DouyuTVIE(InfoExtractor):
@@ -21,7 +29,6 @@ class DouyuTVIE(InfoExtractor):
             'description': 're:.*m7show@163\.com.*',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': '7师傅',
-            'uploader_id': '431925',
             'is_live': True,
         },
         'params': {
@@ -37,7 +44,6 @@ class DouyuTVIE(InfoExtractor):
             'description': 'md5:746a2f7a253966a06755a912f0acc0d2',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': 'douyu小漠',
-            'uploader_id': '3769985',
             'is_live': True,
         },
         'params': {
@@ -54,7 +60,6 @@ class DouyuTVIE(InfoExtractor):
             'description': 're:.*m7show@163\.com.*',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': '7师傅',
-            'uploader_id': '431925',
             'is_live': True,
         },
         'params': {
@@ -65,6 +70,10 @@ class DouyuTVIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    # Decompile core.swf in webpage by ffdec "Search SWFs in memory". core.swf
+    # is encrypted originally, but ffdec can dump memory to get the decrypted one.
+    _API_KEY = 'A12Svb&%1UUmf@hC'
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -75,74 +84,56 @@ class DouyuTVIE(InfoExtractor):
             room_id = self._html_search_regex(
                 r'"room_id"\s*:\s*(\d+),', page, 'room id')
 
-        config = None
-        # Douyu API sometimes returns error "Unable to load the requested class: eticket_redis_cache"
-        # Retry with different parameters - same parameters cause same errors
-        for i in range(5):
-            prefix = 'room/%s?aid=android&client_sys=android&time=%d' % (
-                room_id, int(time.time()))
-            auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest()
+        room = self._download_json(
+            'http://m.douyu.com/html5/live?roomId=%s' % room_id, video_id,
+            note='Downloading room info')['data']
 
-            config_page = self._download_webpage(
-                'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth),
-                video_id)
-            try:
-                config = self._parse_json(config_page, video_id, fatal=False)
-            except ExtractorError:
-                # Wait some time before retrying to get a different time() value
-                self._sleep(1, video_id, msg_template='%(video_id)s: Error occurs. '
-                                                      'Waiting for %(timeout)s seconds before retrying')
-                continue
-            else:
-                break
-        if config is None:
-            raise ExtractorError('Unable to fetch API result')
-
-        data = config['data']
-
-        error_code = config.get('error', 0)
-        if error_code is not 0:
-            error_desc = 'Server reported error %i' % error_code
-            if isinstance(data, (compat_str, compat_basestring)):
-                error_desc += ': ' + data
-            raise ExtractorError(error_desc, expected=True)
-
-        show_status = data.get('show_status')
         # 1 = live, 2 = offline
-        if show_status == '2':
+        if room.get('show_status') == '2':
+            raise ExtractorError('Live stream is offline', expected=True)
+
+        tt = compat_str(int(time.time() / 60))
+        did = uuid.uuid4().hex.upper()
+
+        sign_content = ''.join((room_id, did, self._API_KEY, tt))
+        sign = hashlib.md5((sign_content).encode('utf-8')).hexdigest()
+
+        flv_data = compat_urllib_parse_urlencode({
+            'cdn': 'ws',
+            'rate': '0',
+            'tt': tt,
+            'did': did,
+            'sign': sign,
+        })
+
+        video_info = self._download_json(
+            'http://www.douyu.com/lapi/live/getPlay/%s' % room_id, video_id,
+            data=flv_data, note='Downloading video info',
+            headers={'Content-Type': 'application/x-www-form-urlencoded'})
+
+        error_code = video_info.get('error', 0)
+        if error_code is not 0:
             raise ExtractorError(
-                'Live stream is offline', expected=True)
+                '%s reported error %i' % (self.IE_NAME, error_code),
+                expected=True)
 
-        base_url = data['rtmp_url']
-        live_path = data['rtmp_live']
+        base_url = video_info['data']['rtmp_url']
+        live_path = video_info['data']['rtmp_live']
 
-        title = self._live_title(unescapeHTML(data['room_name']))
-        description = data.get('show_details')
-        thumbnail = data.get('room_src')
+        video_url = '%s/%s' % (base_url, live_path)
 
-        uploader = data.get('nickname')
-        uploader_id = data.get('owner_uid')
-
-        multi_formats = data.get('rtmp_multi_bitrate')
-        if not isinstance(multi_formats, dict):
-            multi_formats = {}
-        multi_formats['live'] = live_path
-
-        formats = [{
-            'url': '%s/%s' % (base_url, format_path),
-            'format_id': format_id,
-            'preference': 1 if format_id == 'live' else 0,
-        } for format_id, format_path in multi_formats.items()]
-        self._sort_formats(formats)
+        title = self._live_title(unescapeHTML(room['room_name']))
+        description = room.get('notice')
+        thumbnail = room.get('room_src')
+        uploader = room.get('nickname')
 
         return {
             'id': room_id,
             'display_id': video_id,
+            'url': video_url,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'uploader_id': uploader_id,
-            'formats': formats,
             'is_live': True,
         }

youtube_dl/extractor/drtv.py

@@ -4,26 +4,45 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    int_or_none,
+    float_or_none,
+    mimetype2ext,
     parse_iso8601,
+    remove_end,
 )
 
 
 class DRTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dr\.dk/tv/se/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
+    _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv/se|nyheder)/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
 
-    _TEST = {
-        'url': 'https://www.dr.dk/tv/se/boern/ultra/panisk-paske/panisk-paske-5',
-        'md5': 'dc515a9ab50577fa14cc4e4b0265168f',
+    _TESTS = [{
+        'url': 'https://www.dr.dk/tv/se/boern/ultra/klassen-ultra/klassen-darlig-taber-10',
+        'md5': '25e659cccc9a2ed956110a299fdf5983',
         'info_dict': {
-            'id': 'panisk-paske-5',
+            'id': 'klassen-darlig-taber-10',
             'ext': 'mp4',
-            'title': 'Panisk Påske (5)',
-            'description': 'md5:ca14173c5ab24cd26b0fcc074dff391c',
-            'timestamp': 1426984612,
-            'upload_date': '20150322',
-            'duration': 1455,
+            'title': 'Klassen - Dårlig taber (10)',
+            'description': 'md5:815fe1b7fa656ed80580f31e8b3c79aa',
+            'timestamp': 1471991907,
+            'upload_date': '20160823',
+            'duration': 606.84,
         },
-    }
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.dr.dk/nyheder/indland/live-christianias-rydning-af-pusher-street-er-i-gang',
+        'md5': '2c37175c718155930f939ef59952474a',
+        'info_dict': {
+            'id': 'christiania-pusher-street-ryddes-drdkrjpo',
+            'ext': 'mp4',
+            'title': 'LIVE Christianias rydning af Pusher Street er i gang',
+            'description': '- Det er det fedeste, der er sket i 20 år, fortæller christianit til DR Nyheder.',
+            'timestamp': 1472800279,
+            'upload_date': '20160902',
+            'duration': 131.4,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -35,7 +54,8 @@ class DRTVIE(InfoExtractor):
                 'Video %s is not available' % video_id, expected=True)
 
         video_id = self._search_regex(
-            r'data-(?:material-identifier|episode-slug)="([^"]+)"',
+            (r'data-(?:material-identifier|episode-slug)="([^"]+)"',
+                r'data-resource="[^>"]+mu/programcard/expanded/([^"]+)"'),
             webpage, 'video id')
 
         programcard = self._download_json(
@@ -43,9 +63,12 @@ class DRTVIE(InfoExtractor):
             video_id, 'Downloading video JSON')
         data = programcard['Data'][0]
 
-        title = data['Title']
-        description = data['Description']
-        timestamp = parse_iso8601(data['CreatedTime'])
+        title = remove_end(self._og_search_title(
+            webpage, default=None), ' | TV | DR') or data['Title']
+        description = self._og_search_description(
+            webpage, default=None) or data.get('Description')
+
+        timestamp = parse_iso8601(data.get('CreatedTime'))
 
         thumbnail = None
         duration = None
@@ -56,16 +79,18 @@ class DRTVIE(InfoExtractor):
         subtitles = {}
 
         for asset in data['Assets']:
-            if asset['Kind'] == 'Image':
-                thumbnail = asset['Uri']
-            elif asset['Kind'] == 'VideoResource':
-                duration = asset['DurationInMilliseconds'] / 1000.0
-                restricted_to_denmark = asset['RestrictedToDenmark']
-                spoken_subtitles = asset['Target'] == 'SpokenSubtitles'
-                for link in asset['Links']:
-                    uri = link['Uri']
-                    target = link['Target']
-                    format_id = target
+            if asset.get('Kind') == 'Image':
+                thumbnail = asset.get('Uri')
+            elif asset.get('Kind') == 'VideoResource':
+                duration = float_or_none(asset.get('DurationInMilliseconds'), 1000)
+                restricted_to_denmark = asset.get('RestrictedToDenmark')
+                spoken_subtitles = asset.get('Target') == 'SpokenSubtitles'
+                for link in asset.get('Links', []):
+                    uri = link.get('Uri')
+                    if not uri:
+                        continue
+                    target = link.get('Target')
+                    format_id = target or ''
                     preference = None
                     if spoken_subtitles:
                         preference = -1
@@ -76,8 +101,8 @@ class DRTVIE(InfoExtractor):
                             video_id, preference, f4m_id=format_id))
                     elif target == 'HLS':
                         formats.extend(self._extract_m3u8_formats(
-                            uri, video_id, 'mp4', preference=preference,
-                            m3u8_id=format_id))
+                            uri, video_id, 'mp4', entry_protocol='m3u8_native',
+                            preference=preference, m3u8_id=format_id))
                     else:
                         bitrate = link.get('Bitrate')
                         if bitrate:
@@ -85,7 +110,7 @@ class DRTVIE(InfoExtractor):
                         formats.append({
                             'url': uri,
                             'format_id': format_id,
-                            'tbr': bitrate,
+                            'tbr': int_or_none(bitrate),
                             'ext': link.get('FileFormat'),
                         })
                 subtitles_list = asset.get('SubtitlesList')
@@ -94,12 +119,18 @@ class DRTVIE(InfoExtractor):
                         'Danish': 'da',
                     }
                     for subs in subtitles_list:
-                        lang = subs['Language']
-                        subtitles[LANGS.get(lang, lang)] = [{'url': subs['Uri'], 'ext': 'vtt'}]
+                        if not subs.get('Uri'):
+                            continue
+                        lang = subs.get('Language') or 'da'
+                        subtitles.setdefault(LANGS.get(lang, lang), []).append({
+                            'url': subs['Uri'],
+                            'ext': mimetype2ext(subs.get('MimeType')) or 'vtt'
+                        })
 
         if not formats and restricted_to_denmark:
-            raise ExtractorError(
-                'Unfortunately, DR is not allowed to show this program outside Denmark.', expected=True)
+            self.raise_geo_restricted(
+                'Unfortunately, DR is not allowed to show this program outside Denmark.',
+                expected=True)
 
         self._sort_formats(formats)
 

youtube_dl/extractor/exfm.py

@@ -1,58 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-
-
-class ExfmIE(InfoExtractor):
-    IE_NAME = 'exfm'
-    IE_DESC = 'ex.fm'
-    _VALID_URL = r'https?://(?:www\.)?ex\.fm/song/(?P<id>[^/]+)'
-    _SOUNDCLOUD_URL = r'http://(?:www\.)?api\.soundcloud\.com/tracks/([^/]+)/stream'
-    _TESTS = [
-        {
-            'url': 'http://ex.fm/song/eh359',
-            'md5': 'e45513df5631e6d760970b14cc0c11e7',
-            'info_dict': {
-                'id': '44216187',
-                'ext': 'mp3',
-                'title': 'Test House "Love Is Not Enough" (Extended Mix) DeadJournalist Exclusive',
-                'uploader': 'deadjournalist',
-                'upload_date': '20120424',
-                'description': 'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive',
-            },
-            'note': 'Soundcloud song',
-            'skip': 'The site is down too often',
-        },
-        {
-            'url': 'http://ex.fm/song/wddt8',
-            'md5': '966bd70741ac5b8570d8e45bfaed3643',
-            'info_dict': {
-                'id': 'wddt8',
-                'ext': 'mp3',
-                'title': 'Safe and Sound',
-                'uploader': 'Capital Cities',
-            },
-            'skip': 'The site is down too often',
-        },
-    ]
-
-    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 = self._download_json(info_url, song_id)['song']
-        song_url = info['url']
-        if re.match(self._SOUNDCLOUD_URL, song_url) is not None:
-            self.to_screen('Soundcloud song detected')
-            return self.url_result(song_url.replace('/stream', ''), 'Soundcloud')
-        return {
-            'id': song_id,
-            'url': song_url,
-            'ext': 'mp3',
-            'title': info['title'],
-            'thumbnail': info['image']['large'],
-            'uploader': info['artist'],
-            'view_count': info['loved_count'],
-        }

