release 2016.08.13

There's no video anymore

.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 *2016.07.30*. 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 Issues with outdated version will be rejected.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 **2016.07.30**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.08.13*. 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 Issues with outdated version will be rejected.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 **2016.08.13**
 
 ### 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 2016.07.30
+[debug] youtube-dl version 2016.08.13
 [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

@@ -179,3 +179,5 @@ Jakub Adam Wieczorek
 Aleksandar Topuzović
 Nehal Patel
 Rob van Bekkum
+Petr Zvoníček
+Pratyush Singh

CONTRIBUTING.md

@@ -46,7 +46,7 @@ Make sure that someone has not already opened the issue you're trying to open. S
 
 ###  Why are existing options not enough?
 
-Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#synopsis). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
+Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#options). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
 
 ###  Is there enough context in your bug report?
 

ChangeLog

@@ -1,3 +1,135 @@
+version 2016.08.13
+
+Core
+* Show progress for curl external downloader
+* Forward more options to curl external downloader
+
+Extractors
+* [pbs] Fix description extraction
+* [franceculture] Fix extraction (#10324)
+* [pornotube] Fix extraction (#10322)
+* [4tube] Fix metadata extraction (#10321)
+* [imgur] Fix width and height extraction (#10325)
+* [expotv] Improve extraction
++ [vbox7] Fix extraction (#10309)
+- [tapely] Remove extractor (#10323)
+* [muenchentv] Fix extraction (#10313)
++ [24video] Add support for .me and .xxx TLDs
+* [24video] Fix comment count extraction
+* [sunporno] Add support for embed URLs
+* [sunporno] Fix metadata extraction (#10316)
++ [hgtv] Add extractor for hgtv.ca (#3999)
+- [pbs] Remove request to unavailable API
++ [pbs] Add support for high quality HTTP formats
++ [crunchyroll] Add support for HLS formats (#10301)
+
+version 2016.08.12
+
+Core
+* Subtitles are now written as is. Newline conversions are disabled. (#10268)
++ Recognize more formats in unified_timestamp
+
+Extractors
+- [goldenmoustache] Remove extractor (#10298)
+* [drtuber] Improve title extraction
+* [drtuber] Make dislike count optional (#10297)
+* [chirbit] Fix extraction (#10296)
+* [francetvinfo] Relax URL regular expression
+* [rtlnl] Relax URL regular expression (#10282)
+* [formula1] Relax URL regular expression (#10283)
+* [wat] Improve extraction (#10281)
+* [ctsnews] Fix extraction
+
+
+version 2016.08.10
+
+Core
+* Make --metadata-from-title non fatal when title does not match the pattern
+* Introduce options for randomized sleep before each download
+  --min-sleep-interval and --max-sleep-interval (#9930)
+* Respect default in _search_json_ld
+
+Extractors
++ [uol] Add extractor for uol.com.br (#4263)
+* [rbmaradio] Fix extraction and extract all formats (#10242)
++ [sonyliv] Add extractor for sonyliv.com (#10258)
+* [aparat] Fix extraction
+* [cwtv] Extract HTTP formats
++ [rozhlas] Add extractor for prehravac.rozhlas.cz (#10253)
+* [kuwo:singer] Fix extraction
+
+
+version 2016.08.07
+
+Core
++ Add support for TV Parental Guidelines ratings in parse_age_limit
++ Add decode_png (#9706)
++ Add support for partOfTVSeries in JSON-LD
+* Lower master M3U8 manifest preference for better format sorting
+
+Extractors
++ [discoverygo] Add extractor (#10245)
+* [flipagram] Make JSON-LD extraction non fatal
+* [generic] Make JSON-LD extraction non fatal
++ [bbc] Add support for morph embeds (#10239)
+* [tnaflixnetworkbase] Improve title extraction
+* [tnaflix] Fix metadata extraction (#10249)
+* [fox] Fix theplatform release URL query
+* [openload] Fix extraction (#9706)
+* [bbc] Skip duplicate manifest URLs
+* [bbc] Improve format code
++ [bbc] Add support for DASH and F4M
+* [bbc] Improve format sorting and listing
+* [bbc] Improve playlist extraction
++ [pokemon] Add extractor (#10093)
++ [condenast] Add fallback scenario for video info extraction
+
+
+version 2016.08.06
+
+Core
+* Add support for JSON-LD root list entries (#10203)
+* Improve unified_timestamp
+* Lower preference of RTSP formats in generic sorting
++ Add support for multiple properties in _og_search_property
+* Improve password hiding from verbose output
+
+Extractors
++ [adultswim] Add support for trailers (#10235)
+* [archiveorg] Improve extraction (#10219)
++ [jwplatform] Add support for playlists
++ [jwplatform] Add support for relative URLs
+* [jwplatform] Improve audio detection
++ [tvplay] Capture and output native error message
++ [tvplay] Extract series metadata
++ [tvplay] Add support for subtitles (#10194)
+* [tvp] Improve extraction (#7799)
+* [cbslocal] Fix timestamp parsing (#10213)
++ [naver] Add support for subtitles (#8096)
+* [naver] Improve extraction
+* [condenast] Improve extraction
+* [engadget] Relax URL regular expression
+* [5min] Fix extraction
++ [nationalgeographic] Add support for Episode Guide
++ [kaltura] Add support for subtitles
+* [kaltura] Optimize network requests
++ [vodplatform] Add extractor for vod-platform.net
+- [gamekings] Remove extractor
+* [limelight] Extract HTTP formats
+* [ntvru] Fix extraction
++ [comedycentral] Re-add :tds and :thedailyshow shortnames
+
+
+version 2016.08.01
+
+Fixed/improved extractors
+- [yandexmusic:track] Adapt to changes in track location JSON (#10193)
+- [bloomberg] Support another form of player (#10187)
+- [limelight] Skip DRM protected videos
+- [safari] Relax regular expressions for URL matching (#10202)
+- [cwtv] Add support for cwtvpr.com (#10196)
+
+
 version 2016.07.30
 
 Fixed/improved extractors

README.md

@@ -330,7 +330,15 @@ which means you can modify it, redistribute it or use it however you like.
                                      bidirectional text support. Requires bidiv
                                      or fribidi executable in PATH
     --sleep-interval SECONDS         Number of seconds to sleep before each
-                                     download.
+                                     download when used alone or a lower bound
+                                     of a range for randomized sleep before each
+                                     download (minimum possible number of
+                                     seconds to sleep) when used along with
+                                     --max-sleep-interval.
+    --max-sleep-interval SECONDS     Upper bound of a range for randomized sleep
+                                     before each download (maximum possible
+                                     number of seconds to sleep). Must only be
+                                     used along with --min-sleep-interval.
 
 ## Video Format Options:
     -f, --format FORMAT              Video format code, see the "FORMAT
@@ -1196,7 +1204,7 @@ Make sure that someone has not already opened the issue you're trying to open. S
 
 ###  Why are existing options not enough?
 
-Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#synopsis). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
+Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#options). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
 
 ###  Is there enough context in your bug report?
 

devscripts/prepare_manpage.py

@@ -54,7 +54,11 @@ def filter_options(readme):
 
         if in_options:
             if line.lstrip().startswith('-'):
-                option, description = re.split(r'\s{2,}', line.lstrip())
+                split = re.split(r'\s{2,}', line.lstrip())
+                # Description string may start with `-` as well. If there is
+                # only one piece then it's a description bit not an option.
+                if len(split) > 1:
+                    option, description = split
                     split_option = option.split(' ')
 
                     if not split_option[-1].startswith('-'):  # metavar
@@ -63,7 +67,7 @@ def filter_options(readme):
                     # Pandoc's definition_lists. See http://pandoc.org/README.html
                     # for more information.
                     ret += '\n%s\n:   %s\n' % (option, description)
-            else:
+                    continue
             ret += line.lstrip() + '\n'
         else:
             ret += line + '\n'

docs/supportedsites.md

@@ -142,6 +142,7 @@
  - **CollegeRama**
  - **ComCarCoff**
  - **ComedyCentral**
+ - **ComedyCentralShortname**
  - **ComedyCentralTV**
  - **CondeNast**: Condé Nast media group: Allure, Architectural Digest, Ars Technica, Bon Appétit, Brides, Condé Nast, Condé Nast Traveler, Details, Epicurious, GQ, Glamour, Golf Digest, SELF, Teen Vogue, The New Yorker, Vanity Fair, Vogue, W Magazine, WIRED
  - **Coub**
@@ -181,6 +182,7 @@
  - **DigitallySpeaking**
  - **Digiteka**
  - **Discovery**
+ - **DiscoveryGo**
  - **Dotsub**
  - **DouyuTV**: 斗鱼
  - **DPlay**
@@ -236,7 +238,6 @@
  - **FoxSports**
  - **france2.fr:generation-quoi**
  - **FranceCulture**
- - **FranceCultureEmission**
  - **FranceInter**
  - **francetv**: France 2, 3, 4, 5 and Ô
  - **francetvinfo.fr**
@@ -247,7 +248,6 @@
  - **FunnyOrDie**
  - **Fusion**
  - **GameInformer**
- - **Gamekings**
  - **GameOne**
  - **gameone:playlist**
  - **Gamersyde**
@@ -264,7 +264,6 @@
  - **GloboArticle**
  - **GodTube**
  - **GodTV**
- - **GoldenMoustache**
  - **Golem**
  - **GoogleDrive**
  - **Goshgay**
@@ -277,6 +276,7 @@
  - **HellPorno**
  - **Helsinki**: helsinki.fi
  - **HentaiStigma**
+ - **HGTV**
  - **HistoricFilms**
  - **history:topic**: History.com Topic
  - **hitbox**
@@ -415,7 +415,8 @@
  - **MyVidster**
  - **n-tv.de**
  - **natgeo**
- - **natgeo:channel**
+ - **natgeo:episodeguide**
+ - **natgeo:video**
  - **Naver**
  - **NBA**
  - **NBC**
@@ -517,6 +518,7 @@
  - **plus.google**: Google Plus
  - **pluzz.francetv.fr**
  - **podomatic**
+ - **Pokemon**
  - **PolskieRadio**
  - **PornHd**
  - **PornHub**: PornHub and Thumbzilla
@@ -561,6 +563,7 @@
  - **RoosterTeeth**
  - **RottenTomatoes**
  - **Roxwel**
+ - **Rozhlas**
  - **RTBF**
  - **rte**: Raidió Teilifís Éireann TV
  - **rte:radio**: Raidió Teilifís Éireann radio
@@ -618,6 +621,7 @@
  - **smotri:user**: Smotri.com user videos
  - **Snotr**
  - **Sohu**
+ - **SonyLIV**
  - **soundcloud**
  - **soundcloud:playlist**
  - **soundcloud:search**: Soundcloud search
@@ -660,7 +664,6 @@
  - **SztvHu**
  - **Tagesschau**
  - **tagesschau:player**
- - **Tapely**
  - **Tass**
  - **TDSLifeway**
  - **teachertube**: teachertube.com videos
@@ -726,6 +729,7 @@
  - **tvigle**: Интернет-телевидение Tvigle.ru
  - **tvland.com**
  - **tvp**: Telewizja Polska
+ - **tvp:embed**: Telewizja Polska
  - **tvp:series**
  - **TVPlay**: TV3Play and related services
  - **Tweakers**
@@ -743,6 +747,7 @@
  - **udemy:course**
  - **UDNEmbed**: 聯合影音
  - **Unistra**
+ - **uol.com.br**
  - **Urort**: NRK P3 Urørt
  - **URPlay**
  - **USAToday**
@@ -805,6 +810,7 @@
  - **vk:wallpost**
  - **vlive**
  - **Vodlocker**
+ - **VODPlatform**
  - **VoiceRepublic**
  - **VoxMedia**
  - **Vporn**

test/test_InfoExtractor.py

@@ -48,6 +48,9 @@ class TestInfoExtractor(unittest.TestCase):
         self.assertEqual(ie._og_search_property('foobar', html), 'Foo')
         self.assertEqual(ie._og_search_property('test1', html), 'foo > < bar')
         self.assertEqual(ie._og_search_property('test2', html), 'foo >//< bar')
+        self.assertEqual(ie._og_search_property(('test0', 'test1'), html), 'foo > < bar')
+        self.assertRaises(RegexNotFoundError, ie._og_search_property, 'test0', html, None, fatal=True)
+        self.assertRaises(RegexNotFoundError, ie._og_search_property, ('test0', 'test00'), html, None, fatal=True)
 
     def test_html_search_meta(self):
         ie = self.ie

test/test_utils.py

@@ -42,6 +42,7 @@ from youtube_dl.utils import (
     ohdave_rsa_encrypt,
     OnDemandPagedList,
     orderedSet,
+    parse_age_limit,
     parse_duration,
     parse_filesize,
     parse_count,
@@ -308,6 +309,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unified_timestamp('25-09-2014'), 1411603200)
         self.assertEqual(unified_timestamp('27.02.2016 17:30'), 1456594200)
         self.assertEqual(unified_timestamp('UNKNOWN DATE FORMAT'), None)
+        self.assertEqual(unified_timestamp('May 16, 2016 11:15 PM'), 1463440500)
 
     def test_determine_ext(self):
         self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4')
@@ -431,6 +433,20 @@ class TestUtil(unittest.TestCase):
             url_basename('http://media.w3.org/2010/05/sintel/trailer.mp4'),
             'trailer.mp4')
 
+    def test_parse_age_limit(self):
+        self.assertEqual(parse_age_limit(None), None)
+        self.assertEqual(parse_age_limit(False), None)
+        self.assertEqual(parse_age_limit('invalid'), None)
+        self.assertEqual(parse_age_limit(0), 0)
+        self.assertEqual(parse_age_limit(18), 18)
+        self.assertEqual(parse_age_limit(21), 21)
+        self.assertEqual(parse_age_limit(22), None)
+        self.assertEqual(parse_age_limit('18'), 18)
+        self.assertEqual(parse_age_limit('18+'), 18)
+        self.assertEqual(parse_age_limit('PG-13'), 13)
+        self.assertEqual(parse_age_limit('TV-14'), 14)
+        self.assertEqual(parse_age_limit('TV-MA'), 17)
+
     def test_parse_duration(self):
         self.assertEqual(parse_duration(None), None)
         self.assertEqual(parse_duration(False), None)
@@ -952,6 +968,7 @@ The first line
         self.assertEqual(cli_option({'proxy': '127.0.0.1:3128'}, '--proxy', 'proxy'), ['--proxy', '127.0.0.1:3128'])
         self.assertEqual(cli_option({'proxy': None}, '--proxy', 'proxy'), [])
         self.assertEqual(cli_option({}, '--proxy', 'proxy'), [])
+        self.assertEqual(cli_option({'retries': 10}, '--retries', 'retries'), ['--retries', '10'])
 
     def test_cli_valueless_option(self):
         self.assertEqual(cli_valueless_option(