youtube_dl/extractor/extractors.py

@@ -127,9 +127,12 @@ from .carambatv import (
     CarambaTVIE,
     CarambaTVPageIE,
 )
+from .cartoonnetwork import CartoonNetworkIE
 from .cbc import (
     CBCIE,
     CBCPlayerIE,
+    CBCWatchVideoIE,
+    CBCWatchIE,
 )
 from .cbs import CBSIE
 from .cbslocal import CBSLocalIE
@@ -191,6 +194,10 @@ from .ctsnews import CtsNewsIE
 from .ctv import CTVIE
 from .ctvnews import CTVNewsIE
 from .cultureunplugged import CultureUnpluggedIE
+from .curiositystream import (
+    CuriosityStreamIE,
+    CuriosityStreamCollectionIE,
+)
 from .cwtv import CWTVIE
 from .dailymail import DailyMailIE
 from .dailymotion import (
@@ -254,11 +261,13 @@ from .espn import ESPNIE
 from .esri import EsriVideoIE
 from .europa import EuropaIE
 from .everyonesmixtape import EveryonesMixtapeIE
-from .exfm import ExfmIE
 from .expotv import ExpoTVIE
 from .extremetube import ExtremeTubeIE
 from .eyedotv import EyedoTVIE
-from .facebook import FacebookIE
+from .facebook import (
+    FacebookIE,
+    FacebookPluginsVideoIE,
+)
 from .faz import FazIE
 from .fc2 import FC2IE
 from .fczenit import FczenitIE
@@ -312,6 +321,7 @@ from .globo import (
     GloboIE,
     GloboArticleIE,
 )
+from .go import GoIE
 from .godtube import GodTubeIE
 from .godtv import GodTVIE
 from .golem import GolemIE
@@ -483,6 +493,7 @@ from .motherless import MotherlessIE
 from .motorsport import MotorsportIE
 from .movieclips import MovieClipsIE
 from .moviezine import MoviezineIE
+from .movingimage import MovingImageIE
 from .msn import MSNIE
 from .mtv import (
     MTVIE,
@@ -539,6 +550,7 @@ from .nextmedia import (
 )
 from .nfb import NFBIE
 from .nfl import NFLIE
+from .nhk import NhkVodIE
 from .nhl import (
     NHLVideocenterIE,
     NHLNewsIE,
@@ -550,7 +562,10 @@ from .nick import (
     NickDeIE,
 )
 from .niconico import NiconicoIE, NiconicoPlaylistIE
-from .ninecninemedia import NineCNineMediaIE
+from .ninecninemedia import (
+    NineCNineMediaStackIE,
+    NineCNineMediaIE,
+)
 from .ninegag import NineGagIE
 from .ninenow import NineNowIE
 from .nintendo import NintendoIE
@@ -632,7 +647,6 @@ from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
 from .pinkbike import PinkbikeIE
 from .pladform import PladformIE
-from .played import PlayedIE
 from .playfm import PlayFMIE
 from .plays import PlaysTVIE
 from .playtvak import PlaytvakIE
@@ -800,7 +814,6 @@ from .srgssr import (
     SRGSSRPlayIE,
 )
 from .srmediathek import SRMediathekIE
-from .ssa import SSAIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .steam import SteamIE
 from .streamable import StreamableIE
@@ -820,6 +833,7 @@ from .tagesschau import (
     TagesschauIE,
 )
 from .tass import TassIE
+from .tbs import TBSIE
 from .tdslifeway import TDSLifewayIE
 from .teachertube import (
     TeacherTubeIE,
@@ -862,16 +876,11 @@ from .tnaflix import (
     MovieFapIE,
 )
 from .toggle import ToggleIE
-from .thvideo import (
-    THVideoIE,
-    THVideoPlaylistIE
-)
 from .toutv import TouTvIE
 from .toypics import ToypicsUserIE, ToypicsIE
 from .traileraddict import TrailerAddictIE
 from .trilulilu import TriluliluIE
 from .trollvids import TrollvidsIE
-from .trutube import TruTubeIE
 from .tube8 import Tube8IE
 from .tubitv import TubiTvIE
 from .tudou import (
@@ -945,6 +954,7 @@ from .uplynk import (
 )
 from .urort import UrortIE
 from .urplay import URPlayIE
+from .usanetwork import USANetworkIE
 from .usatoday import USATodayIE
 from .ustream import UstreamIE, UstreamChannelIE
 from .ustudio import (

youtube_dl/extractor/facebook.py

@@ -15,6 +15,7 @@ from ..compat import (
 from ..utils import (
     error_to_compat_str,
     ExtractorError,
+    int_or_none,
     limit_length,
     sanitized_Request,
     urlencode_postdata,
@@ -62,6 +63,8 @@ class FacebookIE(InfoExtractor):
             'ext': 'mp4',
             'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
             'uploader': 'Tennis on Facebook',
+            'upload_date': '20140908',
+            'timestamp': 1410199200,
         }
     }, {
         'note': 'Video without discernible title',
@@ -71,6 +74,8 @@ class FacebookIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Facebook video #274175099429670',
             'uploader': 'Asif Nawab Butt',
+            'upload_date': '20140506',
+            'timestamp': 1399398998,
         },
         'expected_warnings': [
             'title'
@@ -78,12 +83,14 @@ class FacebookIE(InfoExtractor):
     }, {
         'note': 'Video with DASH manifest',
         'url': 'https://www.facebook.com/video.php?v=957955867617029',
-        'md5': '54706e4db4f5ad58fbad82dde1f1213f',
+        'md5': 'b2c28d528273b323abe5c6ab59f0f030',
         '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',
+            'upload_date': '20160110',
+            'timestamp': 1452431627,
         },
     }, {
         'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
@@ -306,12 +313,16 @@ class FacebookIE(InfoExtractor):
         if not video_title:
             video_title = 'Facebook video #%s' % video_id
         uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage))
+        timestamp = int_or_none(self._search_regex(
+            r'<abbr[^>]+data-utime=["\'](\d+)', webpage,
+            'timestamp', default=None))
 
         info_dict = {
             'id': video_id,
             'title': video_title,
             'formats': formats,
             'uploader': uploader,
+            'timestamp': timestamp,
         }
 
         return webpage, info_dict
@@ -340,3 +351,32 @@ class FacebookIE(InfoExtractor):
                 self._VIDEO_PAGE_TEMPLATE % video_id,
                 video_id, fatal_if_no_video=True)
             return info_dict
+
+
+class FacebookPluginsVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:[\w-]+\.)?facebook\.com/plugins/video\.php\?.*?\bhref=(?P<id>https.+)'
+
+    _TESTS = [{
+        'url': 'https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Fgov.sg%2Fvideos%2F10154383743583686%2F&show_text=0&width=560',
+        'md5': '5954e92cdfe51fe5782ae9bda7058a07',
+        'info_dict': {
+            'id': '10154383743583686',
+            'ext': 'mp4',
+            'title': 'What to do during the haze?',
+            'uploader': 'Gov.sg',
+            'upload_date': '20160826',
+            'timestamp': 1472184808,
+        },
+        'add_ie': [FacebookIE.ie_key()],
+    }, {
+        'url': 'https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Fvideo.php%3Fv%3D10204634152394104',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.facebook.com/plugins/video.php?href=https://www.facebook.com/gov.sg/videos/10154383743583686/&show_text=0&width=560',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        return self.url_result(
+            compat_urllib_parse_unquote(self._match_id(url)),
+            FacebookIE.ie_key())

youtube_dl/extractor/generic.py

@@ -104,7 +104,8 @@ class GenericIE(InfoExtractor):
             },
             'expected_warnings': [
                 'URL could be a direct video link, returning it as such.'
-            ]
+            ],
+            'skip': 'URL invalid',
         },
         # Direct download with broken HEAD
         {
@@ -268,7 +269,8 @@ class GenericIE(InfoExtractor):
             'params': {
                 # m3u8 downloads
                 'skip_download': True,
-            }
+            },
+            'skip': 'video gone',
         },
         # m3u8 served with Content-Type: text/plain
         {
@@ -283,7 +285,8 @@ class GenericIE(InfoExtractor):
             'params': {
                 # m3u8 downloads
                 'skip_download': True,
-            }
+            },
+            'skip': 'video gone',
         },
         # google redirect
         {
@@ -368,6 +371,7 @@ class GenericIE(InfoExtractor):
                 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.',
             },
             'add_ie': ['BrightcoveLegacy'],
+            'skip': 'video gone',
         },
         {
             'url': 'http://www.championat.com/video/football/v/87/87499.html',
@@ -421,6 +425,7 @@ class GenericIE(InfoExtractor):
             'params': {
                 'skip_download': True,
             },
+            'skip': 'movie expired',
         },
         # embed.ly video
         {
@@ -448,6 +453,8 @@ class GenericIE(InfoExtractor):
                 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
             },
+            # HEAD requests lead to endless 301, while GET is OK
+            'expected_warnings': ['301'],
         },
         # RUTV embed
         {
@@ -522,6 +529,9 @@ class GenericIE(InfoExtractor):
                 'title': '[NSFL] [FM15] which pumiscer was this ( vid ) ( alfa as fuck srx )',
             },
             'playlist_mincount': 7,
+            # This forum does not allow <iframe> syntaxes anymore
+            # Now HTML tags are displayed as-is
+            'skip': 'No videos on this page',
         },
         # Embedded TED video
         {
@@ -570,7 +580,8 @@ class GenericIE(InfoExtractor):
             },
             'params': {
                 'skip_download': 'Requires rtmpdump'
-            }
+            },
+            'skip': 'video gone',
         },
         # francetv embed
         {
@@ -2232,11 +2243,11 @@ class GenericIE(InfoExtractor):
 
         # Look for VODPlatform embeds
         mobj = re.search(
-            r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vod-platform\.net/embed/[^/?#]+)',
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vod-platform\.net/[eE]mbed/.+?)\1',
             webpage)
         if mobj is not None:
             return self.url_result(
-                self._proto_relative_url(unescapeHTML(mobj.group(1))), 'VODPlatform')
+                self._proto_relative_url(unescapeHTML(mobj.group('url'))), 'VODPlatform')
 
         # Look for Instagram embeds
         instagram_embed_url = InstagramIE._extract_embed_url(webpage)

youtube_dl/extractor/glide.py

@@ -2,7 +2,6 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import unified_strdate
 
 
 class GlideIE(InfoExtractor):
@@ -14,10 +13,8 @@ class GlideIE(InfoExtractor):
         'info_dict': {
             'id': 'UZF8zlmuQbe4mr+7dCiQ0w==',
             'ext': 'mp4',
-            'title': 'Damon Timm\'s Glide message',
+            'title': "Damon's Glide message",
             'thumbnail': 're:^https?://.*?\.cloudfront\.net/.*\.jpg$',
-            'uploader': 'Damon Timm',
-            'upload_date': '20140919',
         }
     }
 
@@ -27,7 +24,8 @@ class GlideIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         title = self._html_search_regex(
-            r'<title>(.+?)</title>', webpage, 'title')
+            r'<title>(.+?)</title>', webpage,
+            'title', default=None) or self._og_search_title(webpage)
         video_url = self._proto_relative_url(self._search_regex(
             r'<source[^>]+src=(["\'])(?P<url>.+?)\1',
             webpage, 'video URL', default=None,
@@ -36,18 +34,10 @@ class GlideIE(InfoExtractor):
             r'<img[^>]+id=["\']video-thumbnail["\'][^>]+src=(["\'])(?P<url>.+?)\1',
             webpage, 'thumbnail url', default=None,
             group='url')) or self._og_search_thumbnail(webpage)
-        uploader = self._search_regex(
-            r'<div[^>]+class=["\']info-name["\'][^>]*>([^<]+)',
-            webpage, 'uploader', fatal=False)
-        upload_date = unified_strdate(self._search_regex(
-            r'<div[^>]+class="info-date"[^>]*>([^<]+)',
-            webpage, 'upload date', fatal=False))
 
         return {
             'id': video_id,
             'title': title,
             'url': video_url,
             'thumbnail': thumbnail,
-            'uploader': uploader,
-            'upload_date': upload_date,
         }

youtube_dl/extractor/go.py

@@ -0,0 +1,101 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    determine_ext,
+    parse_age_limit,
+)
+
+
+class GoIE(InfoExtractor):
+    _BRANDS = {
+        'abc': '001',
+        'freeform': '002',
+        'watchdisneychannel': '004',
+        'watchdisneyjunior': '008',
+        'watchdisneyxd': '009',
+    }
+    _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/.*?vdka(?P<id>\w+)' % '|'.join(_BRANDS.keys())
+    _TESTS = [{
+        'url': 'http://abc.go.com/shows/castle/video/most-recent/vdka0_g86w5onx',
+        'info_dict': {
+            'id': '0_g86w5onx',
+            'ext': 'mp4',
+            'title': 'Sneak Peek: Language Arts',
+            'description': 'md5:7dcdab3b2d17e5217c953256af964e9c',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://abc.go.com/shows/after-paradise/video/most-recent/vdka3335601',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        sub_domain, video_id = re.match(self._VALID_URL, url).groups()
+        video_data = self._download_json(
+            'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (self._BRANDS[sub_domain], video_id),
+            video_id)['video'][0]
+        title = video_data['title']
+
+        formats = []
+        for asset in video_data.get('assets', {}).get('asset', []):
+            asset_url = asset.get('value')
+            if not asset_url:
+                continue
+            format_id = asset.get('format')
+            ext = determine_ext(asset_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
+            else:
+                formats.append({
+                    'format_id': format_id,
+                    'url': asset_url,
+                    'ext': ext,
+                })
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for cc in video_data.get('closedcaption', {}).get('src', []):
+            cc_url = cc.get('value')
+            if not cc_url:
+                continue
+            ext = determine_ext(cc_url)
+            if ext == 'xml':
+                ext = 'ttml'
+            subtitles.setdefault(cc.get('lang'), []).append({
+                'url': cc_url,
+                'ext': ext,
+            })
+
+        thumbnails = []
+        for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []):
+            thumbnail_url = thumbnail.get('value')
+            if not thumbnail_url:
+                continue
+            thumbnails.append({
+                'url': thumbnail_url,
+                'width': int_or_none(thumbnail.get('width')),
+                'height': int_or_none(thumbnail.get('height')),
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': video_data.get('longdescription') or video_data.get('description'),
+            'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000),
+            'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')),
+            'episode_number': int_or_none(video_data.get('episodenumber')),
+            'series': video_data.get('show', {}).get('title'),
+            'season_number': int_or_none(video_data.get('season', {}).get('num')),
+            'thumbnails': thumbnails,
+            'formats': formats,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/kickstarter.py