test/test_verbose_output.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+from __future__ import unicode_literals
+
+import unittest
+
+import sys
+import os
+import subprocess
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+
+class TestVerboseOutput(unittest.TestCase):
+    def test_private_info_arg(self):
+        outp = subprocess.Popen(
+            [
+                sys.executable, 'youtube_dl/__main__.py', '-v',
+                '--username', 'johnsmith@gmail.com',
+                '--password', 'secret',
+            ], cwd=rootDir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        sout, serr = outp.communicate()
+        self.assertTrue(b'--username' in serr)
+        self.assertTrue(b'johnsmith' not in serr)
+        self.assertTrue(b'--password' in serr)
+        self.assertTrue(b'secret' not in serr)
+
+    def test_private_info_shortarg(self):
+        outp = subprocess.Popen(
+            [
+                sys.executable, 'youtube_dl/__main__.py', '-v',
+                '-u', 'johnsmith@gmail.com',
+                '-p', 'secret',
+            ], cwd=rootDir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        sout, serr = outp.communicate()
+        self.assertTrue(b'-u' in serr)
+        self.assertTrue(b'johnsmith' not in serr)
+        self.assertTrue(b'-p' in serr)
+        self.assertTrue(b'secret' not in serr)
+
+    def test_private_info_eq(self):
+        outp = subprocess.Popen(
+            [
+                sys.executable, 'youtube_dl/__main__.py', '-v',
+                '--username=johnsmith@gmail.com',
+                '--password=secret',
+            ], cwd=rootDir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        sout, serr = outp.communicate()
+        self.assertTrue(b'--username' in serr)
+        self.assertTrue(b'johnsmith' not in serr)
+        self.assertTrue(b'--password' in serr)
+        self.assertTrue(b'secret' not in serr)
+
+    def test_private_info_shortarg_eq(self):
+        outp = subprocess.Popen(
+            [
+                sys.executable, 'youtube_dl/__main__.py', '-v',
+                '-u=johnsmith@gmail.com',
+                '-p=secret',
+            ], cwd=rootDir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        sout, serr = outp.communicate()
+        self.assertTrue(b'-u' in serr)
+        self.assertTrue(b'johnsmith' not in serr)
+        self.assertTrue(b'-p' in serr)
+        self.assertTrue(b'secret' not in serr)
+
+if __name__ == '__main__':
+    unittest.main()

youtube_dl/YoutubeDL.py

@@ -249,7 +249,16 @@ class YoutubeDL(object):
     source_address:    (Experimental) Client-side IP address to bind to.
     call_home:         Boolean, true iff we are allowed to contact the
                        youtube-dl servers for debugging.
-    sleep_interval:    Number of seconds to sleep before each download.
+    sleep_interval:    Number of seconds to sleep before each download when
+                       used alone or a lower bound of a range for randomized
+                       sleep before each download (minimum possible number
+                       of seconds to sleep) when used along with
+                       max_sleep_interval.
+    max_sleep_interval:Upper bound of a range for randomized sleep before each
+                       download (maximum possible number of seconds to sleep).
+                       Must only be used along with sleep_interval.
+                       Actual sleep time will be a random float from range
+                       [sleep_interval; max_sleep_interval].
     listformats:       Print an overview of available video formats and exit.
     list_thumbnails:   Print a table of all thumbnails and exit.
     match_filter:      A function that gets called with the info_dict of
@@ -1594,7 +1603,9 @@ class YoutubeDL(object):
                         self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
                     else:
                         self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
-                        with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
+                        # Use newline='' to prevent conversion of newline characters
+                        # See https://github.com/rg3/youtube-dl/issues/10268
+                        with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
                             subfile.write(sub_data)
                 except (OSError, IOError):
                     self.report_error('Cannot write subtitles file ' + sub_filename)

youtube_dl/__init__.py

@@ -145,6 +145,16 @@ def _real_main(argv=None):
         if numeric_limit is None:
             parser.error('invalid max_filesize specified')
         opts.max_filesize = numeric_limit
+    if opts.sleep_interval is not None:
+        if opts.sleep_interval < 0:
+            parser.error('sleep interval must be positive or 0')
+    if opts.max_sleep_interval is not None:
+        if opts.max_sleep_interval < 0:
+            parser.error('max sleep interval must be positive or 0')
+        if opts.max_sleep_interval < opts.sleep_interval:
+            parser.error('max sleep interval must be greater than or equal to min sleep interval')
+    else:
+        opts.max_sleep_interval = opts.sleep_interval
 
     def parse_retries(retries):
         if retries in ('inf', 'infinite'):
@@ -370,6 +380,7 @@ def _real_main(argv=None):
         'source_address': opts.source_address,
         'call_home': opts.call_home,
         'sleep_interval': opts.sleep_interval,
+        'max_sleep_interval': opts.max_sleep_interval,
         'external_downloader': opts.external_downloader,
         'list_thumbnails': opts.list_thumbnails,
         'playlist_items': opts.playlist_items,

youtube_dl/downloader/common.py

@@ -4,6 +4,7 @@ import os
 import re
 import sys
 import time
+import random
 
 from ..compat import compat_os_name
 from ..utils import (
@@ -342,8 +343,11 @@ class FileDownloader(object):
             })
             return True
 
-        sleep_interval = self.params.get('sleep_interval')
-        if sleep_interval:
+        min_sleep_interval = self.params.get('sleep_interval')
+        if min_sleep_interval:
+            max_sleep_interval = self.params.get('max_sleep_interval', min_sleep_interval)
+            print(min_sleep_interval, max_sleep_interval)
+            sleep_interval = random.uniform(min_sleep_interval, max_sleep_interval)
             self.to_screen('[download] Sleeping %s seconds...' % sleep_interval)
             time.sleep(sleep_interval)
 

youtube_dl/downloader/external.py

@@ -96,6 +96,12 @@ class CurlFD(ExternalFD):
         cmd = [self.exe, '--location', '-o', tmpfilename]
         for key, val in info_dict['http_headers'].items():
             cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._bool_option('--continue-at', 'continuedl', '-', '0')
+        cmd += self._valueless_option('--silent', 'noprogress')
+        cmd += self._valueless_option('--verbose', 'verbose')
+        cmd += self._option('--limit-rate', 'ratelimit')
+        cmd += self._option('--retry', 'retries')
+        cmd += self._option('--max-filesize', 'max_filesize')
         cmd += self._option('--interface', 'source_address')
         cmd += self._option('--proxy', 'proxy')
         cmd += self._valueless_option('--insecure', 'nocheckcertificate')
@@ -103,6 +109,16 @@ class CurlFD(ExternalFD):
         cmd += ['--', info_dict['url']]
         return cmd
 
+    def _call_downloader(self, tmpfilename, info_dict):
+        cmd = [encodeArgument(a) for a in self._make_cmd(tmpfilename, info_dict)]
+
+        self._debug_cmd(cmd)
+
+        # curl writes the progress to stderr so don't capture it.
+        p = subprocess.Popen(cmd)
+        p.communicate()
+        return p.returncode
+
 
 class AxelFD(ExternalFD):
     AVAILABLE_OPT = '-V'

youtube_dl/extractor/adultswim.py

@@ -83,6 +83,20 @@ class AdultSwimIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        # heroMetadata.trailer
+        'url': 'http://www.adultswim.com/videos/decker/inside-decker-a-new-hero/',
+        'info_dict': {
+            'id': 'I0LQFQkaSUaFp8PnAWHhoQ',
+            'ext': 'mp4',
+            'title': 'Decker - Inside Decker: A New Hero',
+            'description': 'md5:c916df071d425d62d70c86d4399d3ee0',
+            'duration': 249.008,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
     }]
 
     @staticmethod
@@ -133,20 +147,26 @@ class AdultSwimIE(InfoExtractor):
             if video_info is None:
                 if bootstrapped_data.get('slugged_video', {}).get('slug') == episode_path:
                     video_info = bootstrapped_data['slugged_video']
-                else:
+            if not video_info:
+                video_info = bootstrapped_data.get('heroMetadata', {}).get('trailer').get('video')
+            if not video_info:
                 raise ExtractorError('Unable to find video info')
 
             show = bootstrapped_data['show']
             show_title = show['title']
             stream = video_info.get('stream')
-            clips = [stream] if stream else video_info.get('clips')
-            if not clips:
+            if stream and stream.get('videoPlaybackID'):
+                segment_ids = [stream['videoPlaybackID']]
+            elif video_info.get('clips'):
+                segment_ids = [clip['videoPlaybackID'] for clip in video_info['clips']]
+            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)
-            segment_ids = [clip['videoPlaybackID'] for clip in clips]
 
         episode_id = video_info['id']
         episode_title = video_info['title']

youtube_dl/extractor/aol.py

@@ -123,6 +123,10 @@ class AolFeaturesIE(InfoExtractor):
             'title': 'What To Watch - February 17, 2016',
         },
         'add_ie': ['FiveMin'],
+        'params': {
+            # encrypted m3u8 download
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/aparat.py

@@ -1,8 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
@@ -15,7 +13,7 @@ class AparatIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.aparat.com/v/wP8On',
-        'md5': '6714e0af7e0d875c5a39c4dc4ab46ad1',
+        'md5': '131aca2e14fe7c4dcb3c4877ba300c89',
         'info_dict': {
             'id': 'wP8On',
             'ext': 'mp4',
@@ -31,13 +29,13 @@ class AparatIE(InfoExtractor):
         # Note: There is an easier-to-parse configuration at
         # http://www.aparat.com/video/video/config/videohash/%video_id
         # but the URL in there does not work
-        embed_url = ('http://www.aparat.com/video/video/embed/videohash/' +
-                     video_id + '/vt/frame')
+        embed_url = 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id
         webpage = self._download_webpage(embed_url, video_id)
 
-        video_urls = [video_url.replace('\\/', '/') for video_url in re.findall(
-            r'(?:fileList\[[0-9]+\]\s*=|"file"\s*:)\s*"([^"]+)"', webpage)]
-        for i, video_url in enumerate(video_urls):
+        file_list = self._parse_json(self._search_regex(
+            r'fileList\s*=\s*JSON\.parse\(\'([^\']+)\'\)', webpage, 'file list'), video_id)
+        for i, item in enumerate(file_list[0]):
+            video_url = item['file']
             req = HEADRequest(video_url)
             res = self._request_webpage(
                 req, video_id, note='Testing video URL %d' % i, errnote=False)

youtube_dl/extractor/archiveorg.py

@@ -1,67 +1,65 @@
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from ..utils import unified_strdate
+from .jwplatform import JWPlatformBaseIE
+from ..utils import (
+    unified_strdate,
+    clean_html,
+)
 
 
-class ArchiveOrgIE(InfoExtractor):
+class ArchiveOrgIE(JWPlatformBaseIE):
     IE_NAME = 'archive.org'
     IE_DESC = 'archive.org videos'
-    _VALID_URL = r'https?://(?:www\.)?archive\.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
+    _VALID_URL = r'https?://(?:www\.)?archive\.org/(?:details|embed)/(?P<id>[^/?#]+)(?:[?].*)?$'
     _TESTS = [{
         'url': 'http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
         'md5': '8af1d4cf447933ed3c7f4871162602db',
         'info_dict': {
             'id': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect',
-            'ext': 'ogv',
+            'ext': 'ogg',
             'title': '1968 Demo - FJCC Conference Presentation Reel #1',
-            'description': 'md5:1780b464abaca9991d8968c877bb53ed',
+            'description': 'md5:da45c349df039f1cc8075268eb1b5c25',
             'upload_date': '19681210',
             'uploader': 'SRI International'
         }
     }, {
         'url': 'https://archive.org/details/Cops1922',
-        'md5': '18f2a19e6d89af8425671da1cf3d4e04',
+        'md5': 'bc73c8ab3838b5a8fc6c6651fa7b58ba',
         'info_dict': {
             'id': 'Cops1922',
-            'ext': 'ogv',
+            'ext': 'mp4',
             'title': 'Buster Keaton\'s "Cops" (1922)',
-            'description': 'md5:70f72ee70882f713d4578725461ffcc3',
+            'description': 'md5:b4544662605877edd99df22f9620d858',
         }
+    }, {
+        'url': 'http://archive.org/embed/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            'http://archive.org/embed/' + video_id, video_id)
+        jwplayer_playlist = self._parse_json(self._search_regex(
+            r"(?s)Play\('[^']+'\s*,\s*(\[.+\])\s*,\s*{.*?}\);",
+            webpage, 'jwplayer playlist'), video_id)
+        info = self._parse_jwplayer_data(
+            {'playlist': jwplayer_playlist}, video_id, base_url=url)
 
-        json_url = url + ('&' if '?' in url else '?') + 'output=json'
-        data = self._download_json(json_url, video_id)
+        def get_optional(metadata, field):
+            return metadata.get(field, [None])[0]
 
-        def get_optional(data_dict, field):
-            return data_dict['metadata'].get(field, [None])[0]
-
-        title = get_optional(data, 'title')
-        description = get_optional(data, 'description')
-        uploader = get_optional(data, 'creator')
-        upload_date = unified_strdate(get_optional(data, 'date'))
-
-        formats = [
-            {
-                'format': fdata['format'],
-                'url': 'http://' + data['server'] + data['dir'] + fn,
-                'file_size': int(fdata['size']),
-            }
-            for fn, fdata in data['files'].items()
-            if 'Video' in fdata['format']]
-
-        self._sort_formats(formats)
-
-        return {
-            '_type': 'video',
-            'id': video_id,
-            'title': title,
-            'formats': formats,
-            'description': description,
-            'uploader': uploader,
-            'upload_date': upload_date,
-            'thumbnail': data.get('misc', {}).get('image'),
-        }
+        metadata = self._download_json(
+            'http://archive.org/details/' + video_id, video_id, query={
+                'output': 'json',
+            })['metadata']
+        info.update({
+            'title': get_optional(metadata, 'title') or info.get('title'),
+            'description': clean_html(get_optional(metadata, 'description')),
+        })
+        if info.get('_type') != 'playlist':
+            info.update({
+                'uploader': get_optional(metadata, 'creator'),
+                'upload_date': unified_strdate(get_optional(metadata, 'date')),
+            })
+        return info

youtube_dl/extractor/bbc.py

@@ -5,11 +5,13 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    dict_get,
     ExtractorError,
     float_or_none,
     int_or_none,
     parse_duration,
     parse_iso8601,
+    try_get,
     unescapeHTML,
 )
 from ..compat import (
@@ -229,51 +231,6 @@ class BBCCoUkIE(InfoExtractor):
         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 
-    def _extract_connection(self, connection, programme_id):
-        formats = []
-        kind = connection.get('kind')
-        protocol = connection.get('protocol')
-        supplier = connection.get('supplier')
-        if protocol == 'http':
-            href = connection.get('href')
-            transfer_format = connection.get('transferFormat')
-            # ASX playlist
-            if supplier == 'asx':
-                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
-                    formats.append({
-                        'url': ref,
-                        'format_id': 'ref%s_%s' % (i, supplier),
-                    })
-            # Skip DASH until supported
-            elif transfer_format == 'dash':
-                pass
-            elif transfer_format == 'hls':
-                formats.extend(self._extract_m3u8_formats(
-                    href, programme_id, ext='mp4', entry_protocol='m3u8_native',
-                    m3u8_id=supplier, fatal=False))
-            # Direct link
-            else:
-                formats.append({
-                    'url': href,
-                    'format_id': supplier or kind or protocol,
-                })
-        elif protocol == 'rtmp':
-            application = connection.get('application', 'ondemand')
-            auth_string = connection.get('authString')
-            identifier = connection.get('identifier')
-            server = connection.get('server')
-            formats.append({
-                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
-                'play_path': identifier,
-                'app': '%s?%s' % (application, auth_string),
-                'page_url': 'http://www.bbc.co.uk',
-                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
-                'rtmp_live': False,
-                'ext': 'flv',
-                'format_id': supplier,
-            })
-        return formats
-
     def _extract_items(self, playlist):
         return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
 
@@ -294,46 +251,6 @@ class BBCCoUkIE(InfoExtractor):
     def _extract_connections(self, media):
         return self._findall_ns(media, './{%s}connection')
 