@@ -37,7 +37,6 @@ class KickStarterIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Power Drive 2000',
         },
-        'expected_warnings': ['OpenGraph description'],
     }]
 
     def _real_extract(self, url):
@@ -67,6 +66,6 @@ class KickStarterIE(InfoExtractor):
             'id': video_id,
             'url': video_url,
             'title': title,
-            'description': self._og_search_description(webpage),
+            'description': self._og_search_description(webpage, default=None),
             'thumbnail': thumbnail,
         }

youtube_dl/extractor/kusi.py

@@ -18,31 +18,20 @@ from ..utils import (
 class KUSIIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?kusi\.com/(?P<path>story/.+|video\?clipId=(?P<clipId>\d+))'
     _TESTS = [{
-        'url': 'http://www.kusi.com/story/31183873/turko-files-case-closed-put-on-hold',
-        'md5': 'f926e7684294cf8cb7bdf8858e1b3988',
+        'url': 'http://www.kusi.com/story/32849881/turko-files-refused-to-help-it-aint-right',
+        'md5': '4e76ce8e53660ce9697d06c0ba6fc47d',
         'info_dict': {
-            'id': '12203019',
+            'id': '12689020',
             'ext': 'mp4',
-            'title': 'Turko Files: Case Closed! & Put On Hold!',
-            'duration': 231.0,
-            'upload_date': '20160210',
-            'timestamp': 1455087571,
+            'title': "Turko Files: Refused to Help, It Ain't Right!",
+            'duration': 223.586,
+            'upload_date': '20160826',
+            'timestamp': 1472233118,
             'thumbnail': 're:^https?://.*\.jpg$'
         },
     }, {
         'url': 'http://kusi.com/video?clipId=12203019',
-        'info_dict': {
-            'id': '12203019',
-            'ext': 'mp4',
-            'title': 'Turko Files: Case Closed! & Put On Hold!',
-            'duration': 231.0,
-            'upload_date': '20160210',
-            'timestamp': 1455087571,
-            'thumbnail': 're:^https?://.*\.jpg$'
-        },
-        'params': {
-            'skip_download': True,  # Same as previous one
-        },
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/limelight.py

@@ -34,11 +34,12 @@ class LimelightBaseIE(InfoExtractor):
     def _extract_info(self, streams, mobile_urls, properties):
         video_id = properties['media_id']
         formats = []
-
+        urls = []
         for stream in streams:
             stream_url = stream.get('url')
-            if not stream_url or stream.get('drmProtected'):
+            if not stream_url or stream.get('drmProtected') or stream_url in urls:
                 continue
+            urls.append(stream_url)
             ext = determine_ext(stream_url)
             if ext == 'f4m':
                 formats.extend(self._extract_f4m_formats(
@@ -58,9 +59,11 @@ class LimelightBaseIE(InfoExtractor):
                     format_id = 'rtmp'
                     if stream.get('videoBitRate'):
                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
+                    http_url = 'http://%s/%s' % (rtmp.group('host').replace('csl.', 'cpl.'), rtmp.group('playpath')[4:])
+                    urls.append(http_url)
                     http_fmt = fmt.copy()
                     http_fmt.update({
-                        'url': 'http://%s/%s' % (rtmp.group('host').replace('csl.', 'cpl.'), rtmp.group('playpath')[4:]),
+                        'url': http_url,
                         'format_id': format_id.replace('rtmp', 'http'),
                     })
                     formats.append(http_fmt)
@@ -76,8 +79,9 @@ class LimelightBaseIE(InfoExtractor):
         for mobile_url in mobile_urls:
             media_url = mobile_url.get('mobileUrl')
             format_id = mobile_url.get('targetMediaPlatform')
-            if not media_url or format_id == 'Widevine':
+            if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls:
                 continue
+            urls.append(media_url)
             ext = determine_ext(media_url)
             if ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(

youtube_dl/extractor/movingimage.py

@@ -7,22 +7,19 @@ from ..utils import (
 )
 
 
-class SSAIE(InfoExtractor):
-    _VALID_URL = r'https?://ssa\.nls\.uk/film/(?P<id>\d+)'
+class MovingImageIE(InfoExtractor):
+    _VALID_URL = r'https?://movingimage\.nls\.uk/film/(?P<id>\d+)'
     _TEST = {
-        'url': 'http://ssa.nls.uk/film/3561',
+        'url': 'http://movingimage.nls.uk/film/3561',
+        'md5': '4caa05c2b38453e6f862197571a7be2f',
         'info_dict': {
             'id': '3561',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'SHETLAND WOOL',
             'description': 'md5:c5afca6871ad59b4271e7704fe50ab04',
             'duration': 900,
             'thumbnail': 're:^https?://.*\.jpg$',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
     }
 
     def _real_extract(self, url):
@@ -30,10 +27,9 @@ class SSAIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        streamer = self._search_regex(
-            r"'streamer'\s*,\S*'(rtmp[^']+)'", webpage, 'streamer')
-        play_path = self._search_regex(
-            r"'file'\s*,\s*'([^']+)'", webpage, 'file').rpartition('.')[0]
+        formats = self._extract_m3u8_formats(
+            self._html_search_regex(r'file\s*:\s*"([^"]+)"', webpage, 'm3u8 manifest URL'),
+            video_id, ext='mp4', entry_protocol='m3u8_native')
 
         def search_field(field_name, fatal=False):
             return self._search_regex(
@@ -44,13 +40,11 @@ class SSAIE(InfoExtractor):
         description = unescapeHTML(search_field('Description'))
         duration = parse_duration(search_field('Running time'))
         thumbnail = self._search_regex(
-            r"'image'\s*,\s*'([^']+)'", webpage, 'thumbnails', fatal=False)
+            r"image\s*:\s*'([^']+)'", webpage, 'thumbnail', fatal=False)
 
         return {
             'id': video_id,
-            'url': streamer,
-            'play_path': play_path,
-            'ext': 'flv',
+            'formats': formats,
             'title': title,
             'description': description,
             'duration': duration,

youtube_dl/extractor/mtv.py

@@ -4,7 +4,6 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse_urlencode,
     compat_str,
     compat_xpath,
 )
@@ -14,12 +13,13 @@ from ..utils import (
     fix_xml_ampersands,
     float_or_none,
     HEADRequest,
+    RegexNotFoundError,
     sanitized_Request,
     strip_or_none,
     timeconvert,
     unescapeHTML,
+    update_url_query,
     url_basename,
-    RegexNotFoundError,
     xpath_text,
 )
 
@@ -36,6 +36,11 @@ class MTVServicesInfoExtractor(InfoExtractor):
     def _id_from_uri(uri):
         return uri.split(':')[-1]
 
+    @staticmethod
+    def _remove_template_parameter(url):
+        # Remove the templates, like &device={device}
+        return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url)
+
     # This was originally implemented for ComedyCentral, but it also works here
     @classmethod
     def _transform_rtmp_url(cls, rtmp_video_url):
@@ -117,9 +122,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         video_id = self._id_from_uri(uri)
         self.report_extraction(video_id)
         content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
-        mediagen_url = content_el.attrib['url']
-        # Remove the templates, like &device={device}
-        mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
+        mediagen_url = self._remove_template_parameter(content_el.attrib['url'])
         if 'acceptMethods' not in mediagen_url:
             mediagen_url += '&' if '?' in mediagen_url else '?'
             mediagen_url += 'acceptMethods=fms'
@@ -178,12 +181,12 @@ class MTVServicesInfoExtractor(InfoExtractor):
         data = {'uri': uri}
         if self._LANG:
             data['lang'] = self._LANG
-        return compat_urllib_parse_urlencode(data)
+        return data
 
     def _get_videos_info(self, uri):
         video_id = self._id_from_uri(uri)
         feed_url = self._get_feed_url(uri)
-        info_url = feed_url + '?' + self._get_feed_query(uri)
+        info_url = update_url_query(feed_url, self._get_feed_query(uri))
         return self._get_videos_info_from_url(info_url, video_id)
 
     def _get_videos_info_from_url(self, url, video_id):
@@ -256,13 +259,9 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
 
     def _get_feed_url(self, uri):
         video_id = self._id_from_uri(uri)
-        site_id = uri.replace(video_id, '')
-        config_url = ('http://media.mtvnservices.com/pmt-arc/e1/players/{0}/'
-                      'context52/config.xml'.format(site_id))
-        config_doc = self._download_xml(config_url, video_id)
-        feed_node = config_doc.find('.//feed')
-        feed_url = feed_node.text.strip().split('?')[0]
-        return feed_url
+        config = self._download_json(
+            'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id)
+        return self._remove_template_parameter(config['feedWithQueryParams'])
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/myvidster.py

@@ -13,7 +13,7 @@ class MyVidsterIE(InfoExtractor):
             'id': '3685814',
             'title': 'md5:7d8427d6d02c4fbcef50fe269980c749',
             'upload_date': '20141027',
-            'uploader_id': 'utkualp',
+            'uploader': 'utkualp',
             'ext': 'mp4',
             'age_limit': 18,
         },

youtube_dl/extractor/nba.py

@@ -1,25 +1,20 @@
 from __future__ import unicode_literals
 
 import functools
-import os.path
 import re
 
-from .common import InfoExtractor
+from .turner import TurnerBaseIE
 from ..compat import (
     compat_urllib_parse_urlencode,
     compat_urlparse,
 )
 from ..utils import (
-    int_or_none,
     OnDemandPagedList,
-    parse_duration,
     remove_start,
-    xpath_text,
-    xpath_attr,
 )
 
 
-class NBAIE(InfoExtractor):
+class NBAIE(TurnerBaseIE):
     _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?P<path>(?:[^/]+/)+(?P<id>[^?]*?))/?(?:/index\.html)?(?:\?.*)?$'
     _TESTS = [{
         'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
@@ -44,28 +39,30 @@ class NBAIE(InfoExtractor):
         'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
         'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
         'info_dict': {
-            'id': '0041400301-cle-atl-recap',
+            'id': 'channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
             'ext': 'mp4',
             'title': 'Hawks vs. Cavaliers Game 1',
             'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
             'duration': 228,
             'timestamp': 1432134543,
             'upload_date': '20150520',
-        }
+        },
+        'expected_warnings': ['Unable to download f4m manifest'],
     }, {
         'url': 'http://www.nba.com/clippers/news/doc-rivers-were-not-trading-blake',
         'info_dict': {
-            'id': '1455672027478-Doc_Feb16_720',
+            'id': 'teams/clippers/2016/02/17/1455672027478-Doc_Feb16_720.mov-297324',
             'ext': 'mp4',
             'title': 'Practice: Doc Rivers - 2/16/16',
             'description': 'Head Coach Doc Rivers addresses the media following practice.',
-            'upload_date': '20160217',
+            'upload_date': '20160216',
             'timestamp': 1455672000,
         },
         'params': {
             # m3u8 download
             'skip_download': True,
         },
+        'expected_warnings': ['Unable to download f4m manifest'],
     }, {
         'url': 'http://www.nba.com/timberwolves/wiggins-shootaround#',
         'info_dict': {
@@ -80,7 +77,7 @@ class NBAIE(InfoExtractor):
     }, {
         'url': 'http://www.nba.com/timberwolves/wiggins-shootaround#',
         'info_dict': {
-            'id': 'Wigginsmp4',
+            'id': 'teams/timberwolves/2014/12/12/Wigginsmp4-3462601',
             'ext': 'mp4',
             'title': 'Shootaround Access - Dec. 12 | Andrew Wiggins',
             'description': 'Wolves rookie Andrew Wiggins addresses the media after Friday\'s shootaround.',
@@ -92,6 +89,7 @@ class NBAIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
+        'expected_warnings': ['Unable to download f4m manifest'],
     }]
 