-    def _extract_video(self, media, programme_id):
-        formats = []
-        vbr = int_or_none(media.get('bitrate'))
-        vcodec = media.get('encoding')
-        service = media.get('service')
-        width = int_or_none(media.get('width'))
-        height = int_or_none(media.get('height'))
-        file_size = int_or_none(media.get('media_file_size'))
-        for connection in self._extract_connections(media):
-            conn_formats = self._extract_connection(connection, programme_id)
-            for format in conn_formats:
-                format.update({
-                    'width': width,
-                    'height': height,
-                    'vbr': vbr,
-                    'vcodec': vcodec,
-                    'filesize': file_size,
-                })
-                if service:
-                    format['format_id'] = '%s_%s' % (service, format['format_id'])
-            formats.extend(conn_formats)
-        return formats
-
-    def _extract_audio(self, media, programme_id):
-        formats = []
-        abr = int_or_none(media.get('bitrate'))
-        acodec = media.get('encoding')
-        service = media.get('service')
-        for connection in self._extract_connections(media):
-            conn_formats = self._extract_connection(connection, programme_id)
-            for format in conn_formats:
-                format.update({
-                    'format_id': '%s_%s' % (service, format['format_id']),
-                    'abr': abr,
-                    'acodec': acodec,
-                    'vcodec': 'none',
-                })
-            formats.extend(conn_formats)
-        return formats
-
     def _get_subtitles(self, media, programme_id):
         subtitles = {}
         for connection in self._extract_connections(media):
@@ -379,13 +296,87 @@ class BBCCoUkIE(InfoExtractor):
     def _process_media_selector(self, media_selection, programme_id):
         formats = []
         subtitles = None
+        urls = []
 
         for media in self._extract_medias(media_selection):
             kind = media.get('kind')
-            if kind == 'audio':
-                formats.extend(self._extract_audio(media, programme_id))
-            elif kind == 'video':
-                formats.extend(self._extract_video(media, programme_id))
+            if kind in ('video', 'audio'):
+                bitrate = int_or_none(media.get('bitrate'))
+                encoding = media.get('encoding')
+                service = media.get('service')
+                width = int_or_none(media.get('width'))
+                height = int_or_none(media.get('height'))
+                file_size = int_or_none(media.get('media_file_size'))
+                for connection in self._extract_connections(media):
+                    href = connection.get('href')
+                    if href in urls:
+                        continue
+                    if href:
+                        urls.append(href)
+                    conn_kind = connection.get('kind')
+                    protocol = connection.get('protocol')
+                    supplier = connection.get('supplier')
+                    transfer_format = connection.get('transferFormat')
+                    format_id = supplier or conn_kind or protocol
+                    if service:
+                        format_id = '%s_%s' % (service, format_id)
+                    # ASX playlist
+                    if supplier == 'asx':
+                        for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
+                            formats.append({
+                                'url': ref,
+                                'format_id': 'ref%s_%s' % (i, format_id),
+                            })
+                    elif transfer_format == 'dash':
+                        formats.extend(self._extract_mpd_formats(
+                            href, programme_id, mpd_id=format_id, fatal=False))
+                    elif transfer_format == 'hls':
+                        formats.extend(self._extract_m3u8_formats(
+                            href, programme_id, ext='mp4', entry_protocol='m3u8_native',
+                            m3u8_id=format_id, fatal=False))
+                    elif transfer_format == 'hds':
+                        formats.extend(self._extract_f4m_formats(
+                            href, programme_id, f4m_id=format_id, fatal=False))
+                    else:
+                        if not service and not supplier and bitrate:
+                            format_id += '-%d' % bitrate
+                        fmt = {
+                            'format_id': format_id,
+                            'filesize': file_size,
+                        }
+                        if kind == 'video':
+                            fmt.update({
+                                'width': width,
+                                'height': height,
+                                'vbr': bitrate,
+                                'vcodec': encoding,
+                            })
+                        else:
+                            fmt.update({
+                                'abr': bitrate,
+                                'acodec': encoding,
+                                'vcodec': 'none',
+                            })
+                        if protocol == 'http':
+                            # Direct link
+                            fmt.update({
+                                'url': href,
+                            })
+                        elif protocol == 'rtmp':
+                            application = connection.get('application', 'ondemand')
+                            auth_string = connection.get('authString')
+                            identifier = connection.get('identifier')
+                            server = connection.get('server')
+                            fmt.update({
+                                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
+                                'play_path': identifier,
+                                'app': '%s?%s' % (application, auth_string),
+                                'page_url': 'http://www.bbc.co.uk',
+                                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
+                                'rtmp_live': False,
+                                'ext': 'flv',
+                            })
+                        formats.append(fmt)
             elif kind == 'captions':
                 subtitles = self.extract_subtitles(media, programme_id)
         return formats, subtitles
@@ -589,7 +580,7 @@ class BBCIE(BBCCoUkIE):
         'info_dict': {
             'id': '150615_telabyad_kentin_cogu',
             'ext': 'mp4',
-            'title': "Tel Abyad'da IŞİD bayrağı indirildi YPG bayrağı çekildi",
+            'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
             'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
             'timestamp': 1434397334,
             'upload_date': '20150615',
@@ -654,6 +645,23 @@ class BBCIE(BBCCoUkIE):
             # rtmp download
             'skip_download': True,
         }
+    }, {
+        # single video embedded with Morph
+        'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
+        'info_dict': {
+            'id': 'p041vhd0',
+            'ext': 'mp4',
+            'title': "Nigeria v Japan - Men's First Round",
+            'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
+            'duration': 7980,
+            'uploader': 'BBC Sport',
+            'uploader_id': 'bbc_sport',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'skip': 'Georestricted to UK',
     }, {
         # single video with playlist.sxml URL in playlist param
         'url': 'http://www.bbc.com/sport/0/football/33653409',
@@ -751,7 +759,7 @@ class BBCIE(BBCCoUkIE):
 
         webpage = self._download_webpage(url, playlist_id)
 
-        json_ld_info = self._search_json_ld(webpage, playlist_id, default=None)
+        json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
         timestamp = json_ld_info.get('timestamp')
 
         playlist_title = json_ld_info.get('title')
@@ -820,13 +828,19 @@ class BBCIE(BBCCoUkIE):
                         # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
                         playlist = data_playable.get('otherSettings', {}).get('playlist', {})
                         if playlist:
-                            for key in ('progressiveDownload', 'streaming'):
+                            entry = None
+                            for key in ('streaming', 'progressiveDownload'):
                                 playlist_url = playlist.get('%sUrl' % key)
                                 if not playlist_url:
                                     continue
                                 try:
-                                    entries.append(self._extract_from_playlist_sxml(
-                                        playlist_url, playlist_id, timestamp))
+                                    info = self._extract_from_playlist_sxml(
+                                        playlist_url, playlist_id, timestamp)
+                                    if not entry:
+                                        entry = info
+                                    else:
+                                        entry['title'] = info['title']
+                                        entry['formats'].extend(info['formats'])
                                 except Exception as e:
                                     # Some playlist URL may fail with 500, at the same time
                                     # the other one may work fine (e.g.
@@ -834,6 +848,9 @@ class BBCIE(BBCCoUkIE):
                                     if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
                                         continue
                                     raise
+                            if entry:
+                                self._sort_formats(entry['formats'])
+                                entries.append(entry)
 
         if entries:
             return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
@@ -866,6 +883,50 @@ class BBCIE(BBCCoUkIE):
                 'subtitles': subtitles,
             }
 
+        # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
+        # There are several setPayload calls may be present but the video
+        # seems to be always related to the first one
+        morph_payload = self._parse_json(
+            self._search_regex(
+                r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
+                webpage, 'morph payload', default='{}'),
+            playlist_id, fatal=False)
+        if morph_payload:
+            components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
+            for component in components:
+                if not isinstance(component, dict):
+                    continue
+                lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
+                if not lead_media:
+                    continue
+                identifiers = lead_media.get('identifiers')
+                if not identifiers or not isinstance(identifiers, dict):
+                    continue
+                programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
+                if not programme_id:
+                    continue
+                title = lead_media.get('title') or self._og_search_title(webpage)
+                formats, subtitles = self._download_media_selector(programme_id)
+                self._sort_formats(formats)
+                description = lead_media.get('summary')
+                uploader = lead_media.get('masterBrand')
+                uploader_id = lead_media.get('mid')
+                duration = None
+                duration_d = lead_media.get('duration')
+                if isinstance(duration_d, dict):
+                    duration = parse_duration(dict_get(
+                        duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
+                return {
+                    'id': programme_id,
+                    'title': title,
+                    'description': description,
+                    'duration': duration,
+                    'uploader': uploader,
+                    'uploader_id': uploader_id,
+                    'formats': formats,
+                    'subtitles': subtitles,
+                }
+
         def extract_all(pattern):
             return list(filter(None, map(
                 lambda s: self._parse_json(s, playlist_id, fatal=False),
@@ -883,7 +944,7 @@ class BBCIE(BBCCoUkIE):
             r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
         if entries:
             return self.playlist_result(
-                [self.url_result(entry, 'BBCCoUk') for entry in entries],
+                [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
                 playlist_id, playlist_title, playlist_description)
 
         # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)

youtube_dl/extractor/bigflix.py

@@ -11,15 +11,6 @@ from ..compat import compat_urllib_parse_unquote
 class BigflixIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?bigflix\.com/.+/(?P<id>[0-9]+)'
     _TESTS = [{
-        'url': 'http://www.bigflix.com/Hindi-movies/Action-movies/Singham-Returns/16537',
-        'md5': 'dc1b4aebb46e3a7077ecc0d9f43f61e3',
-        'info_dict': {
-            'id': '16537',
-            'ext': 'mp4',
-            'title': 'Singham Returns',
-            'description': 'md5:3d2ba5815f14911d5cc6a501ae0cf65d',
-        }
-    }, {
         # 2 formats
         'url': 'http://www.bigflix.com/Tamil-movies/Drama-movies/Madarasapatinam/16070',
         'info_dict': {

youtube_dl/extractor/bilibili.py

@@ -25,13 +25,13 @@ class BiliBiliIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.bilibili.tv/video/av1074402/',
-        'md5': '5f7d29e1a2872f3df0cf76b1f87d3788',
+        'md5': '9fa226fe2b8a9a4d5a69b4c6a183417e',
         'info_dict': {
             'id': '1554319',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': '【金坷垃】金泡沫',
             'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
-            'duration': 308.067,
+            'duration': 308.315,
             'timestamp': 1398012660,
             'upload_date': '20140420',
             'thumbnail': 're:^https?://.+\.jpg',
@@ -41,23 +41,20 @@ class BiliBiliIE(InfoExtractor):
     }, {
         'url': 'http://www.bilibili.com/video/av1041170/',
         'info_dict': {
-            'id': '1041170',
+            'id': '1507019',
+            '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！~',
+            'timestamp': 1396530060,
+            'upload_date': '20140403',
+            'uploader': '枫叶逝去',
+            'uploader_id': '520116',
         },
-        'playlist_count': 9,
     }, {
         'url': 'http://www.bilibili.com/video/av4808130/',
         'info_dict': {
-            'id': '4808130',
-            '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',
-        },
-        'playlist': [{
-            'md5': '55cdadedf3254caaa0d5d27cf20a8f9c',
-            'info_dict': {
-                'id': '4808130_part1',
-                'ext': 'flv',
+            'id': '7802182',
+            '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',
             'timestamp': 1464564180,
@@ -65,49 +62,12 @@ class BiliBiliIE(InfoExtractor):
             'uploader': '喜欢拉面',
             'uploader_id': '151066',
         },
-        }, {
-            'md5': '926f9f67d0c482091872fbd8eca7ea3d',
-            'info_dict': {
-                'id': '4808130_part2',
-                'ext': 'flv',
-                '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',
-                'timestamp': 1464564180,
-                'upload_date': '20160529',
-                'uploader': '喜欢拉面',
-                'uploader_id': '151066',
-            },
-        }, {
-            'md5': '4b7b225b968402d7c32348c646f1fd83',
-            'info_dict': {
-                'id': '4808130_part3',
-                'ext': 'flv',
-                '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',
-                'timestamp': 1464564180,
-                'upload_date': '20160529',
-                'uploader': '喜欢拉面',
-                'uploader_id': '151066',
-            },
-        }, {
-            'md5': '7b795e214166501e9141139eea236e91',
-            'info_dict': {
-                'id': '4808130_part4',
-                'ext': 'flv',
-                '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',
-                'timestamp': 1464564180,
-                'upload_date': '20160529',
-                'uploader': '喜欢拉面',
-                'uploader_id': '151066',
-            },
-        }],
     }, {
         # Missing upload time
         'url': 'http://www.bilibili.com/video/av1867637/',
         'info_dict': {
             'id': '2880301',
-            'ext': 'flv',
+            '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】',
             'uploader': '黑夜为猫',

youtube_dl/extractor/biqle.py

@@ -24,7 +24,8 @@ class BIQLEIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Ребенок в шоке от автоматической мойки',
             'uploader': 'Dmitry Kotov',
-        }
+        },
+        'skip': ' This video was marked as adult.  Embedding adult videos on external sites is prohibited.',
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/bloomberg.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -20,6 +21,18 @@ class BloombergIE(InfoExtractor):
         'params': {
             'format': 'best[format_id^=hds]',
         },
+    }, {
+        # video ID in BPlayer(...)
+        'url': 'http://www.bloomberg.com/features/2016-hello-world-new-zealand/',
+        'info_dict': {
+            'id': '938c7e72-3f25-4ddb-8b85-a9be731baa74',
+            'ext': 'flv',
+            'title': 'Meet the Real-Life Tech Wizards of Middle Earth',
+            'description': 'Hello World, Episode 1: New Zealand’s freaky AI babies, robot exoskeletons, and a virtual you.',
+        },
+        'params': {
+            'format': 'best[format_id^=hds]',
+        },
     }, {
         'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
         'only_matching': True,
@@ -33,7 +46,11 @@ class BloombergIE(InfoExtractor):
         webpage = self._download_webpage(url, name)
         video_id = self._search_regex(
             r'["\']bmmrId["\']\s*:\s*(["\'])(?P<url>.+?)\1',
-            webpage, 'id', group='url')
+            webpage, 'id', group='url', default=None)
+        if not video_id:
+            bplayer_data = self._parse_json(self._search_regex(
+                r'BPlayer\(null,\s*({[^;]+})\);', webpage, 'id'), name)
+            video_id = bplayer_data['id']
         title = re.sub(': Video$', '', self._og_search_title(webpage))
 
         embed_info = self._download_json(

youtube_dl/extractor/cbslocal.py

@@ -1,12 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import calendar
-import datetime
-
 from .anvato import AnvatoIE
 from .sendtonews import SendtoNewsIE
 from ..compat import compat_urlparse
+from ..utils import unified_timestamp
 
 
 class CBSLocalIE(AnvatoIE):
@@ -71,10 +69,7 @@ class CBSLocalIE(AnvatoIE):
 
         time_str = self._html_search_regex(
             r'class="entry-date">([^<]+)<', webpage, 'released date', fatal=False)
-        timestamp = None
-        if time_str:
-            timestamp = calendar.timegm(datetime.datetime.strptime(
-                time_str, '%b %d, %Y %I:%M %p').timetuple())
+        timestamp = unified_timestamp(time_str)
 
         info_dict.update({
             'display_id': display_id,

youtube_dl/extractor/chaturbate.py

@@ -17,7 +17,8 @@ class ChaturbateIE(InfoExtractor):
         },
         'params': {
             'skip_download': True,
-        }
+        },
+        'skip': 'Room is offline',
     }, {
         'url': 'https://en.chaturbate.com/siswet19/',
         'only_matching': True,

youtube_dl/extractor/chirbit.py

@@ -1,30 +1,33 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import base64
+
 from .common import InfoExtractor
-from ..utils import (
-    parse_duration,
-    int_or_none,
-)
+from ..utils import parse_duration
 
 
 class ChirbitIE(InfoExtractor):
     IE_NAME = 'chirbit'
     _VALID_URL = r'https?://(?:www\.)?chirb\.it/(?:(?:wp|pl)/|fb_chirbit_player\.swf\?key=)?(?P<id>[\da-zA-Z]+)'
     _TESTS = [{
-        'url': 'http://chirb.it/PrIPv5',
-        'md5': '9847b0dad6ac3e074568bf2cfb197de8',
+        'url': 'http://chirb.it/be2abG',
         'info_dict': {
-            'id': 'PrIPv5',
+            'id': 'be2abG',
             'ext': 'mp3',
-            'title': 'Фасадстрой',
-            'duration': 52,
-            'view_count': int,
-            'comment_count': int,
+            'title': 'md5:f542ea253f5255240be4da375c6a5d7e',
+            'description': 'md5:f24a4e22a71763e32da5fed59e47c770',
+            'duration': 306,
+        },
+        'params': {
+            'skip_download': True,
         }
     }, {
         'url': 'https://chirb.it/fb_chirbit_player.swf?key=PrIPv5',
         'only_matching': True,
+    }, {
+        'url': 'https://chirb.it/wp/MN58c2',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -33,27 +36,30 @@ class ChirbitIE(InfoExtractor):
         webpage = self._download_webpage(
             'http://chirb.it/%s' % audio_id, audio_id)
 