     _PAGE_SIZE = 30
@@ -145,53 +143,12 @@ class NBAIE(InfoExtractor):
             if path.startswith('video/teams'):
                 path = 'video/channels/proxy/' + path[6:]
 
-        video_info = self._download_xml('http://www.nba.com/%s.xml' % path, video_id)
-        video_id = os.path.splitext(xpath_text(video_info, 'slug'))[0]
-        title = xpath_text(video_info, 'headline')
-        description = xpath_text(video_info, 'description')
-        duration = parse_duration(xpath_text(video_info, 'length'))
-        timestamp = int_or_none(xpath_attr(video_info, 'dateCreated', 'uts'))
-
-        thumbnails = []
-        for image in video_info.find('images'):
-            thumbnails.append({
-                'id': image.attrib.get('cut'),
-                'url': image.text,
-                'width': int_or_none(image.attrib.get('width')),
-                'height': int_or_none(image.attrib.get('height')),
+        return self._extract_cvp_info(
+            'http://www.nba.com/%s.xml' % path, video_id, {
+                'default': {
+                    'media_src': 'http://nba.cdn.turner.com/nba/big',
+                },
+                'm3u8': {
+                    'media_src': 'http://nbavod-f.akamaihd.net',
+                },
             })
-
-        formats = []
-        for video_file in video_info.findall('.//file'):
-            video_url = video_file.text
-            if video_url.startswith('/'):
-                continue
-            if video_url.endswith('.m3u8'):
-                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:
-                key = video_file.attrib.get('bitrate')
-                format_info = {
-                    'format_id': key,
-                    'url': video_url,
-                }
-                mobj = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key)
-                if mobj:
-                    format_info.update({
-                        'width': int(mobj.group(1)),
-                        'height': int(mobj.group(2)),
-                        'tbr': int_or_none(mobj.group(3)),
-                    })
-                formats.append(format_info)
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'duration': duration,
-            'timestamp': timestamp,
-            'thumbnails': thumbnails,
-            'formats': formats,
-        }

youtube_dl/extractor/nhk.py

@@ -0,0 +1,50 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class NhkVodIE(InfoExtractor):
+    _VALID_URL = r'https?://www3\.nhk\.or\.jp/nhkworld/en/vod/(?P<id>.+?)\.html'
+    _TEST = {
+        # Videos available only for a limited period of time. Visit
+        # http://www3.nhk.or.jp/nhkworld/en/vod/ for working samples.
+        'url': 'http://www3.nhk.or.jp/nhkworld/en/vod/tokyofashion/20160815.html',
+        'info_dict': {
+            'id': 'A1bnNiNTE6nY3jLllS-BIISfcC_PpvF5',
+            'ext': 'flv',
+            'title': 'TOKYO FASHION EXPRESS - The Kimono as Global Fashion',
+            'description': 'md5:db338ee6ce8204f415b754782f819824',
+            'series': 'TOKYO FASHION EXPRESS',
+            'episode': 'The Kimono as Global Fashion',
+        },
+        'skip': 'Videos available only for a limited period of time',
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        embed_code = self._search_regex(
+            r'nw_vod_ooplayer\([^,]+,\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
+            webpage, 'ooyala embed code', group='id')
+
+        title = self._search_regex(
+            r'<div[^>]+class=["\']episode-detail["\']>\s*<h\d+>([^<]+)',
+            webpage, 'title', default=None)
+        description = self._html_search_regex(
+            r'(?s)<p[^>]+class=["\']description["\'][^>]*>(.+?)</p>',
+            webpage, 'description', default=None)
+        series = self._search_regex(
+            r'<h2[^>]+class=["\']detail-top-player-title[^>]+><a[^>]+>([^<]+)',
+            webpage, 'series', default=None)
+
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'Ooyala',
+            'url': 'ooyala:%s' % embed_code,
+            'title': '%s - %s' % (series, title) if series and title else title,
+            'description': description,
+            'series': series,
+            'episode': title,
+        }

youtube_dl/extractor/nick.py

@@ -2,7 +2,6 @@
 from __future__ import unicode_literals
 
 from .mtv import MTVServicesInfoExtractor
-from ..compat import compat_urllib_parse_urlencode
 from ..utils import update_url_query
 
 
@@ -59,10 +58,10 @@ class NickIE(MTVServicesInfoExtractor):
     }]
 
     def _get_feed_query(self, uri):
-        return compat_urllib_parse_urlencode({
+        return {
             'feed': 'nick_arc_player_prime',
             'mgid': uri,
-        })
+        }
 
     def _extract_mgid(self, webpage):
         return self._search_regex(r'data-contenturi="([^"]+)', webpage, 'mgid')

youtube_dl/extractor/ninecninemedia.py

@@ -4,40 +4,36 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     parse_iso8601,
-    parse_duration,
-    ExtractorError
+    float_or_none,
+    ExtractorError,
+    int_or_none,
 )
 
 
-class NineCNineMediaIE(InfoExtractor):
-    _VALID_URL = r'9c9media:(?P<destination_code>[^:]+):(?P<id>\d+)'
+class NineCNineMediaBaseIE(InfoExtractor):
+    _API_BASE_TEMPLATE = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/'
+
+
+class NineCNineMediaStackIE(NineCNineMediaBaseIE):
+    IE_NAME = '9c9media:stack'
+    _VALID_URL = r'9c9media:stack:(?P<destination_code>[^:]+):(?P<content_id>\d+):(?P<content_package>\d+):(?P<id>\d+)'
 
     def _real_extract(self, url):
-        destination_code, video_id = re.match(self._VALID_URL, url).groups()
-        api_base_url = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/' % (destination_code, video_id)
-        content = self._download_json(api_base_url, video_id, query={
-            '$include': '[contentpackages]',
-        })
-        title = content['Name']
-        if len(content['ContentPackages']) > 1:
-            raise ExtractorError('multiple content packages')
-        content_package = content['ContentPackages'][0]
-        stacks_base_url = api_base_url + 'contentpackages/%s/stacks/' % content_package['Id']
-        stacks = self._download_json(stacks_base_url, video_id)['Items']
-        if len(stacks) > 1:
-            raise ExtractorError('multiple stacks')
-        stack = stacks[0]
-        stack_base_url = '%s%s/manifest.' % (stacks_base_url, stack['Id'])
+        destination_code, content_id, package_id, stack_id = re.match(self._VALID_URL, url).groups()
+        stack_base_url_template = self._API_BASE_TEMPLATE + 'contentpackages/%s/stacks/%s/manifest.'
+        stack_base_url = stack_base_url_template % (destination_code, content_id, package_id, stack_id)
+
         formats = []
         formats.extend(self._extract_m3u8_formats(
-            stack_base_url + 'm3u8', video_id, 'mp4',
+            stack_base_url + 'm3u8', stack_id, 'mp4',
             'm3u8_native', m3u8_id='hls', fatal=False))
         formats.extend(self._extract_f4m_formats(
-            stack_base_url + 'f4m', video_id,
+            stack_base_url + 'f4m', stack_id,
             f4m_id='hds', fatal=False))
-        mp4_url = self._download_webpage(stack_base_url + 'pd', video_id, fatal=False)
+        mp4_url = self._download_webpage(stack_base_url + 'pd', stack_id, fatal=False)
         if mp4_url:
             formats.append({
                 'url': mp4_url,
@@ -46,10 +42,86 @@ class NineCNineMediaIE(InfoExtractor):
         self._sort_formats(formats)
 
         return {
-            'id': video_id,
-            'title': title,
-            'description': content.get('Desc') or content.get('ShortDesc'),
-            'timestamp': parse_iso8601(content.get('BroadcastDateTime')),
-            'duration': parse_duration(content.get('BroadcastTime')),
+            'id': stack_id,
             'formats': formats,
         }
+
+
+class NineCNineMediaIE(NineCNineMediaBaseIE):
+    IE_NAME = '9c9media'
+    _VALID_URL = r'9c9media:(?P<destination_code>[^:]+):(?P<id>\d+)'
+
+    def _real_extract(self, url):
+        destination_code, content_id = re.match(self._VALID_URL, url).groups()
+        api_base_url = self._API_BASE_TEMPLATE % (destination_code, content_id)
+        content = self._download_json(api_base_url, content_id, query={
+            '$include': '[Media,Season,ContentPackages]',
+        })
+        title = content['Name']
+        if len(content['ContentPackages']) > 1:
+            raise ExtractorError('multiple content packages')
+        content_package = content['ContentPackages'][0]
+        package_id = content_package['Id']
+        content_package_url = api_base_url + 'contentpackages/%s/' % package_id
+        content_package = self._download_json(content_package_url, content_id)
+
+        if content_package.get('Constraints', {}).get('Security', {}).get('Type') == 'adobe-drm':
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        stacks = self._download_json(content_package_url + 'stacks/', package_id)['Items']
+        multistacks = len(stacks) > 1
+
+        thumbnails = []
+        for image in content.get('Images', []):
+            image_url = image.get('Url')
+            if not image_url:
+                continue
+            thumbnails.append({
+                'url': image_url,
+                'width': int_or_none(image.get('Width')),
+                'height': int_or_none(image.get('Height')),
+            })
+
+        tags, categories = [], []
+        for source_name, container in (('Tags', tags), ('Genres', categories)):
+            for e in content.get(source_name, []):
+                e_name = e.get('Name')
+                if not e_name:
+                    continue
+                container.append(e_name)
+
+        description = content.get('Desc') or content.get('ShortDesc')
+        season = content.get('Season', {})
+        base_info = {
+            'description': description,
+            'timestamp': parse_iso8601(content.get('BroadcastDateTime')),
+            'episode_number': int_or_none(content.get('Episode')),
+            'season': season.get('Name'),
+            'season_number': season.get('Number'),
+            'season_id': season.get('Id'),
+            'series': content.get('Media', {}).get('Name'),
+            'tags': tags,
+            'categories': categories,
+        }
+
+        entries = []
+        for stack in stacks:
+            stack_id = compat_str(stack['Id'])
+            entry = {
+                '_type': 'url_transparent',
+                'url': '9c9media:stack:%s:%s:%s:%s' % (destination_code, content_id, package_id, stack_id),
+                'id': stack_id,
+                'title': '%s_part%s' % (title, stack['Name']) if multistacks else title,
+                'duration': float_or_none(stack.get('Duration')),
+                'ie_key': 'NineCNineMediaStack',
+            }
+            entry.update(base_info)
+            entries.append(entry)
+
+        return {
+            '_type': 'multi_video',
+            'id': content_id,
+            'title': title,
+            'description': description,
+            'entries': entries,
+        }

youtube_dl/extractor/nytimes.py

@@ -1,26 +1,37 @@
 from __future__ import unicode_literals
 
+import hmac
+import hashlib
+import base64
+
 from .common import InfoExtractor
 from ..utils import (
     float_or_none,
     int_or_none,
     parse_iso8601,
+    mimetype2ext,
+    determine_ext,
 )
 
 
 class NYTimesBaseIE(InfoExtractor):
+    _SECRET = b'pX(2MbU2);4N{7J8)>YwKRJ+/pQ3JkiU2Q^V>mFYv6g6gYvt6v'
+
     def _extract_video_from_id(self, video_id):
-        video_data = self._download_json(
-            'http://www.nytimes.com/svc/video/api/v2/video/%s' % video_id,
-            video_id, 'Downloading video JSON')
+        # Authorization generation algorithm is reverse engineered from `signer` in
+        # http://graphics8.nytimes.com/video/vhs/vhs-2.x.min.js
+        path = '/svc/video/api/v3/video/' + video_id
+        hm = hmac.new(self._SECRET, (path + ':vhs').encode(), hashlib.sha512).hexdigest()
+        video_data = self._download_json('http://www.nytimes.com' + path, video_id, 'Downloading video JSON', headers={
+            'Authorization': 'NYTV ' + base64.b64encode(hm.encode()).decode(),
+            'X-NYTV': 'vhs',
+        }, fatal=False)
+        if not video_data:
+            video_data = self._download_json(
+                'http://www.nytimes.com/svc/video/api/v2/video/' + video_id,
+                video_id, 'Downloading video JSON')
 
         title = video_data['headline']
-        description = video_data.get('summary')
-        duration = float_or_none(video_data.get('duration'), 1000)
-
-        uploader = video_data.get('byline')
-        publication_date = video_data.get('publication_date')
-        timestamp = parse_iso8601(publication_date[:-8]) if publication_date else None
 
         def get_file_size(file_size):
             if isinstance(file_size, int):
@@ -28,35 +39,59 @@ class NYTimesBaseIE(InfoExtractor):
             elif isinstance(file_size, dict):
                 return int(file_size.get('value', 0))
             else:
-                return 0
+                return None
 