-        audio_url = self._search_regex(
-            r'"setFile"\s*,\s*"([^"]+)"', webpage, 'audio url')
+        data_fd = self._search_regex(
+            r'data-fd=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'data fd', group='url')
+
+        # Reverse engineered from https://chirb.it/js/chirbit.player.js (look
+        # for soundURL)
+        audio_url = base64.b64decode(
+            data_fd[::-1].encode('ascii')).decode('utf-8')
 
         title = self._search_regex(
-            r'itemprop="name">([^<]+)', webpage, 'title')
-        duration = parse_duration(self._html_search_meta(
-            'duration', webpage, 'duration', fatal=False))
-        view_count = int_or_none(self._search_regex(
-            r'itemprop="playCount"\s*>(\d+)', webpage,
-            'listen count', fatal=False))
-        comment_count = int_or_none(self._search_regex(
-            r'>(\d+) Comments?:', webpage,
-            'comment count', fatal=False))
+            r'class=["\']chirbit-title["\'][^>]*>([^<]+)', webpage, 'title')
+        description = self._search_regex(
+            r'<h3>Description</h3>\s*<pre[^>]*>([^<]+)</pre>',
+            webpage, 'description', default=None)
+        duration = parse_duration(self._search_regex(
+            r'class=["\']c-length["\'][^>]*>([^<]+)',
+            webpage, 'duration', fatal=False))
 
         return {
             'id': audio_id,
             'url': audio_url,
             'title': title,
+            'description': description,
             'duration': duration,
-            'view_count': view_count,
-            'comment_count': comment_count,
         }
 
 

youtube_dl/extractor/comedycentral.py

@@ -1,6 +1,7 @@
 from __future__ import unicode_literals
 
 from .mtv import MTVServicesInfoExtractor
+from .common import InfoExtractor
 
 
 class ComedyCentralIE(MTVServicesInfoExtractor):
@@ -96,3 +97,22 @@ class ComedyCentralTVIE(MTVServicesInfoExtractor):
             webpage, 'mrss url', group='url')
 
         return self._get_videos_info_from_url(mrss_url, video_id)
+
+
+class ComedyCentralShortnameIE(InfoExtractor):
+    _VALID_URL = r'^:(?P<id>tds|thedailyshow)$'
+    _TESTS = [{
+        'url': ':tds',
+        'only_matching': True,
+    }, {
+        'url': ':thedailyshow',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        shortcut_map = {
+            'tds': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes',
+            'thedailyshow': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes',
+        }
+        return self.url_result(shortcut_map[video_id])

youtube_dl/extractor/common.py

@@ -727,9 +727,14 @@ class InfoExtractor(object):
                     [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop)
 
     def _og_search_property(self, prop, html, name=None, **kargs):
+        if not isinstance(prop, (list, tuple)):
+            prop = [prop]
         if name is None:
-            name = 'OpenGraph %s' % prop
-        escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs)
+            name = 'OpenGraph %s' % prop[0]
+        og_regexes = []
+        for p in prop:
+            og_regexes.extend(self._og_regexes(p))
+        escaped = self._search_regex(og_regexes, html, name, flags=re.DOTALL, **kargs)
         if escaped is None:
             return None
         return unescapeHTML(escaped)
@@ -811,11 +816,14 @@ class InfoExtractor(object):
         json_ld = self._search_regex(
             r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>',
             html, 'JSON-LD', group='json_ld', **kwargs)
+        default = kwargs.get('default', NO_DEFAULT)
         if not json_ld:
-            return {}
-        return self._json_ld(
-            json_ld, video_id, fatal=kwargs.get('fatal', True),
-            expected_type=expected_type)
+            return default if default is not NO_DEFAULT else {}
+        # JSON-LD may be malformed and thus `fatal` should be respected.
+        # At the same time `default` may be passed that assumes `fatal=False`
+        # for _search_regex. Let's simulate the same behavior here as well.
+        fatal = kwargs.get('fatal', True) if default == NO_DEFAULT else False
+        return self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type)
 
     def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None):
         if isinstance(json_ld, compat_str):
@@ -823,41 +831,47 @@ class InfoExtractor(object):
         if not json_ld:
             return {}
         info = {}
-        if json_ld.get('@context') == 'http://schema.org':
-            item_type = json_ld.get('@type')
+        if not isinstance(json_ld, (list, tuple, dict)):
+            return info
+        if isinstance(json_ld, dict):
+            json_ld = [json_ld]
+        for e in json_ld:
+            if e.get('@context') == 'http://schema.org':
+                item_type = e.get('@type')
                 if expected_type is not None and expected_type != item_type:
                     return info
                 if item_type == 'TVEpisode':
                     info.update({
-                    'episode': unescapeHTML(json_ld.get('name')),
-                    'episode_number': int_or_none(json_ld.get('episodeNumber')),
-                    'description': unescapeHTML(json_ld.get('description')),
+                        'episode': unescapeHTML(e.get('name')),
+                        'episode_number': int_or_none(e.get('episodeNumber')),
+                        'description': unescapeHTML(e.get('description')),
                     })
-                part_of_season = json_ld.get('partOfSeason')
+                    part_of_season = e.get('partOfSeason')
                     if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason':
                         info['season_number'] = int_or_none(part_of_season.get('seasonNumber'))
-                part_of_series = json_ld.get('partOfSeries')
+                    part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries')
                     if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries':
                         info['series'] = unescapeHTML(part_of_series.get('name'))
                 elif item_type == 'Article':
                     info.update({
-                    'timestamp': parse_iso8601(json_ld.get('datePublished')),
-                    'title': unescapeHTML(json_ld.get('headline')),
-                    'description': unescapeHTML(json_ld.get('articleBody')),
+                        'timestamp': parse_iso8601(e.get('datePublished')),
+                        'title': unescapeHTML(e.get('headline')),
+                        'description': unescapeHTML(e.get('articleBody')),
                     })
                 elif item_type == 'VideoObject':
                     info.update({
-                    'url': json_ld.get('contentUrl'),
-                    'title': unescapeHTML(json_ld.get('name')),
-                    'description': unescapeHTML(json_ld.get('description')),
-                    'thumbnail': json_ld.get('thumbnailUrl'),
-                    'duration': parse_duration(json_ld.get('duration')),
-                    'timestamp': unified_timestamp(json_ld.get('uploadDate')),
-                    'filesize': float_or_none(json_ld.get('contentSize')),
-                    'tbr': int_or_none(json_ld.get('bitrate')),
-                    'width': int_or_none(json_ld.get('width')),
-                    'height': int_or_none(json_ld.get('height')),
+                        'url': e.get('contentUrl'),
+                        'title': unescapeHTML(e.get('name')),
+                        'description': unescapeHTML(e.get('description')),
+                        'thumbnail': e.get('thumbnailUrl'),
+                        'duration': parse_duration(e.get('duration')),
+                        'timestamp': unified_timestamp(e.get('uploadDate')),
+                        'filesize': float_or_none(e.get('contentSize')),
+                        'tbr': int_or_none(e.get('bitrate')),
+                        'width': int_or_none(e.get('width')),
+                        'height': int_or_none(e.get('height')),
                     })
+                break
         return dict((k, v) for k, v in info.items() if v is not None)
 
     @staticmethod
@@ -911,7 +925,8 @@ class InfoExtractor(object):
                 if f.get('ext') in ['f4f', 'f4m']:  # Not yet supported
                     preference -= 0.5
 
-            proto_preference = 0 if determine_protocol(f) in ['http', 'https'] else -0.1
+            protocol = f.get('protocol') or determine_protocol(f)
+            proto_preference = 0 if protocol in ['http', 'https'] else (-0.5 if protocol == 'rtsp' else -0.1)
 
             if f.get('vcodec') == 'none':  # audio only
                 preference -= 50
@@ -1128,7 +1143,7 @@ class InfoExtractor(object):
             'url': m3u8_url,
             'ext': ext,
             'protocol': 'm3u8',
-            'preference': preference - 1 if preference else -1,
+            'preference': preference - 100 if preference else -100,
             'resolution': 'multiple',
             'format_note': 'Quality selection URL',
         }

youtube_dl/extractor/condenast.py

@@ -5,13 +5,17 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse_urlencode,
     compat_urllib_parse_urlparse,
     compat_urlparse,
 )
 from ..utils import (
     orderedSet,
     remove_end,
+    extract_attributes,
+    mimetype2ext,
+    determine_ext,
+    int_or_none,
+    parse_iso8601,
 )
 
 
@@ -58,6 +62,9 @@ class CondeNastIE(InfoExtractor):
             'ext': 'mp4',
             'title': '3D Printed Speakers Lit With LED',
             'description': 'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
+            'uploader': 'wired',
+            'upload_date': '20130314',
+            'timestamp': 1363219200,
         }
     }, {
         # JS embed
@@ -67,70 +74,93 @@ class CondeNastIE(InfoExtractor):
             'id': '55f9cf8b61646d1acf00000c',
             'ext': 'mp4',
             'title': '3D printed TSA Travel Sentry keys really do open TSA locks',
+            'uploader': 'arstechnica',
+            'upload_date': '20150916',
+            'timestamp': 1442434955,
         }
     }]
 
     def _extract_series(self, url, webpage):
-        title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
-                                        webpage, 'series title', flags=re.DOTALL)
+        title = self._html_search_regex(
+            r'(?s)<div class="cne-series-info">.*?<h1>(.+?)</h1>',
+            webpage, 'series title')
         url_object = compat_urllib_parse_urlparse(url)
         base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
-        m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]',
-                              webpage, flags=re.DOTALL)
+        m_paths = re.finditer(
+            r'(?s)<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]', webpage)
         paths = orderedSet(m.group(1) for m in m_paths)
         build_url = lambda path: compat_urlparse.urljoin(base_url, path)
         entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
         return self.playlist_result(entries, playlist_title=title)
 
     def _extract_video(self, webpage, url_type):
-        if url_type != 'embed':
-            description = self._html_search_regex(
-                [
-                    r'<div class="cne-video-description">(.+?)</div>',
-                    r'<div class="video-post-content">(.+?)</div>',
-                ],
-                webpage, 'description', fatal=False, flags=re.DOTALL)
-        else:
-            description = None
-        params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
-                                    'player params', flags=re.DOTALL)
-        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
-        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id')
-        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target')
-        data = compat_urllib_parse_urlencode({'videoId': video_id,
-                                              'playerId': player_id,
-                                              'target': target,
+        query = {}
+        params = self._search_regex(
+            r'(?s)var params = {(.+?)}[;,]', webpage, 'player params', default=None)
+        if params:
+            query.update({
+                'videoId': self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id'),
+                'playerId': self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id'),
+                'target': self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target'),
             })
-        base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
-                                           webpage, 'base info url',
-                                           default='http://player.cnevids.com/player/loader.js?')
-        info_url = base_info_url + data
-        info_page = self._download_webpage(info_url, video_id,
-                                           'Downloading video info')
-        video_info = self._search_regex(r'var\s+video\s*=\s*({.+?});', info_page, 'video info')
-        video_info = self._parse_json(video_info, video_id)
+        else:
+            params = extract_attributes(self._search_regex(
+                r'(<[^>]+data-js="video-player"[^>]+>)',
+                webpage, 'player params element'))
+            query.update({
+                'videoId': params['data-video'],
+                'playerId': params['data-player'],
+                'target': params['id'],
+            })
+        video_id = query['videoId']
+        video_info = None
+        info_page = self._download_webpage(
+            'http://player.cnevids.com/player/video.js',
+            video_id, 'Downloading video info', query=query, fatal=False)
+        if info_page:
+            video_info = self._parse_json(self._search_regex(
+                r'loadCallback\(({.+})\)', info_page, 'video info'), video_id)['video']
+        else:
+            info_page = self._download_webpage(
+                'http://player.cnevids.com/player/loader.js',
+                video_id, 'Downloading loader info', query=query)
+            video_info = self._parse_json(self._search_regex(
+                r'var\s+video\s*=\s*({.+?});', info_page, 'video info'), video_id)
+        title = video_info['title']
 
-        formats = [{
-            'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
-            'url': fdata['src'],
-            'ext': fdata['type'].split('/')[-1],
-            'quality': 1 if fdata['quality'] == 'high' else 0,
-        } for fdata in video_info['sources'][0]]
+        formats = []
+        for fdata in video_info.get('sources', [{}])[0]:
+            src = fdata.get('src')
+            if not src:
+                continue
+            ext = mimetype2ext(fdata.get('type')) or determine_ext(src)
+            quality = fdata.get('quality')
+            formats.append({
+                'format_id': ext + ('-%s' % quality if quality else ''),
+                'url': src,
+                'ext': ext,
+                'quality': 1 if quality == 'high' else 0,
+            })
         self._sort_formats(formats)
 
-        return {
+        info = self._search_json_ld(
+            webpage, video_id, fatal=False) if url_type != 'embed' else {}
+        info.update({
             'id': video_id,
             'formats': formats,
-            'title': video_info['title'],
-            'thumbnail': video_info['poster_frame'],
-            'description': description,
-        }
+            'title': title,
+            'thumbnail': video_info.get('poster_frame'),
+            'uploader': video_info.get('brand'),
+            'duration': int_or_none(video_info.get('duration')),
+            'tags': video_info.get('tags'),
+            'series': video_info.get('series_title'),
+            'season': video_info.get('season_title'),
+            'timestamp': parse_iso8601(video_info.get('premiere_date')),
+        })
+        return info
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        site = mobj.group('site')
-        url_type = mobj.group('type')
-        item_id = mobj.group('id')
+        site, url_type, item_id = re.match(self._VALID_URL, url).groups()
 