-        formats = [
-            {
-                'url': video['url'],
-                'format_id': video.get('type'),
-                'vcodec': video.get('video_codec'),
-                'width': int_or_none(video.get('width')),
-                'height': int_or_none(video.get('height')),
-                'filesize': get_file_size(video.get('fileSize')),
-            } for video in video_data['renditions'] if video.get('url')
-        ]
+        urls = []
+        formats = []
+        for video in video_data.get('renditions', []):
+            video_url = video.get('url')
+            format_id = video.get('type')
+            if not video_url or format_id == 'thumbs' or video_url in urls:
+                continue
+            urls.append(video_url)
+            ext = mimetype2ext(video.get('mimetype')) or determine_ext(video_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id=format_id or 'hls', fatal=False))
+            elif ext == 'mpd':
+                continue
+            #     formats.extend(self._extract_mpd_formats(
+            #         video_url, video_id, format_id or 'dash', fatal=False))
+            else:
+                formats.append({
+                    'url': video_url,
+                    'format_id': format_id,
+                    'vcodec': video.get('videoencoding') or video.get('video_codec'),
+                    'width': int_or_none(video.get('width')),
+                    'height': int_or_none(video.get('height')),
+                    'filesize': get_file_size(video.get('file_size') or video.get('fileSize')),
+                    'tbr': int_or_none(video.get('bitrate'), 1000),
+                    'ext': ext,
+                })
         self._sort_formats(formats)
 
-        thumbnails = [
-            {
-                'url': 'http://www.nytimes.com/%s' % image['url'],
+        thumbnails = []
+        for image in video_data.get('images', []):
+            image_url = image.get('url')
+            if not image_url:
+                continue
+            thumbnails.append({
+                'url': 'http://www.nytimes.com/' + image_url,
                 'width': int_or_none(image.get('width')),
                 'height': int_or_none(image.get('height')),
-            } for image in video_data.get('images', []) if image.get('url')
-        ]
+            })
+
+        publication_date = video_data.get('publication_date')
+        timestamp = parse_iso8601(publication_date[:-8]) if publication_date else None
 
         return {
             'id': video_id,
             'title': title,
-            'description': description,
+            'description': video_data.get('summary'),
             'timestamp': timestamp,
-            'uploader': uploader,
-            'duration': duration,
+            'uploader': video_data.get('byline'),
+            'duration': float_or_none(video_data.get('duration'), 1000),
             'formats': formats,
             'thumbnails': thumbnails,
         }
@@ -67,7 +102,7 @@ class NYTimesIE(NYTimesBaseIE):
 
     _TESTS = [{
         'url': 'http://www.nytimes.com/video/opinion/100000002847155/verbatim-what-is-a-photocopier.html?playlistId=100000001150263',
-        'md5': '18a525a510f942ada2720db5f31644c0',
+        'md5': 'd665342765db043f7e225cff19df0f2d',
         'info_dict': {
             'id': '100000002847155',
             'ext': 'mov',

youtube_dl/extractor/openload.py

@@ -55,10 +55,12 @@ class OpenloadIE(InfoExtractor):
 
         video_url_chars = []
 
-        for c in enc_data:
+        for idx, c in enumerate(enc_data):
             j = compat_ord(c)
             if j >= 33 and j <= 126:
                 j = ((j + 14) % 94) + 33
+            if idx == len(enc_data) - 1:
+                j += 1
             video_url_chars += compat_chr(j)
 
         video_url = 'https://openload.co/stream/%s?mime=true' % ''.join(video_url_chars)

youtube_dl/extractor/periscope.py

@@ -8,7 +8,14 @@ from ..utils import (
 )
 
 
-class PeriscopeIE(InfoExtractor):
+class PeriscopeBaseIE(InfoExtractor):
+    def _call_api(self, method, query, item_id):
+        return self._download_json(
+            'https://api.periscope.tv/api/v2/%s' % method,
+            item_id, query=query)
+
+
+class PeriscopeIE(PeriscopeBaseIE):
     IE_DESC = 'Periscope'
     IE_NAME = 'periscope'
     _VALID_URL = r'https?://(?:www\.)?periscope\.tv/[^/]+/(?P<id>[^/?#]+)'
@@ -34,14 +41,11 @@ class PeriscopeIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    def _call_api(self, method, value):
-        return self._download_json(
-            'https://api.periscope.tv/api/v2/%s?broadcast_id=%s' % (method, value), value)
-
     def _real_extract(self, url):
         token = self._match_id(url)
 
-        broadcast_data = self._call_api('getBroadcastPublic', token)
+        broadcast_data = self._call_api(
+            'getBroadcastPublic', {'broadcast_id': token}, token)
         broadcast = broadcast_data['broadcast']
         status = broadcast['status']
 
@@ -61,7 +65,8 @@ class PeriscopeIE(InfoExtractor):
             'url': broadcast[image],
         } for image in ('image_url', 'image_url_small') if broadcast.get(image)]
 
-        stream = self._call_api('getAccessPublic', token)
+        stream = self._call_api(
+            'getAccessPublic', {'broadcast_id': token}, token)
 
         formats = []
         for format_id in ('replay', 'rtmp', 'hls', 'https_hls'):
@@ -88,7 +93,7 @@ class PeriscopeIE(InfoExtractor):
         }
 
 
-class PeriscopeUserIE(InfoExtractor):
+class PeriscopeUserIE(PeriscopeBaseIE):
     _VALID_URL = r'https?://www\.periscope\.tv/(?P<id>[^/]+)/?$'
     IE_DESC = 'Periscope user videos'
     IE_NAME = 'periscope:user'
@@ -106,26 +111,34 @@ class PeriscopeUserIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        user_id = self._match_id(url)
+        user_name = self._match_id(url)
 
-        webpage = self._download_webpage(url, user_id)
+        webpage = self._download_webpage(url, user_name)
 
         data_store = self._parse_json(
             unescapeHTML(self._search_regex(
                 r'data-store=(["\'])(?P<data>.+?)\1',
                 webpage, 'data store', default='{}', group='data')),
-            user_id)
+            user_name)
 
-        user = data_store.get('User', {}).get('user', {})
-        title = user.get('display_name') or user.get('username')
+        user = list(data_store['UserCache']['users'].values())[0]['user']
+        user_id = user['id']
+        session_id = data_store['SessionToken']['broadcastHistory']['token']['session_id']
+
+        broadcasts = self._call_api(
+            'getUserBroadcastsPublic',
+            {'user_id': user_id, 'session_id': session_id},
+            user_name)['broadcasts']
+
+        broadcast_ids = [
+            broadcast['id'] for broadcast in broadcasts if broadcast.get('id')]
+
+        title = user.get('display_name') or user.get('username') or user_name
         description = user.get('description')
 
-        broadcast_ids = (data_store.get('UserBroadcastHistory', {}).get('broadcastIds') or
-                         data_store.get('BroadcastCache', {}).get('broadcastIds', []))
-
         entries = [
             self.url_result(
-                'https://www.periscope.tv/%s/%s' % (user_id, broadcast_id))
+                'https://www.periscope.tv/%s/%s' % (user_name, broadcast_id))
             for broadcast_id in broadcast_ids]
 
         return self.playlist_result(entries, user_id, title, description)

youtube_dl/extractor/played.py

@@ -1,60 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-import os.path
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    sanitized_Request,
-    urlencode_postdata,
-)
-
-
-class PlayedIE(InfoExtractor):
-    IE_NAME = 'played.to'
-    _VALID_URL = r'https?://(?:www\.)?played\.to/(?P<id>[a-zA-Z0-9_-]+)'
-
-    _TEST = {
-        'url': 'http://played.to/j2f2sfiiukgt',
-        'md5': 'c2bd75a368e82980e7257bf500c00637',
-        'info_dict': {
-            'id': 'j2f2sfiiukgt',
-            'ext': 'flv',
-            'title': 'youtube-dl_test_video.mp4',
-        },
-        'skip': 'Removed for copyright infringement.',  # oh wow
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        orig_webpage = self._download_webpage(url, video_id)
-
-        m_error = re.search(
-            r'(?s)Reason for deletion:.*?<b class="err"[^>]*>(?P<msg>[^<]+)</b>', orig_webpage)
-        if m_error:
-            raise ExtractorError(m_error.group('msg'), expected=True)
-
-        data = self._hidden_inputs(orig_webpage)
-
-        self._sleep(2, video_id)
-
-        post = urlencode_postdata(data)
-        headers = {
-            b'Content-Type': b'application/x-www-form-urlencoded',
-        }
-        req = sanitized_Request(url, post, headers)
-        webpage = self._download_webpage(
-            req, video_id, note='Downloading video page ...')
-
-        title = os.path.splitext(data['fname'])[0]
-
-        video_url = self._search_regex(
-            r'file: "?(.+?)",', webpage, 'video URL')
-
-        return {
-            'id': video_id,
-            'title': title,
-            'url': video_url,
-        }

youtube_dl/extractor/porncom.py

@@ -26,6 +26,8 @@ class PornComIE(InfoExtractor):
             'duration': 551,
             'view_count': int,
             'age_limit': 18,
+            'categories': list,
+            'tags': list,
         },
     }, {
         'url': 'http://se.porn.com/videos/marsha-may-rides-seth-on-top-of-his-thick-cock-2658067',
@@ -75,7 +77,14 @@ class PornComIE(InfoExtractor):
         self._sort_formats(formats)
 
         view_count = str_to_int(self._search_regex(
-            r'class=["\']views["\'][^>]*><p>([\d,.]+)', webpage, 'view count'))
+            r'class=["\']views["\'][^>]*><p>([\d,.]+)', webpage,
+            'view count', fatal=False))
+
+        def extract_list(kind):
+            s = self._search_regex(
+                r'(?s)<p[^>]*>%s:(.+?)</p>' % kind.capitalize(),
+                webpage, kind, fatal=False)
+            return re.findall(r'<a[^>]+>([^<]+)</a>', s or '')
 
         return {
             'id': video_id,
@@ -86,4 +95,6 @@ class PornComIE(InfoExtractor):
             'view_count': view_count,
             'formats': formats,
             'age_limit': 18,
+            'categories': extract_list('categories'),
+            'tags': extract_list('tags'),
         }

youtube_dl/extractor/pyvideo.py

@@ -1,59 +1,72 @@
 from __future__ import unicode_literals
 
 import re
-import os
 
 from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import int_or_none
 
 
 class PyvideoIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?pyvideo\.org/video/(?P<id>\d+)/(.*)'
+    _VALID_URL = r'https?://(?:www\.)?pyvideo\.org/(?P<category>[^/]+)/(?P<id>[^/?#&.]+)'
 
-    _TESTS = [
-        {
-            'url': 'http://pyvideo.org/video/1737/become-a-logging-expert-in-30-minutes',
-            'md5': '520915673e53a5c5d487c36e0c4d85b5',
-            'info_dict': {
-                'id': '24_4WWkSmNo',
-                'ext': 'webm',
-                'title': 'Become a logging expert in 30 minutes',
-                'description': 'md5:9665350d466c67fb5b1598de379021f7',
-                'upload_date': '20130320',
-                'uploader': 'Next Day Video',
-                'uploader_id': 'NextDayVideo',
-            },
-            'add_ie': ['Youtube'],
+    _TESTS = [{
+        'url': 'http://pyvideo.org/pycon-us-2013/become-a-logging-expert-in-30-minutes.html',
+        'info_dict': {
+            'id': 'become-a-logging-expert-in-30-minutes',
         },
-        {
-            'url': 'http://pyvideo.org/video/2542/gloriajw-spotifywitherikbernhardsson182m4v',
-            'md5': '5fe1c7e0a8aa5570330784c847ff6d12',
-            'info_dict': {
-                'id': '2542',
-                'ext': 'm4v',
-                'title': 'Gloriajw-SpotifyWithErikBernhardsson182',
-            },
+        'playlist_count': 2,
+    }, {
+        'url': 'http://pyvideo.org/pygotham-2012/gloriajw-spotifywitherikbernhardsson182m4v.html',
+        'md5': '5fe1c7e0a8aa5570330784c847ff6d12',
+        'info_dict': {
+            'id': '2542',
+            'ext': 'm4v',
+            'title': 'Gloriajw-SpotifyWithErikBernhardsson182.m4v',
         },
-    ]
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
+        category = mobj.group('category')
         video_id = mobj.group('id')
 
-        webpage = self._download_webpage(url, video_id)
+        entries = []
 
-        m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', webpage)
-        if m_youtube is not None:
-            return self.url_result(m_youtube.group(1), 'Youtube')
+        data = self._download_json(
+            'https://raw.githubusercontent.com/pyvideo/data/master/%s/videos/%s.json'
+            % (category, video_id), video_id, fatal=False)
 