         # Convert JS embed to regular embed
         if url_type == 'embedjs':

youtube_dl/extractor/crunchyroll.py

@@ -114,6 +114,21 @@ class CrunchyrollIE(CrunchyrollBaseIE):
             # rtmp
             'skip_download': True,
         },
+    }, {
+        'url': 'http://www.crunchyroll.com/rezero-starting-life-in-another-world-/episode-5-the-morning-of-our-promise-is-still-distant-702409',
+        'info_dict': {
+            'id': '702409',
+            'ext': 'mp4',
+            'title': 'Re:ZERO -Starting Life in Another World- Episode 5 – The Morning of Our Promise Is Still Distant',
+            'description': 'md5:97664de1ab24bbf77a9c01918cb7dca9',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'TV TOKYO',
+            'upload_date': '20160508',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }, {
         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
         'only_matching': True,
@@ -336,9 +351,18 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             if video_encode_id in video_encode_ids:
                 continue
             video_encode_ids.append(video_encode_id)
+
+            video_file = xpath_text(stream_info, './file')
+            if not video_file:
+                continue
+            if video_file.startswith('http'):
+                formats.extend(self._extract_m3u8_formats(
+                    video_file, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+                continue
+
             video_url = xpath_text(stream_info, './host')
-            video_play_path = xpath_text(stream_info, './file')
-            if not video_url or not video_play_path:
+            if not video_url:
                 continue
             metadata = stream_info.find('./metadata')
             format_info = {
@@ -353,7 +377,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                 parsed_video_url = compat_urlparse.urlparse(video_url)
                 direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
                     netloc='v.lvlt.crcdn.net',
-                    path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_play_path.split(':')[-1])))
+                    path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_file.split(':')[-1])))
                 if self._is_valid_url(direct_video_url, video_id, video_format):
                     format_info.update({
                         'url': direct_video_url,
@@ -363,7 +387,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
             format_info.update({
                 'url': video_url,
-                'play_path': video_play_path,
+                'play_path': video_file,
                 'ext': 'flv',
             })
             formats.append(format_info)

youtube_dl/extractor/ctsnews.py

@@ -1,13 +1,12 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import parse_iso8601, ExtractorError
+from ..utils import unified_timestamp
 
 
 class CtsNewsIE(InfoExtractor):
     IE_DESC = '華視新聞'
-    # https connection failed (Connection reset)
     _VALID_URL = r'https?://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html'
     _TESTS = [{
         'url': 'http://news.cts.com.tw/cts/international/201501/201501291578109.html',
@@ -16,7 +15,7 @@ class CtsNewsIE(InfoExtractor):
             'id': '201501291578109',
             'ext': 'mp4',
             'title': '以色列.真主黨交火 3人死亡',
-            'description': 'md5:95e9b295c898b7ff294f09d450178d7d',
+            'description': '以色列和黎巴嫩真主黨，爆發五年最嚴重衝突，雙方砲轟交火，兩名以軍死亡，還有一名西班牙籍的聯合國維和人...',
             'timestamp': 1422528540,
             'upload_date': '20150129',
         }
@@ -28,7 +27,7 @@ class CtsNewsIE(InfoExtractor):
             'id': '201309031304098',
             'ext': 'mp4',
             'title': '韓國31歲童顏男 貌如十多歲小孩',
-            'description': 'md5:f183feeba3752b683827aab71adad584',
+            'description': '越有年紀的人，越希望看起來年輕一點，而南韓卻有一位31歲的男子，看起來像是11、12歲的小孩，身...',
             'thumbnail': 're:^https?://.*\.jpg$',
             'timestamp': 1378205880,
             'upload_date': '20130903',
@@ -36,8 +35,7 @@ class CtsNewsIE(InfoExtractor):
     }, {
         # With Youtube embedded video
         'url': 'http://news.cts.com.tw/cts/money/201501/201501291578003.html',
-        'md5': '1d842c771dc94c8c3bca5af2cc1db9c5',
-        'add_ie': ['Youtube'],
+        'md5': 'e4726b2ccd70ba2c319865e28f0a91d1',
         'info_dict': {
             'id': 'OVbfO7d0_hQ',
             'ext': 'mp4',
@@ -47,42 +45,37 @@ class CtsNewsIE(InfoExtractor):
             'upload_date': '20150128',
             'uploader_id': 'TBSCTS',
             'uploader': '中華電視公司',
-        }
+        },
+        'add_ie': ['Youtube'],
     }]
 
     def _real_extract(self, url):
         news_id = self._match_id(url)
         page = self._download_webpage(url, news_id)
 
-        if self._search_regex(r'(CTSPlayer2)', page, 'CTSPlayer2 identifier', default=None):
-            feed_url = self._html_search_regex(
-                r'(http://news\.cts\.com\.tw/action/mp4feed\.php\?news_id=\d+)',
-                page, 'feed url')
-            video_url = self._download_webpage(
-                feed_url, news_id, note='Fetching feed')
+        news_id = self._hidden_inputs(page).get('get_id')
+
+        if news_id:
+            mp4_feed = self._download_json(
+                'http://news.cts.com.tw/action/test_mp4feed.php',
+                news_id, note='Fetching feed', query={'news_id': news_id})
+            video_url = mp4_feed['source_url']
         else:
             self.to_screen('Not CTSPlayer video, trying Youtube...')
             youtube_url = self._search_regex(
-                r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url',
-                default=None)
-            if not youtube_url:
-                raise ExtractorError('The news includes no videos!', expected=True)
+                r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url')
 
-            return {
-                '_type': 'url',
-                'url': youtube_url,
-                'ie_key': 'Youtube',
-            }
+            return self.url_result(youtube_url, ie='Youtube')
 
         description = self._html_search_meta('description', page)
-        title = self._html_search_meta('title', page)
+        title = self._html_search_meta('title', page, fatal=True)
         thumbnail = self._html_search_meta('image', page)
 
         datetime_str = self._html_search_regex(
-            r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time')
-        # Transform into ISO 8601 format with timezone info
-        datetime_str = datetime_str.replace('/', '-') + ':00+0800'
-        timestamp = parse_iso8601(datetime_str, delimiter=' ')
+            r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time', fatal=False)
+        timestamp = None
+        if datetime_str:
+            timestamp = unified_timestamp(datetime_str) - 8 * 3600
 
         return {
             'id': news_id,

youtube_dl/extractor/cwtv.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CWTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?cw(?:tv|seed)\.com/(?:shows/)?(?:[^/]+/){2}\?.*\bplay=(?P<id>[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12})'
+    _VALID_URL = r'https?://(?:www\.)?cw(?:tv(?:pr)?|seed)\.com/(?:shows/)?(?:[^/]+/)+[^?]*\?.*\b(?:play|watch)=(?P<id>[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12})'
     _TESTS = [{
         'url': 'http://cwtv.com/shows/arrow/legends-of-yesterday/?play=6b15e985-9345-4f60-baf8-56e96be57c63',
         'info_dict': {
@@ -28,7 +28,8 @@ class CWTVIE(InfoExtractor):
         'params': {
             # m3u8 download
             'skip_download': True,
-        }
+        },
+        'skip': 'redirect to http://cwtv.com/shows/arrow/',
     }, {
         'url': 'http://www.cwseed.com/shows/whose-line-is-it-anyway/jeff-davis-4/?play=24282b12-ead2-42f2-95ad-26770c2c6088',
         'info_dict': {
@@ -44,22 +45,43 @@ class CWTVIE(InfoExtractor):
             'upload_date': '20151006',
             'timestamp': 1444107300,
         },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        }
     }, {
         'url': 'http://cwtv.com/thecw/chroniclesofcisco/?play=8adebe35-f447-465f-ab52-e863506ff6d6',
         'only_matching': True,
+    }, {
+        'url': 'http://cwtvpr.com/the-cw/video?watch=9eee3f60-ef4e-440b-b3b2-49428ac9c54e',
+        'only_matching': True,
+    }, {
+        'url': 'http://cwtv.com/shows/arrow/legends-of-yesterday/?watch=6b15e985-9345-4f60-baf8-56e96be57c63',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        video_data = self._download_json(
-            'http://metaframe.digitalsmiths.tv/v2/CWtv/assets/%s/partner/132?format=json' % video_id, video_id)
-
-        formats = self._extract_m3u8_formats(
-            video_data['videos']['variantplaylist']['uri'], video_id, 'mp4')
+        video_data = None
+        formats = []
+        for partner in (154, 213):
+            vdata = self._download_json(
+                'http://metaframe.digitalsmiths.tv/v2/CWtv/assets/%s/partner/%d?format=json' % (video_id, partner), video_id, fatal=False)
+            if not vdata:
+                continue
+            video_data = vdata
+            for quality, quality_data in vdata.get('videos', {}).items():
+                quality_url = quality_data.get('uri')
+                if not quality_url:
+                    continue
+                if quality == 'variantplaylist':
+                    formats.extend(self._extract_m3u8_formats(
+                        quality_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
+                else:
+                    tbr = int_or_none(quality_data.get('bitrate'))
+                    format_id = 'http' + ('-%d' % tbr if tbr else '')
+                    if self._is_valid_url(quality_url, video_id, format_id):
+                        formats.append({
+                            'format_id': format_id,
+                            'url': quality_url,
+                            'tbr': tbr,
+                        })
         self._sort_formats(formats)
 
         thumbnails = [{

youtube_dl/extractor/discoverygo.py

@@ -0,0 +1,98 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    extract_attributes,
+    int_or_none,
+    parse_age_limit,
+    unescapeHTML,
+)
+
+
+class DiscoveryGoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?discoverygo\.com/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://www.discoverygo.com/love-at-first-kiss/kiss-first-ask-questions-later/',
+        'info_dict': {
+            'id': '57a33c536b66d1cd0345eeb1',
+            'ext': 'mp4',
+            'title': 'Kiss First, Ask Questions Later!',
+            'description': 'md5:fe923ba34050eae468bffae10831cb22',
+            'duration': 2579,
+            'series': 'Love at First Kiss',
+            'season_number': 1,
+            'episode_number': 1,
+            'age_limit': 14,
+        },
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        container = extract_attributes(
+            self._search_regex(
+                r'(<div[^>]+class=["\']video-player-container[^>]+>)',
+                webpage, 'video container'))
+
+        video = self._parse_json(
+            unescapeHTML(container.get('data-video') or container.get('data-json')),
+            display_id)
+
+        title = video['name']
+
+        stream = video['stream']
+        STREAM_URL_SUFFIX = 'streamUrl'
+        formats = []
+        for stream_kind in ('', 'hds'):
+            suffix = STREAM_URL_SUFFIX.capitalize() if stream_kind else STREAM_URL_SUFFIX
+            stream_url = stream.get('%s%s' % (stream_kind, suffix))
+            if not stream_url:
+                continue
+            if stream_kind == '':
+                formats.extend(self._extract_m3u8_formats(
+                    stream_url, display_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif stream_kind == 'hds':
+                formats.extend(self._extract_f4m_formats(
+                    stream_url, display_id, f4m_id=stream_kind, fatal=False))
+        self._sort_formats(formats)
+
+        video_id = video.get('id') or display_id
+        description = video.get('description', {}).get('detailed')
+        duration = int_or_none(video.get('duration'))
+
+        series = video.get('show', {}).get('name')
+        season_number = int_or_none(video.get('season', {}).get('number'))
+        episode_number = int_or_none(video.get('episodeNumber'))
+
+        tags = video.get('tags')
+        age_limit = parse_age_limit(video.get('parental', {}).get('rating'))
+
+        subtitles = {}
+        captions = stream.get('captions')
+        if isinstance(captions, list):
+            for caption in captions:
+                subtitle_url = caption.get('fileUrl')
+                if (not subtitle_url or not isinstance(subtitle_url, compat_str) or
+                        not subtitle_url.startswith('http')):
+                    continue
+                lang = caption.get('fileLang', 'en')
+                subtitles.setdefault(lang, []).append({'url': subtitle_url})
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'series': series,
+            'season_number': season_number,
+            'episode_number': episode_number,
+            'tags': tags,
+            'age_limit': age_limit,
+            'formats': formats,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/drtuber.py

@@ -3,7 +3,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import str_to_int
+from ..utils import (
+    NO_DEFAULT,
+    str_to_int,
+)
 
 
 class DrTuberIE(InfoExtractor):
@@ -17,7 +20,6 @@ class DrTuberIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'hot perky blonde naked golf',
             'like_count': int,
-            'dislike_count': int,
             'comment_count': int,
             'categories': ['Babe', 'Blonde', 'Erotic', 'Outdoor', 'Softcore', 'Solo'],
             'thumbnail': 're:https?://.*\.jpg$',
@@ -36,25 +38,29 @@ class DrTuberIE(InfoExtractor):
             r'<source src="([^"]+)"', webpage, 'video URL')
 
         title = self._html_search_regex(
-            [r'<p[^>]+class="title_substrate">([^<]+)</p>', r'<title>([^<]+) - \d+'],
+            (r'class="title_watch"[^>]*><p>([^<]+)<',
+             r'<p[^>]+class="title_substrate">([^<]+)</p>',
+             r'<title>([^<]+) - \d+'),
             webpage, 'title')
 
         thumbnail = self._html_search_regex(
             r'poster="([^"]+)"',
             webpage, 'thumbnail', fatal=False)
 
-        def extract_count(id_, name):
+        def extract_count(id_, name, default=NO_DEFAULT):
             return str_to_int(self._html_search_regex(
                 r'<span[^>]+(?:class|id)="%s"[^>]*>([\d,\.]+)</span>' % id_,
-                webpage, '%s count' % name, fatal=False))
+                webpage, '%s count' % name, default=default, fatal=False))
 
         like_count = extract_count('rate_likes', 'like')
-        dislike_count = extract_count('rate_dislikes', 'dislike')
+        dislike_count = extract_count('rate_dislikes', 'dislike', default=None)
         comment_count = extract_count('comments_count', 'comment')
 
         cats_str = self._search_regex(
-            r'<div[^>]+class="categories_list">(.+?)</div>', webpage, 'categories', fatal=False)
-        categories = [] if not cats_str else re.findall(r'<a title="([^"]+)"', cats_str)
+            r'<div[^>]+class="categories_list">(.+?)</div>',
+            webpage, 'categories', fatal=False)
+        categories = [] if not cats_str else re.findall(
+            r'<a title="([^"]+)"', cats_str)
 
         return {
             'id': video_id,

youtube_dl/extractor/engadget.py

@@ -4,9 +4,10 @@ from .common import InfoExtractor
 
 
 class EngadgetIE(InfoExtractor):
-    _VALID_URL = r'https?://www.engadget.com/video/(?P<id>\d+)'
+    _VALID_URL = r'https?://www.engadget.com/video/(?P<id>[^/?#]+)'
 
-    _TEST = {
+    _TESTS = [{
+        # video with 5min ID
         'url': 'http://www.engadget.com/video/518153925/',
         'md5': 'c6820d4828a5064447a4d9fc73f312c9',
         'info_dict': {
@@ -15,8 +16,12 @@ class EngadgetIE(InfoExtractor):
             'title': 'Samsung Galaxy Tab Pro 8.4 Review',
         },
         'add_ie': ['FiveMin'],
-    }
+    }, {
+        # video with vidible ID
+        'url': 'https://www.engadget.com/video/57a28462134aa15a39f0421a/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        return self.url_result('5min:%s' % video_id)
+        return self.url_result('aol-video:%s' % video_id)

youtube_dl/extractor/expotv.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -12,23 +10,22 @@ from ..utils import (
 class ExpoTVIE(InfoExtractor):
     _VALID_URL = r'https?://www\.expotv\.com/videos/[^?#]*/(?P<id>[0-9]+)($|[?#])'
     _TEST = {
-        'url': 'http://www.expotv.com/videos/reviews/1/24/LinneCardscom/17561',
-        'md5': '2985e6d7a392b2f7a05e0ca350fe41d0',
+        'url': 'http://www.expotv.com/videos/reviews/3/40/NYX-Butter-lipstick/667916',
+        'md5': 'fe1d728c3a813ff78f595bc8b7a707a8',
         'info_dict': {
-            'id': '17561',
+            'id': '667916',
             'ext': 'mp4',
-            'upload_date': '20060212',
-            'title': 'My Favorite Online Scrapbook Store',
-            'view_count': int,
-            'description': 'You\'ll find most everything you need at this virtual store front.',
-            'uploader': 'Anna T.',
+            'title': 'NYX Butter Lipstick Little Susie',
+            'description': 'Goes on like butter, but looks better!',
             'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Stephanie S.',
+            'upload_date': '20150520',
+            'view_count': int,
         }
     }
 