-        title = self._html_search_regex(
-            r'<div class="section">\s*<h3(?:\s+class="[^"]*"[^>]*)?>([^>]+?)</h3>',
-            webpage, 'title', flags=re.DOTALL)
-        video_url = self._search_regex(
-            [r'<source src="(.*?)"', r'<dt>Download</dt>.*?<a href="(.+?)"'],
-            webpage, 'video url', flags=re.DOTALL)
+        if data:
+            for video in data['videos']:
+                video_url = video.get('url')
+                if video_url:
+                    if video.get('type') == 'youtube':
+                        entries.append(self.url_result(video_url, 'Youtube'))
+                    else:
+                        entries.append({
+                            'id': compat_str(data.get('id') or video_id),
+                            'url': video_url,
+                            'title': data['title'],
+                            'description': data.get('description') or data.get('summary'),
+                            'thumbnail': data.get('thumbnail_url'),
+                            'duration': int_or_none(data.get('duration')),
+                        })
+        else:
+            webpage = self._download_webpage(url, video_id)
+            title = self._og_search_title(webpage)
+            media_urls = self._search_regex(
+                r'(?s)Media URL:(.+?)</li>', webpage, 'media urls')
+            for m in re.finditer(
+                    r'<a[^>]+href=(["\'])(?P<url>http.+?)\1', media_urls):
+                media_url = m.group('url')
+                if re.match(r'https?://www\.youtube\.com/watch\?v=.*', media_url):
+                    entries.append(self.url_result(media_url, 'Youtube'))
+                else:
+                    entries.append({
+                        'id': video_id,
+                        'url': media_url,
+                        'title': title,
+                    })
 
-        return {
-            'id': video_id,
-            'title': os.path.splitext(title)[0],
-            'url': video_url,
-        }
+        return self.playlist_result(entries, video_id)

youtube_dl/extractor/soundcloud.py

@@ -32,7 +32,7 @@ class SoundcloudIE(InfoExtractor):
     _VALID_URL = r'''(?x)^(?:https?://)?
                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
                             (?P<uploader>[\w\d-]+)/
-                            (?!(?:tracks|sets(?:/[^/?#]+)?|reposts|likes|spotlight)/?(?:$|[?#]))
+                            (?!(?:tracks|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#]))
                             (?P<title>[\w\d-]+)/?
                             (?P<token>[^?]+?)?(?:[?].*)?$)
                        |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+)
@@ -265,6 +265,9 @@ class SoundcloudSetIE(SoundcloudIE):
             'title': 'The Royal Concept EP',
         },
         'playlist_mincount': 6,
+    }, {
+        'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/southpark.py

@@ -35,6 +35,7 @@ class SouthParkEsIE(SouthParkIE):
             'description': 'Cartman Consigue Una Sonda Anal',
         },
         'playlist_count': 4,
+        'skip': 'Geo-restricted',
     }]
 
 

youtube_dl/extractor/spankbang.py

@@ -14,7 +14,7 @@ class SpankBangIE(InfoExtractor):
             'id': '3vvn',
             'ext': 'mp4',
             'title': 'fantasy solo',
-            'description': 'dillion harper masturbates on a bed',
+            'description': 'Watch fantasy solo free HD porn video - 05 minutes - dillion harper masturbates on a bed free adult movies.',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': 'silly2587',
             'age_limit': 18,
@@ -44,12 +44,10 @@ class SpankBangIE(InfoExtractor):
 
         title = self._html_search_regex(
             r'(?s)<h1[^>]*>(.+?)</h1>', webpage, 'title')
-        description = self._search_regex(
-            r'class="desc"[^>]*>([^<]+)',
-            webpage, 'description', default=None)
+        description = self._og_search_description(webpage)
         thumbnail = self._og_search_thumbnail(webpage)
         uploader = self._search_regex(
-            r'class="user"[^>]*>([^<]+)',
+            r'class="user"[^>]*><img[^>]+>([^<]+)',
             webpage, 'uploader', fatal=False)
 
         age_limit = self._rta_search(webpage)

youtube_dl/extractor/tbs.py

@@ -0,0 +1,59 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .turner import TurnerBaseIE
+from ..utils import (
+    extract_attributes,
+    ExtractorError,
+)
+
+
+class TBSIE(TurnerBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?(?P<site>tbs|tntdrama)\.com/videos/(?:[^/]+/)+(?P<id>[^/?#]+)\.html'
+    _TESTS = [{
+        'url': 'http://www.tbs.com/videos/people-of-earth/season-1/extras/2007318/theatrical-trailer.html',
+        'md5': '9e61d680e2285066ade7199e6408b2ee',
+        'info_dict': {
+            'id': '2007318',
+            'ext': 'mp4',
+            'title': 'Theatrical Trailer',
+            'description': 'Catch the latest comedy from TBS, People of Earth, premiering Halloween night--Monday, October 31, at 9/8c.',
+        }
+    }, {
+        'url': 'http://www.tntdrama.com/videos/good-behavior/season-1/extras/1538823/you-better-run.html',
+        'md5': 'ce53c6ead5e9f3280b4ad2031a6fab56',
+        'info_dict': {
+            'id': '1538823',
+            'ext': 'mp4',
+            'title': 'You Better Run',
+            'description': 'Letty Raines must figure out what she\'s running toward while running away from her past. Good Behavior premieres November 15 at 9/8c.',
+        }
+    }]
+
+    def _real_extract(self, url):
+        domain, display_id = re.match(self._VALID_URL, url).groups()
+        site = domain[:3]
+        webpage = self._download_webpage(url, display_id)
+        video_params = extract_attributes(self._search_regex(r'(<[^>]+id="page-video"[^>]*>)', webpage, 'video params'))
+        if video_params.get('isAuthRequired') == 'true':
+            raise ExtractorError(
+                'This video is only available via cable service provider subscription that'
+                ' is not currently supported.', expected=True)
+        query = None
+        clip_id = video_params.get('clipid')
+        if clip_id:
+            query = 'id=' + clip_id
+        else:
+            query = 'titleId=' + video_params['titleid']
+        return self._extract_cvp_info(
+            'http://www.%s.com/service/cvpXml?%s' % (domain, query), display_id, {
+                'default': {
+                    'media_src': 'http://ht.cdn.turner.com/%s/big' % site,
+                },
+                'secure': {
+                    'media_src': 'http://apple-secure.cdn.turner.com/%s/big' % site,
+                    'tokenizer_src': 'http://www.%s.com/video/processors/services/token_ipadAdobe.do' % domain,
+                },
+            })

youtube_dl/extractor/thestar.py

@@ -2,8 +2,6 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from .brightcove import BrightcoveLegacyIE
-from ..compat import compat_parse_qs
 
 
 class TheStarIE(InfoExtractor):
@@ -30,6 +28,9 @@ class TheStarIE(InfoExtractor):
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
-        brightcove_legacy_url = BrightcoveLegacyIE._extract_brightcove_url(webpage)
-        brightcove_id = compat_parse_qs(brightcove_legacy_url)['@videoPlayer'][0]
-        return self.url_result(self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew', brightcove_id)
+        brightcove_id = self._search_regex(
+            r'mainartBrightcoveVideoId["\']?\s*:\s*["\']?(\d+)',
+            webpage, 'brightcove id')
+        return self.url_result(
+            self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id,
+            'BrightcoveNew', brightcove_id)

youtube_dl/extractor/thvideo.py

@@ -1,84 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    unified_strdate
-)
-
-
-class THVideoIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?thvideo\.tv/(?:v/th|mobile\.php\?cid=)(?P<id>[0-9]+)'
-    _TEST = {
-        'url': 'http://thvideo.tv/v/th1987/',
-        'md5': 'fa107b1f73817e325e9433505a70db50',
-        'info_dict': {
-            'id': '1987',
-            'ext': 'mp4',
-            'title': '【动画】秘封活动记录 ～ The Sealed Esoteric History.分镜稿预览',
-            'display_id': 'th1987',
-            'thumbnail': 'http://thvideo.tv/uploadfile/2014/0722/20140722013459856.jpg',
-            'description': '社团京都幻想剧团的第一个东方二次同人动画作品「秘封活动记录 ～ The Sealed Esoteric History.」 本视频是该动画第一期的分镜草稿...',
-            'upload_date': '20140722'
-        }
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        # extract download link from mobile player page
-        webpage_player = self._download_webpage(
-            'http://thvideo.tv/mobile.php?cid=%s-0' % (video_id),
-            video_id, note='Downloading video source page')
-        video_url = self._html_search_regex(
-            r'<source src="(.*?)" type', webpage_player, 'video url')
-
-        # extract video info from main page
-        webpage = self._download_webpage(
-            'http://thvideo.tv/v/th%s' % (video_id), video_id)
-        title = self._og_search_title(webpage)
-        display_id = 'th%s' % video_id
-        thumbnail = self._og_search_thumbnail(webpage)
-        description = self._og_search_description(webpage)
-        upload_date = unified_strdate(self._html_search_regex(
-            r'span itemprop="datePublished" content="(.*?)">', webpage,
-            'upload date', fatal=False))
-
-        return {
-            'id': video_id,
-            'ext': 'mp4',
-            'url': video_url,
-            'title': title,
-            'display_id': display_id,
-            'thumbnail': thumbnail,
-            'description': description,
-            'upload_date': upload_date
-        }
-
-
-class THVideoPlaylistIE(InfoExtractor):
-    _VALID_URL = r'http?://(?:www\.)?thvideo\.tv/mylist(?P<id>[0-9]+)'
-    _TEST = {
-        'url': 'http://thvideo.tv/mylist2',
-        'info_dict': {
-            'id': '2',
-            'title': '幻想万華鏡',
-        },
-        'playlist_mincount': 23,
-    }
-
-    def _real_extract(self, url):
-        playlist_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, playlist_id)
-        list_title = self._html_search_regex(
-            r'<h1 class="show_title">(.*?)<b id', webpage, 'playlist title',
-            fatal=False)
-
-        entries = [
-            self.url_result('http://thvideo.tv/v/th' + id, 'THVideo')
-            for id in re.findall(r'<dd><a href="http://thvideo.tv/v/th(\d+)/" target=', webpage)]
-
-        return self.playlist_result(entries, playlist_id, list_title)

youtube_dl/extractor/tnaflix.py

@@ -10,6 +10,7 @@ from ..utils import (
     int_or_none,
     parse_duration,
     str_to_int,
+    unescapeHTML,
     xpath_text,
 )
 
@@ -80,7 +81,8 @@ class TNAFlixNetworkBaseIE(InfoExtractor):
 
         if not cfg_url:
             inputs = self._hidden_inputs(webpage)
-            cfg_url = 'https://cdn-fck.tnaflix.com/tnaflix/%s.fid?key=%s' % (inputs['vkey'], inputs['nkey'])
+            cfg_url = ('https://cdn-fck.tnaflix.com/tnaflix/%s.fid?key=%s&VID=%s&premium=1&vip=1&alpha'
+                       % (inputs['vkey'], inputs['nkey'], video_id))
 
         cfg_xml = self._download_xml(
             cfg_url, display_id, 'Downloading metadata',
@@ -89,7 +91,7 @@ class TNAFlixNetworkBaseIE(InfoExtractor):
         formats = []
 
         def extract_video_url(vl):
-            return re.sub('speed=\d+', 'speed=', vl.text)
+            return re.sub('speed=\d+', 'speed=', unescapeHTML(vl.text))
 
         video_link = cfg_xml.find('./videoLink')
         if video_link is not None:
@@ -201,7 +203,7 @@ class TNAFlixIE(TNAFlixNetworkBaseIE):
     _TESTS = [{
         # anonymous uploader, no categories
         'url': 'http://www.tnaflix.com/porn-stars/Carmella-Decesare-striptease/video553878',
-        'md5': '7e569419fe6d69543d01e6be22f5f7c4',
+        'md5': 'ecf3498417d09216374fc5907f9c6ec0',
         'info_dict': {
             'id': '553878',
             'display_id': 'Carmella-Decesare-striptease',
@@ -215,11 +217,11 @@ class TNAFlixIE(TNAFlixNetworkBaseIE):
     }, {
         # non-anonymous uploader, categories
         'url': 'https://www.tnaflix.com/teen-porn/Educational-xxx-video/video6538',
-        'md5': 'fcba2636572895aba116171a899a5658',
+        'md5': '0f5d4d490dbfd117b8607054248a07c0',
         'info_dict': {
             'id': '6538',
             'display_id': 'Educational-xxx-video',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Educational xxx video',
             'description': 'md5:b4fab8f88a8621c8fabd361a173fe5b8',
             'thumbnail': 're:https?://.*\.jpg$',