     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)
         player_key = self._search_regex(
@@ -66,7 +63,7 @@ class ExpoTVIE(InfoExtractor):
             fatal=False)
         upload_date = unified_strdate(self._search_regex(
             r'<h5>Reviewed on ([0-9/.]+)</h5>', webpage, 'upload date',
-            fatal=False))
+            fatal=False), day_first=False)
 
         return {
             'id': video_id,

youtube_dl/extractor/extractors.py

@@ -159,6 +159,7 @@ from .coub import CoubIE
 from .collegerama import CollegeRamaIE
 from .comedycentral import (
     ComedyCentralIE,
+    ComedyCentralShortnameIE,
     ComedyCentralTVIE,
     ToshIE,
 )
@@ -220,6 +221,7 @@ from .dvtv import DVTVIE
 from .dumpert import DumpertIE
 from .defense import DefenseGouvFrIE
 from .discovery import DiscoveryIE
+from .discoverygo import DiscoveryGoIE
 from .dispeak import DigitallySpeakingIE
 from .dropbox import DropboxIE
 from .dw import (
@@ -270,10 +272,7 @@ from .fox import FOXIE
 from .foxgay import FoxgayIE
 from .foxnews import FoxNewsIE
 from .foxsports import FoxSportsIE
-from .franceculture import (
-    FranceCultureIE,
-    FranceCultureEmissionIE,
-)
+from .franceculture import FranceCultureIE
 from .franceinter import FranceInterIE
 from .francetv import (
     PluzzIE,
@@ -289,7 +288,6 @@ from .funimation import FunimationIE
 from .funnyordie import FunnyOrDieIE
 from .fusion import FusionIE
 from .gameinformer import GameInformerIE
-from .gamekings import GamekingsIE
 from .gameone import (
     GameOneIE,
     GameOnePlaylistIE,
@@ -310,7 +308,6 @@ from .globo import (
 )
 from .godtube import GodTubeIE
 from .godtv import GodTVIE
-from .goldenmoustache import GoldenMoustacheIE
 from .golem import GolemIE
 from .googledrive import GoogleDriveIE
 from .googleplus import GooglePlusIE
@@ -325,6 +322,7 @@ from .heise import HeiseIE
 from .hellporno import HellPornoIE
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
+from .hgtv import HGTVIE
 from .historicfilms import HistoricFilmsIE
 from .hitbox import HitboxIE, HitboxLiveIE
 from .hornbunny import HornBunnyIE
@@ -491,8 +489,9 @@ from .myvi import MyviIE
 from .myvideo import MyVideoIE
 from .myvidster import MyVidsterIE
 from .nationalgeographic import (
+    NationalGeographicVideoIE,
     NationalGeographicIE,
-    NationalGeographicChannelIE,
+    NationalGeographicEpisodeGuideIE,
 )
 from .naver import NaverIE
 from .nba import NBAIE
@@ -635,6 +634,7 @@ from .pluralsight import (
     PluralsightCourseIE,
 )
 from .podomatic import PodomaticIE
+from .pokemon import PokemonIE
 from .polskieradio import PolskieRadioIE
 from .porn91 import Porn91IE
 from .pornhd import PornHdIE
@@ -693,6 +693,7 @@ from .rockstargames import RockstarGamesIE
 from .roosterteeth import RoosterTeethIE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
+from .rozhlas import RozhlasIE
 from .rtbf import RTBFIE
 from .rte import RteIE, RteRadioIE
 from .rtlnl import RtlNlIE
@@ -752,6 +753,7 @@ from .smotri import (
 )
 from .snotr import SnotrIE
 from .sohu import SohuIE
+from .sonyliv import SonyLIVIE
 from .soundcloud import (
     SoundcloudIE,
     SoundcloudSetIE,
@@ -807,7 +809,6 @@ from .tagesschau import (
     TagesschauPlayerIE,
     TagesschauIE,
 )
-from .tapely import TapelyIE
 from .tass import TassIE
 from .tdslifeway import TDSLifewayIE
 from .teachertube import (
@@ -891,6 +892,7 @@ from .tvc import (
 from .tvigle import TvigleIE
 from .tvland import TVLandIE
 from .tvp import (
+    TVPEmbedIE,
     TVPIE,
     TVPSeriesIE,
 )
@@ -923,6 +925,7 @@ from .udemy import (
 from .udn import UDNEmbedIE
 from .digiteka import DigitekaIE
 from .unistra import UnistraIE
+from .uol import UOLIE
 from .urort import UrortIE
 from .urplay import URPlayIE
 from .usatoday import USATodayIE
@@ -1004,6 +1007,7 @@ from .vk import (
 )
 from .vlive import VLiveIE
 from .vodlocker import VodlockerIE
+from .vodplatform import VODPlatformIE
 from .voicerepublic import VoiceRepublicIE
 from .voxmedia import VoxMediaIE
 from .vporn import VpornIE

youtube_dl/extractor/fivemin.py

@@ -1,24 +1,11 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..compat import (
-    compat_parse_qs,
-    compat_urllib_parse_urlencode,
-    compat_urllib_parse_urlparse,
-    compat_urlparse,
-)
-from ..utils import (
-    ExtractorError,
-    parse_duration,
-    replace_extension,
-)
 
 
 class FiveMinIE(InfoExtractor):
     IE_NAME = '5min'
-    _VALID_URL = r'(?:5min:(?P<id>\d+)(?::(?P<sid>\d+))?|https?://[^/]*?5min\.com/Scripts/PlayerSeed\.js\?(?P<query>.*))'
+    _VALID_URL = r'(?:5min:|https?://(?:[^/]*?5min\.com/|delivery\.vidible\.tv/aol)(?:(?:Scripts/PlayerSeed\.js|playerseed/?)?\?.*?playList=)?)(?P<id>\d+)'
 
     _TESTS = [
         {
@@ -29,8 +16,16 @@ class FiveMinIE(InfoExtractor):
                 'id': '518013791',
                 'ext': 'mp4',
                 'title': 'iPad Mini with Retina Display Review',
+                'description': 'iPad mini with Retina Display review',
                 'duration': 177,
+                'uploader': 'engadget',
+                'upload_date': '20131115',
+                'timestamp': 1384515288,
             },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            }
         },
         {
             # From http://on.aol.com/video/how-to-make-a-next-level-fruit-salad-518086247
@@ -44,108 +39,16 @@ class FiveMinIE(InfoExtractor):
             },
             'skip': 'no longer available',
         },
+        {
+            'url': 'http://embed.5min.com/518726732/',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://delivery.vidible.tv/aol?playList=518013791',
+            'only_matching': True,
+        }
     ]
-    _ERRORS = {
-        'ErrorVideoNotExist': 'We\'re sorry, but the video you are trying to watch does not exist.',
-        'ErrorVideoNoLongerAvailable': 'We\'re sorry, but the video you are trying to watch is no longer available.',
-        'ErrorVideoRejected': 'We\'re sorry, but the video you are trying to watch has been removed.',
-        'ErrorVideoUserNotGeo': 'We\'re sorry, but the video you are trying to watch cannot be viewed from your current location.',
-        'ErrorVideoLibraryRestriction': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.',
-        'ErrorExposurePermission': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.',
-    }
-    _QUALITIES = {
-        1: {
-            'width': 640,
-            'height': 360,
-        },
-        2: {
-            'width': 854,
-            'height': 480,
-        },
-        4: {
-            'width': 1280,
-            'height': 720,
-        },
-        8: {
-            'width': 1920,
-            'height': 1080,
-        },
-        16: {
-            'width': 640,
-            'height': 360,
-        },
-        32: {
-            'width': 854,
-            'height': 480,
-        },
-        64: {
-            'width': 1280,
-            'height': 720,
-        },
-        128: {
-            'width': 640,
-            'height': 360,
-        },
-    }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        sid = mobj.group('sid')
-
-        if mobj.group('query'):
-            qs = compat_parse_qs(mobj.group('query'))
-            if not qs.get('playList'):
-                raise ExtractorError('Invalid URL', expected=True)
-            video_id = qs['playList'][0]
-            if qs.get('sid'):
-                sid = qs['sid'][0]
-
-        embed_url = 'https://embed.5min.com/playerseed/?playList=%s' % video_id
-        if not sid:
-            embed_page = self._download_webpage(embed_url, video_id,
-                                                'Downloading embed page')
-            sid = self._search_regex(r'sid=(\d+)', embed_page, 'sid')
-
-        response = self._download_json(
-            'https://syn.5min.com/handlers/SenseHandler.ashx?' +
-            compat_urllib_parse_urlencode({
-                'func': 'GetResults',
-                'playlist': video_id,
-                'sid': sid,
-                'isPlayerSeed': 'true',
-                'url': embed_url,
-            }),
-            video_id)
-        if not response['success']:
-            raise ExtractorError(
-                '%s said: %s' % (
-                    self.IE_NAME,
-                    self._ERRORS.get(response['errorMessage'], response['errorMessage'])),
-                expected=True)
-        info = response['binding'][0]
-
-        formats = []
-        parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs(
-            compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0])
-        for rendition in info['Renditions']:
-            if rendition['RenditionType'] == 'aac' or rendition['RenditionType'] == 'm3u8':
-                continue
-            else:
-                rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType'])))
-                quality = self._QUALITIES.get(rendition['ID'], {})
-                formats.append({
-                    'format_id': '%s-%d' % (rendition['RenditionType'], rendition['ID']),
-                    'url': rendition_url,
-                    'width': quality.get('width'),
-                    'height': quality.get('height'),
-                })
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': info['Title'],
-            'thumbnail': info.get('ThumbURL'),
-            'duration': parse_duration(info.get('Duration')),
-            'formats': formats,
-        }
+        video_id = self._match_id(url)
+        return self.url_result('aol-video:%s' % video_id)

youtube_dl/extractor/flipagram.py

@@ -48,7 +48,7 @@ class FlipagramIE(InfoExtractor):
         flipagram = video_data['flipagram']
         video = flipagram['video']
 
-        json_ld = self._search_json_ld(webpage, video_id, default=False)
+        json_ld = self._search_json_ld(webpage, video_id, default={})
         title = json_ld.get('title') or flipagram['captionText']
         description = json_ld.get('description') or flipagram.get('captionText')
 

youtube_dl/extractor/formula1.py

@@ -5,8 +5,8 @@ from .common import InfoExtractor
 
 
 class Formula1IE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?formula1\.com/content/fom-website/en/video/\d{4}/\d{1,2}/(?P<id>.+?)\.html'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?formula1\.com/(?:content/fom-website/)?en/video/\d{4}/\d{1,2}/(?P<id>.+?)\.html'
+    _TESTS = [{
         'url': 'http://www.formula1.com/content/fom-website/en/video/2016/5/Race_highlights_-_Spain_2016.html',
         'md5': '8c79e54be72078b26b89e0e111c0502b',
         'info_dict': {
@@ -15,7 +15,10 @@ class Formula1IE(InfoExtractor):
             'title': 'Race highlights - Spain 2016',
         },
         'add_ie': ['Ooyala'],
-    }
+    }, {
+        'url': 'http://www.formula1.com/en/video/2016/5/Race_highlights_-_Spain_2016.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)

youtube_dl/extractor/fourtube.py

@@ -43,14 +43,14 @@ class FourTubeIE(InfoExtractor):
             'uploadDate', webpage))
         thumbnail = self._html_search_meta('thumbnailUrl', webpage)
         uploader_id = self._html_search_regex(
-            r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">',
+            r'<a class="item-to-subscribe" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">',
             webpage, 'uploader id', fatal=False)
         uploader = self._html_search_regex(
-            r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">',
+            r'<a class="item-to-subscribe" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">',
             webpage, 'uploader', fatal=False)
 
         categories_html = self._search_regex(
-            r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>',
+            r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="[^"]*?list[^"]*?">(.*?)</ul>',
             webpage, 'categories', fatal=False)
         categories = None
         if categories_html:
@@ -59,10 +59,10 @@ class FourTubeIE(InfoExtractor):
                     r'(?s)<li><a.*?>(.*?)</a>', categories_html)]
 
         view_count = str_to_int(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserPlays:([0-9,]+)">',
+            r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([0-9,]+)">',
             webpage, 'view count', fatal=False))
         like_count = str_to_int(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserLikes:([0-9,]+)">',
+            r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserLikes:([0-9,]+)">',
             webpage, 'like count', fatal=False))
         duration = parse_duration(self._html_search_meta('duration', webpage))
 

youtube_dl/extractor/fox.py

@@ -2,7 +2,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import smuggle_url
+from ..utils import (
+    smuggle_url,
+    update_url_query,
+)
 
 
 class FOXIE(InfoExtractor):
@@ -29,11 +32,12 @@ class FOXIE(InfoExtractor):
 
         release_url = self._parse_json(self._search_regex(
             r'"fox_pdk_player"\s*:\s*({[^}]+?})', webpage, 'fox_pdk_player'),
-            video_id)['release_url'] + '&switch=http'
+            video_id)['release_url']
 
         return {
             '_type': 'url_transparent',
             'ie_key': 'ThePlatform',
-            'url': smuggle_url(release_url, {'force_smil_url': True}),
+            'url': smuggle_url(update_url_query(
+                release_url, {'switch': 'http'}), {'force_smil_url': True}),
             'id': video_id,
         }

youtube_dl/extractor/franceculture.py

@@ -2,104 +2,56 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urlparse,
-)
 from ..utils import (
     determine_ext,
-    int_or_none,
-    ExtractorError,
+    unified_strdate,
 )
 
 
 class FranceCultureIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/player/reecouter\?play=(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/emissions/(?:[^/]+/)*(?P<id>[^/?#&]+)'
     _TEST = {
-        'url': 'http://www.franceculture.fr/player/reecouter?play=4795174',
+        'url': 'http://www.franceculture.fr/emissions/carnet-nomade/rendez-vous-au-pays-des-geeks',
         'info_dict': {
-            'id': '4795174',
+            'id': 'rendez-vous-au-pays-des-geeks',
+            'display_id': 'rendez-vous-au-pays-des-geeks',
             'ext': 'mp3',
             'title': 'Rendez-vous au pays des geeks',
-            'alt_title': 'Carnet nomade | 13-14',
-            'vcodec': 'none',
+            'thumbnail': 're:^https?://.*\\.jpg$',
             'upload_date': '20140301',
-            'thumbnail': r're:^http://static\.franceculture\.fr/.*/images/player/Carnet-nomade\.jpg$',
-            'description': 'startswith:Avec :Jean-Baptiste Péretié pour son documentaire sur Arte "La revanche',
-            'timestamp': 1393700400,
+            'vcodec': 'none',
         }
     }
 
-    def _extract_from_player(self, url, video_id):
-        webpage = self._download_webpage(url, video_id)
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
 
-        video_path = self._search_regex(
-            r'<a id="player".*?href="([^"]+)"', webpage, 'video path')
-        video_url = compat_urlparse.urljoin(url, video_path)
-        timestamp = int_or_none(self._search_regex(
-            r'<a id="player".*?data-date="([0-9]+)"',
+        webpage = self._download_webpage(url, display_id)
+
+        video_url = self._search_regex(
+            r'(?s)<div[^>]+class="[^"]*?title-zone-diffusion[^"]*?"[^>]*>.*?<a[^>]+href="([^"]+)"',
+            webpage, 'video path')
+
+        title = self._og_search_title(webpage)
+
+        upload_date = unified_strdate(self._search_regex(
+            '(?s)<div[^>]+class="date"[^>]*>.*?<span[^>]+class="inner"[^>]*>([^<]+)<',
             webpage, 'upload date', fatal=False))
         thumbnail = self._search_regex(
-            r'<a id="player".*?>\s+<img src="([^"]+)"',
+            r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+data-pagespeed-(?:lazy|high-res)-src="([^"]+)"',
             webpage, 'thumbnail', fatal=False)
-
-        display_id = self._search_regex(
-            r'<span class="path-diffusion">emission-(.*?)</span>', webpage, 'display_id')
-
-        title = self._html_search_regex(
-            r'<span class="title-diffusion">(.*?)</span>', webpage, 'title')
-        alt_title = self._html_search_regex(
-            r'<span class="title">(.*?)</span>',
-            webpage, 'alt_title', fatal=False)
-        description = self._html_search_regex(
-            r'<span class="description">(.*?)</span>',
-            webpage, 'description', fatal=False)
-
         uploader = self._html_search_regex(
             r'(?s)<div id="emission".*?<span class="author">(.*?)</span>',
             webpage, 'uploader', default=None)
         vcodec = 'none' if determine_ext(video_url.lower()) == 'mp3' else None
 
         return {
-            'id': video_id,
+            'id': display_id,
+            'display_id': display_id,
             'url': video_url,
+            'title': title,
+            'thumbnail': thumbnail,
             'vcodec': vcodec,
             'uploader': uploader,
-            'timestamp': timestamp,
-            'title': title,
-            'alt_title': alt_title,
-            'thumbnail': thumbnail,
-            'description': description,
-            'display_id': display_id,
+            'upload_date': upload_date,
         }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        return self._extract_from_player(url, video_id)
-
-
-class FranceCultureEmissionIE(FranceCultureIE):
-    _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/emission-(?P<id>[^?#]+)'
-    _TEST = {
-        'url': 'http://www.franceculture.fr/emission-les-carnets-de-la-creation-jean-gabriel-periot-cineaste-2015-10-13',
-        'info_dict': {
-            'title': 'Jean-Gabriel Périot, cinéaste',
-            'alt_title': 'Les Carnets de la création',
-            'id': '5093239',
-            'display_id': 'les-carnets-de-la-creation-jean-gabriel-periot-cineaste-2015-10-13',
-            'ext': 'mp3',
-            'timestamp': 1444762500,
-            'upload_date': '20151013',
-            'description': 'startswith:Aujourd\'hui dans "Les carnets de la création", le cinéaste',
-        },
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        video_path = self._html_search_regex(
-            r'<a class="rf-player-open".*?href="([^"]+)"', webpage, 'video path', 'no_path_player')
-        if video_path == 'no_path_player':
-            raise ExtractorError('no player : no sound in this page.', expected=True)
-        new_id = self._search_regex('play=(?P<id>[0-9]+)', video_path, 'new_id', group='id')
-        video_url = compat_urlparse.urljoin(url, video_path)
-        return self._extract_from_player(video_url, new_id)

youtube_dl/extractor/francetv.py

@@ -131,7 +131,7 @@ class PluzzIE(FranceTVBaseInfoExtractor):
 
 class FranceTvInfoIE(FranceTVBaseInfoExtractor):
     IE_NAME = 'francetvinfo.fr'
-    _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/.*/(?P<title>.+)\.html'
+    _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<title>[^/?#&.]+)'
 