youtube_dl/extractor/trutube.py

@@ -1,26 +0,0 @@
-from __future__ import unicode_literals
-
-from .nuevo import NuevoBaseIE
-
-
-class TruTubeIE(NuevoBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?trutube\.tv/(?:video/|nuevo/player/embed\.php\?v=)(?P<id>\d+)'
-    _TESTS = [{
-        'url': 'http://trutube.tv/video/14880/Ramses-II-Proven-To-Be-A-Red-Headed-Caucasoid-',
-        'md5': 'c5b6e301b0a2040b074746cbeaa26ca1',
-        'info_dict': {
-            'id': '14880',
-            'ext': 'flv',
-            'title': 'Ramses II - Proven To Be A Red Headed Caucasoid',
-            'thumbnail': 're:^http:.*\.jpg$',
-        }
-    }, {
-        'url': 'https://trutube.tv/nuevo/player/embed.php?v=14880',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        return self._extract_nuevo(
-            'https://trutube.tv/nuevo/player/config.php?v=%s' % video_id,
-            video_id)

youtube_dl/extractor/turner.py

@@ -0,0 +1,178 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    xpath_text,
+    int_or_none,
+    determine_ext,
+    parse_duration,
+    xpath_attr,
+    update_url_query,
+    compat_urlparse,
+)
+
+
+class TurnerBaseIE(InfoExtractor):
+    def _extract_timestamp(self, video_data):
+        return int_or_none(xpath_attr(video_data, 'dateCreated', 'uts'))
+
+    def _extract_cvp_info(self, data_src, video_id, path_data={}):
+        video_data = self._download_xml(data_src, video_id)
+        video_id = video_data.attrib['id']
+        title = xpath_text(video_data, 'headline', fatal=True)
+        # rtmp_src = xpath_text(video_data, 'akamai/src')
+        # if rtmp_src:
+        #     splited_rtmp_src = rtmp_src.split(',')
+        #     if len(splited_rtmp_src) == 2:
+        #         rtmp_src = splited_rtmp_src[1]
+        # aifp = xpath_text(video_data, 'akamai/aifp', default='')
+
+        tokens = {}
+        urls = []
+        formats = []
+        rex = re.compile(
+            r'(?P<width>[0-9]+)x(?P<height>[0-9]+)(?:_(?P<bitrate>[0-9]+))?')
+        # Possible formats locations: files/file, files/groupFiles/files
+        # and maybe others
+        for video_file in video_data.findall('.//file'):
+            video_url = video_file.text.strip()
+            if not video_url:
+                continue
+            ext = determine_ext(video_url)
+            if video_url.startswith('/mp4:protected/'):
+                continue
+                # TODO Correct extraction for these files
+                # protected_path_data = path_data.get('protected')
+                # if not protected_path_data or not rtmp_src:
+                #     continue
+                # protected_path = self._search_regex(
+                #     r'/mp4:(.+)\.[a-z0-9]', video_url, 'secure path')
+                # auth = self._download_webpage(
+                #     protected_path_data['tokenizer_src'], query={
+                #         'path': protected_path,
+                #         'videoId': video_id,
+                #         'aifp': aifp,
+                #     })
+                # token = xpath_text(auth, 'token')
+                # if not token:
+                #     continue
+                # video_url = rtmp_src + video_url + '?' + token
+            elif video_url.startswith('/secure/'):
+                secure_path_data = path_data.get('secure')
+                if not secure_path_data:
+                    continue
+                video_url = secure_path_data['media_src'] + video_url
+                secure_path = self._search_regex(r'https?://[^/]+(.+/)', video_url, 'secure path') + '*'
+                token = tokens.get(secure_path)
+                if not token:
+                    auth = self._download_xml(
+                        secure_path_data['tokenizer_src'], video_id, query={
+                            'path': secure_path,
+                            'videoId': video_id,
+                        })
+                    token = xpath_text(auth, 'token')
+                    if not token:
+                        continue
+                    tokens[secure_path] = token
+                video_url = video_url + '?hdnea=' + token
+            elif not re.match('https?://', video_url):
+                base_path_data = path_data.get(ext, path_data.get('default', {}))
+                media_src = base_path_data.get('media_src')
+                if not media_src:
+                    continue
+                video_url = media_src + video_url
+            if video_url in urls:
+                continue
+            urls.append(video_url)
+            format_id = video_file.get('bitrate')
+            if ext == 'smil':
+                formats.extend(self._extract_smil_formats(
+                    video_url, video_id, fatal=False))
+            elif ext == 'm3u8':
+                m3u8_formats = self._extract_m3u8_formats(
+                    video_url, video_id, 'mp4', m3u8_id=format_id or 'hls',
+                    fatal=False)
+                if m3u8_formats:
+                    # Sometimes final URLs inside m3u8 are unsigned, let's fix this
+                    # ourselves
+                    qs = compat_urlparse.urlparse(video_url).query
+                    if qs:
+                        query = compat_urlparse.parse_qs(qs)
+                        for m3u8_format in m3u8_formats:
+                            m3u8_format['url'] = update_url_query(m3u8_format['url'], query)
+                            m3u8_format['extra_param_to_segment_url'] = qs
+                    formats.extend(m3u8_formats)
+            elif ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(
+                    update_url_query(video_url, {'hdcore': '3.7.0'}),
+                    video_id, f4m_id=format_id or 'hds', fatal=False))
+            else:
+                f = {
+                    'format_id': format_id,
+                    'url': video_url,
+                    'ext': ext,
+                }
+                mobj = rex.search(format_id + video_url)
+                if mobj:
+                    f.update({
+                        'width': int(mobj.group('width')),
+                        'height': int(mobj.group('height')),
+                        'tbr': int_or_none(mobj.group('bitrate')),
+                    })
+                elif isinstance(format_id, compat_str):
+                    if format_id.isdigit():
+                        f['tbr'] = int(format_id)
+                    else:
+                        mobj = re.match(r'ios_(audio|[0-9]+)$', format_id)
+                        if mobj:
+                            if mobj.group(1) == 'audio':
+                                f.update({
+                                    'vcodec': 'none',
+                                    'ext': 'm4a',
+                                })
+                            else:
+                                f['tbr'] = int(mobj.group(1))
+                formats.append(f)
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for source in video_data.findall('closedCaptions/source'):
+            for track in source.findall('track'):
+                track_url = track.get('url')
+                if not isinstance(track_url, compat_str) or track_url.endswith('/big'):
+                    continue
+                lang = track.get('lang') or track.get('label') or 'en'
+                subtitles.setdefault(lang, []).append({
+                    'url': track_url,
+                    'ext': {
+                        'scc': 'scc',
+                        'webvtt': 'vtt',
+                        'smptett': 'tt',
+                    }.get(source.get('format'))
+                })
+
+        thumbnails = [{
+            'id': image.get('cut'),
+            'url': image.text,
+            'width': int_or_none(image.get('width')),
+            'height': int_or_none(image.get('height')),
+        } for image in video_data.findall('images/image')]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'subtitles': subtitles,
+            'thumbnails': thumbnails,
+            'description': xpath_text(video_data, 'description'),
+            'duration': parse_duration(xpath_text(video_data, 'length') or xpath_text(video_data, 'trt')),
+            'timestamp': self._extract_timestamp(video_data),
+            'upload_date': xpath_attr(video_data, 'metas', 'version'),
+            'series': xpath_text(video_data, 'showTitle'),
+            'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
+            'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
+        }

youtube_dl/extractor/uplynk.py

@@ -33,9 +33,7 @@ class UplynkIE(InfoExtractor):
         formats = self._extract_m3u8_formats('http://content.uplynk.com/%s.m3u8' % path, display_id, 'mp4')
         if session_id:
             for f in formats:
-                f['extra_param_to_segment_url'] = {
-                    'pbs': session_id,
-                }
+                f['extra_param_to_segment_url'] = 'pbs=' + session_id
         self._sort_formats(formats)
         asset = self._download_json('http://content.uplynk.com/player/assetinfo/%s.json' % path, display_id)
         if asset.get('error') == 1:

youtube_dl/extractor/usanetwork.py

@@ -0,0 +1,76 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .adobepass import AdobePassIE
+from ..utils import (
+    extract_attributes,
+    smuggle_url,
+    update_url_query,
+)
+
+
+class USANetworkIE(AdobePassIE):
+    _VALID_URL = r'https?://(?:www\.)?usanetwork\.com/(?:[^/]+/videos|movies)/(?P<id>[^/?#]+)'
+    _TEST = {
+        'url': 'http://www.usanetwork.com/mrrobot/videos/hpe-cybersecurity',
+        'md5': '33c0d2ba381571b414024440d08d57fd',
+        'info_dict': {
+            'id': '3086229',
+            'ext': 'mp4',
+            'title': 'HPE Cybersecurity',
+            'description': 'The more we digitize our world, the more vulnerable we are.',
+            'upload_date': '20160818',
+            'timestamp': 1471535460,
+            'uploader': 'NBCU-USA',
+        },
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        player_params = extract_attributes(self._search_regex(
+            r'(<div[^>]+data-usa-tve-player-container[^>]*>)', webpage, 'player params'))
+        video_id = player_params['data-mpx-guid']
+        title = player_params['data-episode-title']
+
+        account_pid, path = re.search(
+            r'data-src="(?:https?)?//player\.theplatform\.com/p/([^/]+)/.*?/(media/guid/\d+/\d+)',
+            webpage).groups()
+
+        query = {
+            'mbr': 'true',
+        }
+        if player_params.get('data-is-full-episode') == '1':
+            query['manifest'] = 'm3u'
+
+        if player_params.get('data-entitlement') == 'auth':
+            adobe_pass = {}
+            drupal_settings = self._search_regex(
+                r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
+                webpage, 'drupal settings', fatal=False)
+            if drupal_settings:
+                drupal_settings = self._parse_json(drupal_settings, video_id, fatal=False)
+                if drupal_settings:
+                    adobe_pass = drupal_settings.get('adobePass', {})
+            resource = self._get_mvpd_resource(
+                adobe_pass.get('adobePassResourceId', 'usa'),
+                title, video_id, player_params.get('data-episode-rating', 'TV-14'))
+            query['auth'] = self._extract_mvpd_auth(
+                url, video_id, adobe_pass.get('adobePassRequestorId', 'usa'), resource)
+
+        info = self._search_json_ld(webpage, video_id, default={})
+        info.update({
+            '_type': 'url_transparent',
+            'url': smuggle_url(update_url_query(
+                'http://link.theplatform.com/s/%s/%s' % (account_pid, path),
+                query), {'force_smil_url': True}),
+            'id': video_id,
+            'title': title,
+            'series': player_params.get('data-show-title'),
+            'episode': title,
+            'ie_key': 'ThePlatform',
+        })
+        return info

youtube_dl/extractor/vodplatform.py

@@ -6,7 +6,7 @@ from ..utils import unescapeHTML
 
 
 class VODPlatformIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vod-platform\.net/embed/(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?:www\.)?vod-platform\.net/[eE]mbed/(?P<id>[^/?#]+)'
     _TEST = {
         # from http://www.lbcgroup.tv/watch/chapter/29143/52844/%D8%A7%D9%84%D9%86%D8%B5%D8%B1%D8%A9-%D9%81%D9%8A-%D8%B6%D9%8A%D8%A7%D9%81%D8%A9-%D8%A7%D9%84%D9%80-cnn/ar
         'url': 'http://vod-platform.net/embed/RufMcytHDolTH1MuKHY9Fw',

youtube_dl/extractor/yahoo.py

@@ -8,7 +8,6 @@ import re
 from .common import InfoExtractor, SearchInfoExtractor
 from ..compat import (
     compat_urllib_parse,
-    compat_urllib_parse_urlencode,
     compat_urlparse,
 )
 from ..utils import (
@@ -17,6 +16,7 @@ from ..utils import (
     ExtractorError,
     int_or_none,
     mimetype2ext,
+    determine_ext,
 )
 
 from .brightcove import BrightcoveNewIE
@@ -39,7 +39,7 @@ class YahooIE(InfoExtractor):
         },
         {
             'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
-            'md5': 'c3466d2b6d5dd6b9f41ba9ed04c24b23',
+            'md5': '251af144a19ebc4a033e8ba91ac726bb',
             'info_dict': {
                 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
                 'ext': 'mp4',
@@ -50,7 +50,7 @@ class YahooIE(InfoExtractor):
         },
         {
             'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
-            'md5': '75ffabdb87c16d4ffe8c036dc4d1c136',
+            'md5': '7993e572fac98e044588d0b5260f4352',
             'info_dict': {
                 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
                 'ext': 'mp4',
@@ -61,7 +61,7 @@ class YahooIE(InfoExtractor):
         },
         {
             'url': 'https://tw.news.yahoo.com/%E6%95%A2%E5%95%8F%E5%B8%82%E9%95%B7%20%E9%BB%83%E7%A7%80%E9%9C%9C%E6%89%B9%E8%B3%B4%E6%B8%85%E5%BE%B7%20%E9%9D%9E%E5%B8%B8%E9%AB%98%E5%82%B2-034024051.html',
-            'md5': '9035d38f88b1782682a3e89f985be5bb',
+            'md5': '45c024bad51e63e9b6f6fad7a43a8c23',
             'info_dict': {
                 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f',
                 'ext': 'mp4',
@@ -72,10 +72,10 @@ class YahooIE(InfoExtractor):
         },
         {
             'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
-            'md5': '0b51660361f0e27c9789e7037ef76f4b',
+            'md5': '71298482f7c64cbb7fa064e4553ff1c1',
             'info_dict': {
                 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
-                'ext': 'mp4',
+                'ext': 'webm',
                 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
                 'description': 'md5:f66c890e1490f4910a9953c941dee944',
                 'duration': 97,
@@ -98,7 +98,7 @@ class YahooIE(InfoExtractor):
                 'id': '154609075',
             },
             'playlist': [{
-                'md5': 'f8e336c6b66f503282e5f719641d6565',
+                'md5': '000887d0dc609bc3a47c974151a40fb8',
                 'info_dict': {
                     'id': 'e624c4bc-3389-34de-9dfc-025f74943409',
                     'ext': 'mp4',
@@ -107,7 +107,7 @@ class YahooIE(InfoExtractor):
                     'duration': 30,
                 },
             }, {
-                'md5': '958bcb90b4d6df71c56312137ee1cd5a',
+                'md5': '81bc74faf10750fe36e4542f9a184c66',
                 'info_dict': {
                     'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9',
                     'ext': 'mp4',
@@ -139,7 +139,7 @@ class YahooIE(InfoExtractor):
             'skip': 'Domain name in.lifestyle.yahoo.com gone',
         }, {
             'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
-            'md5': 'b17ac378b1134fa44370fb27db09a744',
+            'md5': '2a9752f74cb898af5d1083ea9f661b58',
             'info_dict': {
                 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
                 'ext': 'mp4',
@@ -168,7 +168,7 @@ class YahooIE(InfoExtractor):
         }, {
             # Query result is embedded in webpage, but explicit request to video API fails with geo restriction
             'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html',
-            'md5': '1ddbf7c850777548438e5c4f147c7b8c',
+            'md5': '4fbafb9c9b6f07aa8f870629f6671b35',
             'info_dict': {
                 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504',
                 'ext': 'mp4',
@@ -196,6 +196,7 @@ class YahooIE(InfoExtractor):
                 'description': 'Galactic',
                 'title': 'Dolla Diva (feat. Maggie Koerner)',
             },
+            'skip': 'redirect to https://www.yahoo.com/music',
         },
     ]
 
@@ -213,15 +214,7 @@ class YahooIE(InfoExtractor):
         entries = []
         iframe_urls = re.findall(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage)
         for idx, iframe_url in enumerate(iframe_urls):
-            iframepage = self._download_webpage(
-                host + iframe_url, display_id,
-                note='Downloading iframe webpage for video #%d' % idx)
-            items_json = self._search_regex(
-                r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None)
-            if items_json:
-                items = json.loads(items_json)
-                video_id = items[0]['id']
-                entries.append(self._get_info(video_id, display_id, webpage))
+            entries.append(self.url_result(host + iframe_url, 'Yahoo'))
         if entries:
             return self.playlist_result(entries, page_id)
 
@@ -246,7 +239,9 @@ class YahooIE(InfoExtractor):
             if config:
                 sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi')
                 if sapi and 'query' in sapi:
-                    return self._extract_info(display_id, sapi, webpage)
+                    info = self._extract_info(display_id, sapi, webpage)
+                    self._sort_formats(info['formats'])
+                    return info
 
         items_json = self._search_regex(
             r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
@@ -292,15 +287,17 @@ class YahooIE(InfoExtractor):
 
         formats = []
         for s in info['streams']:
+            tbr = int_or_none(s.get('bitrate'))
             format_info = {
                 'width': int_or_none(s.get('width')),
                 'height': int_or_none(s.get('height')),
-                'tbr': int_or_none(s.get('bitrate')),
+                'tbr': tbr,
             }
 
             host = s['host']
             path = s['path']
             if host.startswith('rtmp'):
+                fmt = 'rtmp'
                 format_info.update({
                     'url': host,
                     'play_path': path,
@@ -308,14 +305,18 @@ class YahooIE(InfoExtractor):
                 })
             else:
                 if s.get('format') == 'm3u8_playlist':
-                    format_info['protocol'] = 'm3u8_native'
-                    format_info['ext'] = 'mp4'
+                    fmt = 'hls'
+                    format_info.update({
+                        'protocol': 'm3u8_native',
+                        'ext': 'mp4',
+                    })
+                else:
+                    fmt = format_info['ext'] = determine_ext(path)
                 format_url = compat_urlparse.urljoin(host, path)
                 format_info['url'] = format_url
+            format_info['format_id'] = fmt + ('-%d' % tbr if tbr else '')
             formats.append(format_info)
 
-        self._sort_formats(formats)
-
         closed_captions = self._html_search_regex(
             r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions',
             default='[]')
@@ -346,17 +347,25 @@ class YahooIE(InfoExtractor):
     def _get_info(self, video_id, display_id, webpage):
         region = self._search_regex(
             r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"',
-            webpage, 'region', fatal=False, default='US')
-        data = compat_urllib_parse_urlencode({
-            'protocol': 'http',
-            'region': region.upper(),
-        })
-        query_url = (
-            'https://video.media.yql.yahoo.com/v1/video/sapi/streams/'
-            '{id}?{data}'.format(id=video_id, data=data))
-        query_result = self._download_json(
-            query_url, display_id, 'Downloading video info')
-        return self._extract_info(display_id, query_result, webpage)
+            webpage, 'region', fatal=False, default='US').upper()
+        formats = []
+        info = {}
+        for fmt in ('webm', 'mp4'):
+            query_result = self._download_json(
+                'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' + video_id,
+                display_id, 'Downloading %s video info' % fmt, query={
+                    'protocol': 'http',
+                    'region': region,
+                    'format': fmt,
+                })
+            info = self._extract_info(display_id, query_result, webpage)
+            formats.extend(info['formats'])
+        formats.extend(self._extract_m3u8_formats(
+            'http://video.media.yql.yahoo.com/v1/hls/%s?region=%s' % (video_id, region),
+            video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+        self._sort_formats(formats)
+        info['formats'] = formats
+        return info
 
 
 class YahooSearchIE(SearchInfoExtractor):

youtube_dl/extractor/youporn.py

@@ -35,7 +35,7 @@ class YouPornIE(InfoExtractor):
             'age_limit': 18,
         },
     }, {
-        # Anonymous User uploader
+        # Unknown uploader
         'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4',
         'info_dict': {
             'id': '561726',
@@ -44,7 +44,7 @@ class YouPornIE(InfoExtractor):
             'title': 'Big Tits Awesome Brunette On amazing webcam show',
             'description': 'http://sweetlivegirls.com Big Tits Awesome Brunette On amazing webcam show.mp4',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'uploader': 'Anonymous User',
+            'uploader': 'Unknown',
             'upload_date': '20111125',
             'average_rating': int,
             'view_count': int,
@@ -140,17 +140,17 @@ class YouPornIE(InfoExtractor):
             r'>All [Cc]omments? \(([\d,.]+)\)',
             webpage, 'comment count', fatal=False))
 
-        def extract_tag_box(title):
-            tag_box = self._search_regex(
-                (r'<div[^>]+class=["\']tagBoxTitle["\'][^>]*>\s*%s\b.*?</div>\s*'
-                 '<div[^>]+class=["\']tagBoxContent["\']>(.+?)</div>') % re.escape(title),
-                webpage, '%s tag box' % title, default=None)
+        def extract_tag_box(regex, title):
+            tag_box = self._search_regex(regex, webpage, title, default=None)
             if not tag_box:
                 return []
             return re.findall(r'<a[^>]+href=[^>]+>([^<]+)', tag_box)
 
-        categories = extract_tag_box('Category')
-        tags = extract_tag_box('Tags')
+        categories = extract_tag_box(
+            r'(?s)Categories:.*?</[^>]+>(.+?)</div>', 'categories')
+        tags = extract_tag_box(
+            r'(?s)Tags:.*?</div>\s*<div[^>]+class=["\']tagBoxContent["\'][^>]*>(.+?)</div>',
+            'tags')
 
         return {
             'id': video_id,

youtube_dl/extractor/youtube.py

@@ -844,6 +844,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
             'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
             'only_matching': True,
+        },
+        {
+            # Rental video preview
+            'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
+            'info_dict': {
+                'id': 'uGpuVWrhIzE',
+                'ext': 'mp4',
+                'title': 'Piku - Trailer',
+                'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
+                'upload_date': '20150811',
+                'uploader': 'FlixMatrix',
+                'uploader_id': 'FlixMatrixKaravan',
+                'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
+                'license': 'Standard YouTube License',
+            },
+            'params': {
+                'skip_download': True,
+            },
         }
     ]
 
@@ -1254,6 +1272,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     # Convert to the same format returned by compat_parse_qs
                     video_info = dict((k, [v]) for k, v in args.items())
                     add_dash_mpd(video_info)
+                # Rental video is not rented but preview is available (e.g.
+                # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
+                # https://github.com/rg3/youtube-dl/issues/10532)
+                if not video_info and args.get('ypc_vid'):
+                    return self.url_result(
+                        args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
                 if args.get('livestream') == '1' or args.get('live_playback') == 1:
                     is_live = True
             if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
@@ -1841,6 +1865,28 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
         },
         'playlist_mincout': 21,
+    }, {
+        # Playlist URL that does not actually serve a playlist
+        'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
+        'info_dict': {
+            'id': 'FqZTN594JQw',
+            'ext': 'webm',
+            'title': "Smiley's People 01 detective, Adventure Series, Action",
+            'uploader': 'STREEM',
+            'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
+            'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
+            'upload_date': '20150526',
+            'license': 'Standard YouTube License',
+            'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
+            'categories': ['People & Blogs'],
+            'tags': list,
+            'like_count': int,
+            'dislike_count': int,
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': [YoutubeIE.ie_key()],
     }]
 
     def _real_initialize(self):
@@ -1901,9 +1947,20 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
 
         playlist_title = self._html_search_regex(
             r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
-            page, 'title')
+            page, 'title', default=None)
 
-        return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
+        has_videos = True
+
+        if not playlist_title:
+            try:
+                # Some playlist URLs don't actually serve a playlist (e.g.
+                # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
+                next(self._entries(page, playlist_id))
+            except StopIteration:
+                has_videos = False
+
+        return has_videos, self.playlist_result(
+            self._entries(page, playlist_id), playlist_id, playlist_title)
 
     def _check_download_just_video(self, url, playlist_id):
         # Check if it's a video-specific URL
@@ -1912,9 +1969,11 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             video_id = query_dict['v'][0]
             if self._downloader.params.get('noplaylist'):
                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
-                return self.url_result(video_id, 'Youtube', video_id=video_id)
+                return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
             else:
                 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
+                return video_id, None
+        return None, None
 
     def _real_extract(self, url):
         # Extract playlist id
@@ -1923,7 +1982,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             raise ExtractorError('Invalid URL: %s' % url)
         playlist_id = mobj.group(1) or mobj.group(2)
 
-        video = self._check_download_just_video(url, playlist_id)
+        video_id, video = self._check_download_just_video(url, playlist_id)
         if video:
             return video
 
@@ -1931,7 +1990,15 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             # Mixes require a custom extraction process
             return self._extract_mix(playlist_id)
 
-        return self._extract_playlist(playlist_id)
+        has_videos, playlist = self._extract_playlist(playlist_id)
+        if has_videos or not video_id:
+            return playlist
+
+        # Some playlist URLs don't actually serve a playlist (see
+        # https://github.com/rg3/youtube-dl/issues/10537).
+        # Fallback to plain video extraction if there is a video id
+        # along with playlist id.
+        return self.url_result(video_id, 'Youtube', video_id=video_id)
 
 
 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
@@ -2312,7 +2379,8 @@ class YoutubeWatchLaterIE(YoutubePlaylistIE):
         video = self._check_download_just_video(url, 'WL')
         if video:
             return video
-        return self._extract_playlist('WL')
+        _, playlist = self._extract_playlist('WL')
+        return playlist
 
 
 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):

youtube_dl/utils.py

@@ -2148,7 +2148,7 @@ def mimetype2ext(mt):
         return ext
 
     _, _, res = mt.rpartition('/')
-    res = res.lower()
+    res = res.split(';')[0].strip().lower()
 
     return {
         '3gpp': '3gp',
@@ -2168,6 +2168,7 @@ def mimetype2ext(mt):
         'f4m+xml': 'f4m',
         'hds+xml': 'f4m',
         'vnd.ms-sstr+xml': 'ism',
+        'quicktime': 'mov',
     }.get(res, res)
 
 
@@ -2183,7 +2184,7 @@ def parse_codecs(codecs_str):
         if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v'):
             if not vcodec:
                 vcodec = full_codec
-        elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac'):
+        elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3'):
             if not acodec:
                 acodec = full_codec
         else:

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.08.24.1'
+__version__ = '2016.09.03'