     _TESTS = [{
         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
@@ -206,6 +206,9 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
             'uploader_id': 'x2q2ez',
         },
         'add_ie': ['Dailymotion'],
+    }, {
+        'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/gamekings.py

@@ -1,76 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from ..utils import (
-    xpath_text,
-    xpath_with_ns,
-)
-from .youtube import YoutubeIE
-
-
-class GamekingsIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.gamekings\.nl/(?:videos|nieuws)/(?P<id>[^/]+)'
-    _TESTS = [{
-        # YouTube embed video
-        'url': 'http://www.gamekings.nl/videos/phoenix-wright-ace-attorney-dual-destinies-review/',
-        'md5': '5208d3a17adeaef829a7861887cb9029',
-        'info_dict': {
-            'id': 'HkSQKetlGOU',
-            'ext': 'mp4',
-            'title': 'Phoenix Wright: Ace Attorney - Dual Destinies Review',
-            'description': 'md5:db88c0e7f47e9ea50df3271b9dc72e1d',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'uploader_id': 'UCJugRGo4STYMeFr5RoOShtQ',
-            'uploader': 'Gamekings Vault',
-            'upload_date': '20151123',
-        },
-        'add_ie': ['Youtube'],
-    }, {
-        # vimeo video
-        'url': 'http://www.gamekings.nl/videos/the-legend-of-zelda-majoras-mask/',
-        'md5': '12bf04dfd238e70058046937657ea68d',
-        'info_dict': {
-            'id': 'the-legend-of-zelda-majoras-mask',
-            'ext': 'mp4',
-            'title': 'The Legend of Zelda: Majora’s Mask',
-            'description': 'md5:9917825fe0e9f4057601fe1e38860de3',
-            'thumbnail': 're:^https?://.*\.jpg$',
-        },
-    }, {
-        'url': 'http://www.gamekings.nl/nieuws/gamekings-extra-shelly-en-david-bereiden-zich-voor-op-de-livestream/',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-
-        playlist_id = self._search_regex(
-            r'gogoVideo\([^,]+,\s*"([^"]+)', webpage, 'playlist id')
-
-        # Check if a YouTube embed is used
-        if YoutubeIE.suitable(playlist_id):
-            return self.url_result(playlist_id, ie='Youtube')
-
-        playlist = self._download_xml(
-            'http://www.gamekings.tv/wp-content/themes/gk2010/rss_playlist.php?id=%s' % playlist_id,
-            video_id)
-
-        NS_MAP = {
-            'jwplayer': 'http://rss.jwpcdn.com/'
-        }
-
-        item = playlist.find('./channel/item')
-
-        thumbnail = xpath_text(item, xpath_with_ns('./jwplayer:image', NS_MAP), 'thumbnail')
-        video_url = item.find(xpath_with_ns('./jwplayer:source', NS_MAP)).get('file')
-
-        return {
-            'id': video_id,
-            'url': video_url,
-            'title': self._og_search_title(webpage),
-            'description': self._og_search_description(webpage),
-            'thumbnail': thumbnail,
-        }

youtube_dl/extractor/generic.py

@@ -2207,6 +2207,14 @@ class GenericIE(InfoExtractor):
             return self.url_result(
                 self._proto_relative_url(unescapeHTML(mobj.group(1))), 'Vine')
 
+        # Look for VODPlatform embeds
+        mobj = re.search(
+            r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vod-platform\.net/embed/[^/?#]+)',
+            webpage)
+        if mobj is not None:
+            return self.url_result(
+                self._proto_relative_url(unescapeHTML(mobj.group(1))), 'VODPlatform')
+
         # Look for Instagram embeds
         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
         if instagram_embed_url is not None:
@@ -2233,8 +2241,8 @@ class GenericIE(InfoExtractor):
 
         # Looking for http://schema.org/VideoObject
         json_ld = self._search_json_ld(
-            webpage, video_id, default=None, expected_type='VideoObject')
-        if json_ld and json_ld.get('url'):
+            webpage, video_id, default={}, expected_type='VideoObject')
+        if json_ld.get('url'):
             info_dict.update({
                 'title': video_title or info_dict['title'],
                 'description': video_description,

youtube_dl/extractor/goldenmoustache.py

@@ -1,48 +0,0 @@
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-
-
-class GoldenMoustacheIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?goldenmoustache\.com/(?P<display_id>[\w-]+)-(?P<id>\d+)'
-    _TESTS = [{
-        'url': 'http://www.goldenmoustache.com/suricate-le-poker-3700/',
-        'md5': '0f904432fa07da5054d6c8beb5efb51a',
-        'info_dict': {
-            'id': '3700',
-            'ext': 'mp4',
-            'title': 'Suricate - Le Poker',
-            'description': 'md5:3d1f242f44f8c8cb0a106f1fd08e5dc9',
-            'thumbnail': 're:^https?://.*\.jpg$',
-        }
-    }, {
-        'url': 'http://www.goldenmoustache.com/le-lab-tout-effacer-mc-fly-et-carlito-55249/',
-        'md5': '27f0c50fb4dd5f01dc9082fc67cd5700',
-        'info_dict': {
-            'id': '55249',
-            'ext': 'mp4',
-            'title': 'Le LAB - Tout Effacer (Mc Fly et Carlito)',
-            'description': 'md5:9b7fbf11023fb2250bd4b185e3de3b2a',
-            'thumbnail': 're:^https?://.*\.(?:png|jpg)$',
-        }
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        video_url = self._html_search_regex(
-            r'data-src-type="mp4" data-src="([^"]+)"', webpage, 'video URL')
-        title = self._html_search_regex(
-            r'<title>(.*?)(?: - Golden Moustache)?</title>', webpage, 'title')
-        thumbnail = self._og_search_thumbnail(webpage)
-        description = self._og_search_description(webpage)
-
-        return {
-            'id': video_id,
-            'url': video_url,
-            'ext': 'mp4',
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-        }

youtube_dl/extractor/hgtv.py

@@ -0,0 +1,48 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    smuggle_url,
+)
+
+
+class HGTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?hgtv\.ca/[^/]+/video/(?P<id>[^/]+)/video.html'
+    _TEST = {
+        'url': 'http://www.hgtv.ca/homefree/video/overnight-success/video.html?v=738081859718&p=1&s=da#video',
+        'md5': '',
+        'info_dict': {
+            'id': 'aFH__I_5FBOX',
+            'ext': 'mp4',
+            'title': 'Overnight Success',
+            'description': 'After weeks of hard work, high stakes, breakdowns and pep talks, the final 2 contestants compete to win the ultimate dream.',
+            'uploader': 'SHWM-NEW',
+            'timestamp': 1470320034,
+            'upload_date': '20160804',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        embed_vars = self._parse_json(self._search_regex(
+            r'(?s)embed_vars\s*=\s*({.*?});',
+            webpage, 'embed vars'), display_id, js_to_json)
+        return {
+            '_type': 'url_transparent',
+            'url': smuggle_url(
+                'http://link.theplatform.com/s/dtjsEC/%s?mbr=true&manifest=m3u' % embed_vars['pid'], {
+                    'force_smil_url': True
+                }),
+            'series': embed_vars.get('show'),
+            'season_number': int_or_none(embed_vars.get('season')),
+            'episode_number': int_or_none(embed_vars.get('episode')),
+            'ie_key': 'ThePlatform',
+        }

youtube_dl/extractor/imgur.py

@@ -50,12 +50,10 @@ class ImgurIE(InfoExtractor):
         webpage = self._download_webpage(
             compat_urlparse.urljoin(url, video_id), video_id)
 
-        width = int_or_none(self._search_regex(
-            r'<param name="width" value="([0-9]+)"',
-            webpage, 'width', fatal=False))
-        height = int_or_none(self._search_regex(
-            r'<param name="height" value="([0-9]+)"',
-            webpage, 'height', fatal=False))
+        width = int_or_none(self._og_search_property(
+            'video:width', webpage, default=None))
+        height = int_or_none(self._og_search_property(
+            'video:height', webpage, default=None))
 
         video_elements = self._search_regex(
             r'(?s)<div class="video-elements">(.*?)</div>',

youtube_dl/extractor/jwplatform.py

@@ -4,10 +4,12 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     determine_ext,
     float_or_none,
     int_or_none,
+    mimetype2ext,
 )
 
 
@@ -28,14 +30,14 @@ class JWPlatformBaseIE(InfoExtractor):
         return self._parse_jwplayer_data(
             jwplayer_data, video_id, *args, **kwargs)
 
-    def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True, m3u8_id=None, rtmp_params=None):
+    def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True, m3u8_id=None, rtmp_params=None, base_url=None):
         # JWPlayer backward compatibility: flattened playlists
         # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/api/config.js#L81-L96
         if 'playlist' not in jwplayer_data:
             jwplayer_data = {'playlist': [jwplayer_data]}
 
-        video_data = jwplayer_data['playlist'][0]
-
+        entries = []
+        for video_data in jwplayer_data['playlist']:
             # JWPlayer backward compatibility: flattened sources
             # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35
             if 'sources' not in video_data:
@@ -44,20 +46,26 @@ class JWPlatformBaseIE(InfoExtractor):
             formats = []
             for source in video_data['sources']:
                 source_url = self._proto_relative_url(source['file'])
+                if base_url:
+                    source_url = compat_urlparse.urljoin(base_url, source_url)
                 source_type = source.get('type') or ''
-            if source_type in ('application/vnd.apple.mpegurl', 'hls') or determine_ext(source_url) == 'm3u8':
+                ext = mimetype2ext(source_type) or determine_ext(source_url)
+                if source_type == 'hls' or ext == 'm3u8':
                     formats.extend(self._extract_m3u8_formats(
                         source_url, video_id, 'mp4', 'm3u8_native', m3u8_id=m3u8_id, fatal=False))
-            elif source_type.startswith('audio'):
+                # https://github.com/jwplayer/jwplayer/blob/master/src/js/providers/default.js#L67
+                elif source_type.startswith('audio') or ext in ('oga', 'aac', 'mp3', 'mpeg', 'vorbis'):
                     formats.append({
                         'url': source_url,
                         'vcodec': 'none',
+                        'ext': ext,
                     })
                 else:
                     a_format = {
                         'url': source_url,
                         'width': int_or_none(source.get('width')),
                         'height': int_or_none(source.get('height')),
+                        'ext': ext,
                     }
                     if source_url.startswith('rtmp'):
                         a_format['ext'] = 'flv',
@@ -86,7 +94,7 @@ class JWPlatformBaseIE(InfoExtractor):
                             'url': self._proto_relative_url(track['file'])
                         })
 
-        return {
+            entries.append({
                 'id': video_id,
                 'title': video_data['title'] if require_title else video_data.get('title'),
                 'description': video_data.get('description'),
@@ -95,7 +103,11 @@ class JWPlatformBaseIE(InfoExtractor):
                 'duration': float_or_none(jwplayer_data.get('duration')),
                 'subtitles': subtitles,
                 'formats': formats,
-        }
+            })
+        if len(entries) == 1:
+            return entries[0]
+        else:
+            return self.playlist_result(entries)
 
 
 class JWPlatformIE(JWPlatformBaseIE):

youtube_dl/extractor/kaltura.py

@@ -62,6 +62,11 @@ class KalturaIE(InfoExtractor):
         {
             'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.30.2/mwEmbedFrame.php/p/1337/uiconf_id/20540612/entry_id/1_sf5ovm7u?wid=_243342',
             'only_matching': True,
+        },
+        {
+            # video with subtitles
+            'url': 'kaltura:111032:1_cw786r8q',
+            'only_matching': True,
         }
     ]
 
@@ -130,7 +135,6 @@ class KalturaIE(InfoExtractor):
             video_id, actions, service_url, note='Downloading Kaltura signature')['ks']
 
     def _get_video_info(self, video_id, partner_id, service_url=None):
-        signature = self._get_kaltura_signature(video_id, partner_id, service_url)
         actions = [
             {
                 'action': 'null',
@@ -138,18 +142,30 @@ class KalturaIE(InfoExtractor):
                 'clientTag': 'kdp:v3.8.5',
                 'format': 1,  # JSON, 2 = XML, 3 = PHP
                 'service': 'multirequest',
-                'ks': signature,
+            },
+            {
+                'expiry': 86400,
+                'service': 'session',
+                'action': 'startWidgetSession',
+                'widgetId': '_%s' % partner_id,
             },
             {
                 'action': 'get',
                 'entryId': video_id,
                 'service': 'baseentry',
-                'version': '-1',
+                'ks': '{1:result:ks}',
             },
             {
                 'action': 'getbyentryid',
                 'entryId': video_id,
                 'service': 'flavorAsset',
+                'ks': '{1:result:ks}',
+            },
+            {
+                'action': 'list',
+                'filter:entryIdEqual': video_id,
+                'service': 'caption_captionasset',
+                'ks': '{1:result:ks}',
             },
         ]
         return self._kaltura_api_call(
@@ -161,8 +177,9 @@ class KalturaIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         partner_id, entry_id = mobj.group('partner_id', 'id')
         ks = None
+        captions = None
         if partner_id and entry_id:
-            info, flavor_assets = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'))
+            _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'))
         else:
             path, query = mobj.group('path', 'query')
             if not path and not query:
@@ -181,7 +198,7 @@ class KalturaIE(InfoExtractor):
                 raise ExtractorError('Invalid URL', expected=True)
             if 'entry_id' in params:
                 entry_id = params['entry_id'][0]
-                info, flavor_assets = self._get_video_info(entry_id, partner_id)
+                _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id)
             elif 'uiconf_id' in params and 'flashvars[referenceId]' in params:
                 reference_id = params['flashvars[referenceId]'][0]
                 webpage = self._download_webpage(url, reference_id)
@@ -217,7 +234,7 @@ class KalturaIE(InfoExtractor):
         formats = []
         for f in flavor_assets:
             # Continue if asset is not ready
-            if f['status'] != 2:
+            if f.get('status') != 2:
                 continue
             video_url = sign_url(
                 '%s/flavorId/%s' % (data_url, f['id']))
@@ -240,13 +257,24 @@ class KalturaIE(InfoExtractor):
                 m3u8_url, entry_id, 'mp4', 'm3u8_native',
                 m3u8_id='hls', fatal=False))
 
-        self._check_formats(formats, entry_id)
         self._sort_formats(formats)
 
+        subtitles = {}
+        if captions:
+            for caption in captions.get('objects', []):
+                # Continue if caption is not ready
+                if f.get('status') != 2:
+                    continue
+                subtitles.setdefault(caption.get('languageCode') or caption.get('language'), []).append({
+                    'url': '%s/api_v3/service/caption_captionasset/action/serve/captionAssetId/%s' % (self._SERVICE_URL, caption['id']),
+                    'ext': caption.get('fileExt'),
+                })
+
         return {
             'id': entry_id,
             'title': info['name'],
             'formats': formats,
+            'subtitles': subtitles,
             'description': clean_html(info.get('description')),
             'thumbnail': info.get('thumbnailUrl'),
             'duration': info.get('duration'),

youtube_dl/extractor/kuwo.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     get_element_by_id,
     clean_html,
@@ -242,8 +243,9 @@ class KuwoSingerIE(InfoExtractor):
                 query={'artistId': artist_id, 'pn': page_num, 'rn': self.PAGE_SIZE})
 
             return [
-                self.url_result(song_url, 'Kuwo') for song_url in re.findall(
-                    r'<div[^>]+class="name"><a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+)',
+                self.url_result(compat_urlparse.urljoin(url, song_url), 'Kuwo')
+                for song_url in re.findall(
+                    r'<div[^>]+class="name"><a[^>]+href="(/yinyue/\d+)',
                     webpage)
             ]
 

youtube_dl/extractor/limelight.py

@@ -37,11 +37,12 @@ class LimelightBaseIE(InfoExtractor):
 
         for stream in streams:
             stream_url = stream.get('url')
-            if not stream_url:
+            if not stream_url or stream.get('drmProtected'):
                 continue
-            if '.f4m' in stream_url:
+            ext = determine_ext(stream_url)
+            if ext == 'f4m':
                 formats.extend(self._extract_f4m_formats(
-                    stream_url, video_id, fatal=False))
+                    stream_url, video_id, f4m_id='hds', fatal=False))
             else:
                 fmt = {
                     'url': stream_url,
@@ -50,13 +51,19 @@ class LimelightBaseIE(InfoExtractor):
                     'fps': float_or_none(stream.get('videoFrameRate')),
                     'width': int_or_none(stream.get('videoWidthInPixels')),
                     'height': int_or_none(stream.get('videoHeightInPixels')),
-                    'ext': determine_ext(stream_url)
+                    'ext': ext,
                 }
-                rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
+                rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
                 if rtmp:
                     format_id = 'rtmp'
                     if stream.get('videoBitRate'):
                         format_id += '-%d' % int_or_none(stream['videoBitRate'])
+                    http_fmt = fmt.copy()
+                    http_fmt.update({
+                        'url': 'http://%s/%s' % (rtmp.group('host').replace('csl.', 'cpl.'), rtmp.group('playpath')[4:]),
+                        'format_id': format_id.replace('rtmp', 'http'),
+                    })
+                    formats.append(http_fmt)
                     fmt.update({
                         'url': rtmp.group('url'),
                         'play_path': rtmp.group('playpath'),
@@ -68,18 +75,23 @@ class LimelightBaseIE(InfoExtractor):
 
         for mobile_url in mobile_urls:
             media_url = mobile_url.get('mobileUrl')
-            if not media_url:
-                continue
             format_id = mobile_url.get('targetMediaPlatform')
-            if determine_ext(media_url) == 'm3u8':
+            if not media_url or format_id == 'Widevine':
+                continue
+            ext = determine_ext(media_url)
+            if ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
                     media_url, video_id, 'mp4', 'm3u8_native',
                     m3u8_id=format_id, fatal=False))
+            elif ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(
+                    stream_url, video_id, f4m_id=format_id, fatal=False))
             else:
                 formats.append({
                     'url': media_url,
                     'format_id': format_id,
                     'preference': -1,
+                    'ext': ext,
                 })
 
         self._sort_formats(formats)
@@ -145,7 +157,7 @@ class LimelightMediaIE(LimelightBaseIE):
         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
         'info_dict': {
             'id': '3ffd040b522b4485b6d84effc750cd86',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'HaP and the HB Prince Trailer',
             'description': 'md5:8005b944181778e313d95c1237ddb640',
             'thumbnail': 're:^https?://.*\.jpeg$',
@@ -154,27 +166,23 @@ class LimelightMediaIE(LimelightBaseIE):
             'upload_date': '20090604',
         },
         'params': {
-            # rtmp download
+            # m3u8 download
             'skip_download': True,
         },
     }, {
         # video with subtitles
         'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
+        'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
         'info_dict': {
             'id': 'a3e00274d4564ec4a9b29b9466432335',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': '3Play Media Overview Video',
-            'description': '',
             'thumbnail': 're:^https?://.*\.jpeg$',
             'duration': 78.101,
             'timestamp': 1338929955,
             'upload_date': '20120605',
             'subtitles': 'mincount:9',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
     }, {
         'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
         'only_matching': True,

youtube_dl/extractor/muenchentv.py

@@ -36,7 +36,7 @@ class MuenchenTVIE(InfoExtractor):
         title = self._live_title(self._og_search_title(webpage))
 
         data_js = self._search_regex(
-            r'(?s)\nplaylist:\s*(\[.*?}\]),related:',
+            r'(?s)\nplaylist:\s*(\[.*?}\]),',
             webpage, 'playlist configuration')
         data_json = js_to_json(data_js)
         data = json.loads(data_json)[0]

youtube_dl/extractor/nationalgeographic.py

@@ -1,16 +1,19 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from .theplatform import ThePlatformIE
 from ..utils import (
     smuggle_url,
     url_basename,
     update_url_query,
+    get_element_by_class,
 )
 
 
-class NationalGeographicIE(InfoExtractor):
-    IE_NAME = 'natgeo'
+class NationalGeographicVideoIE(InfoExtractor):
+    IE_NAME = 'natgeo:video'
     _VALID_URL = r'https?://video\.nationalgeographic\.com/.*?'
 
     _TESTS = [
@@ -62,16 +65,16 @@ class NationalGeographicIE(InfoExtractor):
         }
 
 
-class NationalGeographicChannelIE(ThePlatformIE):
-    IE_NAME = 'natgeo:channel'
-    _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:wild/)?[^/]+/videos/(?P<id>[^/?]+)'
+class NationalGeographicIE(ThePlatformIE):
+    IE_NAME = 'natgeo'
+    _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:wild/)?[^/]+/(?:videos|episodes)/(?P<id>[^/?]+)'
 
     _TESTS = [
         {
             'url': 'http://channel.nationalgeographic.com/the-story-of-god-with-morgan-freeman/videos/uncovering-a-universal-knowledge/',
             'md5': '518c9aa655686cf81493af5cc21e2a04',
             'info_dict': {
-                'id': 'nB5vIAfmyllm',
+                'id': 'vKInpacll2pC',
                 'ext': 'mp4',
                 'title': 'Uncovering a Universal Knowledge',
                 'description': 'md5:1a89148475bf931b3661fcd6ddb2ae3a',
@@ -85,7 +88,7 @@ class NationalGeographicChannelIE(ThePlatformIE):
             'url': 'http://channel.nationalgeographic.com/wild/destination-wild/videos/the-stunning-red-bird-of-paradise/',
             'md5': 'c4912f656b4cbe58f3e000c489360989',
             'info_dict': {
-                'id': '3TmMv9OvGwIR',
+                'id': 'Pok5lWCkiEFA',
                 'ext': 'mp4',
                 'title': 'The Stunning Red Bird of Paradise',
                 'description': 'md5:7bc8cd1da29686be4d17ad1230f0140c',
@@ -95,6 +98,10 @@ class NationalGeographicChannelIE(ThePlatformIE):
             },
             'add_ie': ['ThePlatform'],
         },
+        {
+            'url': 'http://channel.nationalgeographic.com/the-story-of-god-with-morgan-freeman/episodes/the-power-of-miracles/',
+            'only_matching': True,
+        }
     ]
 
     def _real_extract(self, url):
@@ -122,3 +129,40 @@ class NationalGeographicChannelIE(ThePlatformIE):
                 {'force_smil_url': True}),
             'display_id': display_id,
         }
+
+
+class NationalGeographicEpisodeGuideIE(ThePlatformIE):
+    IE_NAME = 'natgeo:episodeguide'
+    _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:wild/)?(?P<id>[^/]+)/episode-guide'
+    _TESTS = [
+        {
+            'url': 'http://channel.nationalgeographic.com/the-story-of-god-with-morgan-freeman/episode-guide/',
+            'info_dict': {
+                'id': 'the-story-of-god-with-morgan-freeman-season-1',
+                'title': 'The Story of God with Morgan Freeman - Season 1',
+            },
+            'playlist_mincount': 6,
+        },
+        {
+            'url': 'http://channel.nationalgeographic.com/underworld-inc/episode-guide/?s=2',
+            'info_dict': {
+                'id': 'underworld-inc-season-2',
+                'title': 'Underworld, Inc. - Season 2',
+            },
+            'playlist_mincount': 7,
+        },
+    ]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        show = get_element_by_class('show', webpage)
+        selected_season = self._search_regex(
+            r'<div[^>]+class="select-seasons[^"]*".*?<a[^>]*>(.*?)</a>',
+            webpage, 'selected season')
+        entries = [
+            self.url_result(self._proto_relative_url(entry_url), 'NationalGeographic')
+            for entry_url in re.findall('(?s)<div[^>]+class="col-inner"[^>]*?>.*?<a[^>]+href="([^"]+)"', webpage)]
+        return self.playlist_result(
+            entries, '%s-%s' % (display_id, selected_season.lower().replace(' ', '-')),
+            '%s - %s' % (show, selected_season))

youtube_dl/extractor/naver.py

@@ -4,12 +4,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse_urlencode,
-    compat_urlparse,
-)
 from ..utils import (
     ExtractorError,
+    int_or_none,
+    update_url_query,
 )
 
 
@@ -51,48 +49,74 @@ class NaverIE(InfoExtractor):
             if error:
                 raise ExtractorError(error, expected=True)
             raise ExtractorError('couldn\'t extract vid and key')
-        vid = m_id.group(1)
-        key = m_id.group(2)
-        query = compat_urllib_parse_urlencode({'vid': vid, 'inKey': key, })
-        query_urls = compat_urllib_parse_urlencode({
-            'masterVid': vid,
-            'protocol': 'p2p',
-            'inKey': key,
+        video_data = self._download_json(
+            'http://play.rmcnmv.naver.com/vod/play/v2.0/' + m_id.group(1),
+            video_id, query={
+                'key': m_id.group(2),
             })
-        info = self._download_xml(
-            'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
-            video_id, 'Downloading video info')
-        urls = self._download_xml(
-            'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
-            video_id, 'Downloading video formats info')
-
+        meta = video_data['meta']
+        title = meta['subject']
         formats = []
-        for format_el in urls.findall('EncodingOptions/EncodingOption'):
-            domain = format_el.find('Domain').text
-            uri = format_el.find('uri').text
-            f = {
-                'url': compat_urlparse.urljoin(domain, uri),
-                'ext': 'mp4',
-                'width': int(format_el.find('width').text),
-                'height': int(format_el.find('height').text),
-            }
-            if domain.startswith('rtmp'):
-                # urlparse does not support custom schemes
-                # https://bugs.python.org/issue18828
-                f.update({
-                    'url': domain + uri,
-                    'ext': 'flv',
-                    'rtmp_protocol': '1',  # rtmpt
+
+        def extract_formats(streams, stream_type, query={}):
+            for stream in streams:
+                stream_url = stream.get('source')
+                if not stream_url:
+                    continue
+                stream_url = update_url_query(stream_url, query)
+                encoding_option = stream.get('encodingOption', {})
+                bitrate = stream.get('bitrate', {})
+                formats.append({
+                    'format_id': '%s_%s' % (stream.get('type') or stream_type, encoding_option.get('id') or encoding_option.get('name')),
+                    'url': stream_url,
+                    'width': int_or_none(encoding_option.get('width')),
+                    'height': int_or_none(encoding_option.get('height')),
+                    'vbr': int_or_none(bitrate.get('video')),
+                    'abr': int_or_none(bitrate.get('audio')),
+                    'filesize': int_or_none(stream.get('size')),
+                    'protocol': 'm3u8_native' if stream_type == 'HLS' else None,
                 })
-            formats.append(f)
+
+        extract_formats(video_data.get('videos', {}).get('list', []), 'H264')
+        for stream_set in video_data.get('streams', []):
+            query = {}
+            for param in stream_set.get('keys', []):
+                query[param['name']] = param['value']
+            stream_type = stream_set.get('type')
+            videos = stream_set.get('videos')
+            if videos:
+                extract_formats(videos, stream_type, query)
+            elif stream_type == 'HLS':
+                stream_url = stream_set.get('source')
+                if not stream_url:
+                    continue
+                formats.extend(self._extract_m3u8_formats(
+                    update_url_query(stream_url, query), video_id,
+                    'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False))
         self._sort_formats(formats)
 
+        subtitles = {}
+        for caption in video_data.get('captions', {}).get('list', []):
+            caption_url = caption.get('source')
+            if not caption_url:
+                continue
+            subtitles.setdefault(caption.get('language') or caption.get('locale'), []).append({
+                'url': caption_url,
+            })
+
+        upload_date = self._search_regex(
+            r'<span[^>]+class="date".*?(\d{4}\.\d{2}\.\d{2})',
+            webpage, 'upload date', fatal=False)
+        if upload_date:
+            upload_date = upload_date.replace('.', '')
+
         return {
             'id': video_id,
-            'title': info.find('Subject').text,
+            'title': title,
             'formats': formats,
+            'subtitles': subtitles,
             'description': self._og_search_description(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'upload_date': info.find('WriteDate').text.replace('.', ''),
-            'view_count': int(info.find('PlayCount').text),
+            'thumbnail': meta.get('cover', {}).get('source') or self._og_search_thumbnail(webpage),
+            'view_count': int_or_none(meta.get('count')),
+            'upload_date': upload_date,
         }

youtube_dl/extractor/ntvru.py

@@ -11,10 +11,9 @@ from ..utils import (
 
 class NTVRuIE(InfoExtractor):
     IE_NAME = 'ntv.ru'
-    _VALID_URL = r'https?://(?:www\.)?ntv\.ru/(?P<id>.+)'
+    _VALID_URL = r'https?://(?:www\.)?ntv\.ru/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
-    _TESTS = [
-        {
+    _TESTS = [{
         'url': 'http://www.ntv.ru/novosti/863142/',
         'md5': 'ba7ea172a91cb83eb734cad18c10e723',
         'info_dict': {
@@ -25,8 +24,7 @@ class NTVRuIE(InfoExtractor):
             'thumbnail': 're:^http://.*\.jpg',
             'duration': 136,
         },
-        },
-        {
+    }, {
         'url': 'http://www.ntv.ru/video/novosti/750370/',
         'md5': 'adecff79691b4d71e25220a191477124',
         'info_dict': {
@@ -37,8 +35,7 @@ class NTVRuIE(InfoExtractor):
             'thumbnail': 're:^http://.*\.jpg',
             'duration': 172,
         },
-        },
-        {
+    }, {
         'url': 'http://www.ntv.ru/peredacha/segodnya/m23700/o232416',
         'md5': '82dbd49b38e3af1d00df16acbeab260c',
         'info_dict': {
@@ -49,8 +46,7 @@ class NTVRuIE(InfoExtractor):
             'thumbnail': 're:^http://.*\.jpg',
             'duration': 1496,
         },
-        },
-        {
+    }, {
         'url': 'http://www.ntv.ru/kino/Koma_film',
         'md5': 'f825770930937aa7e5aca0dc0d29319a',
         'info_dict': {
@@ -61,8 +57,7 @@ class NTVRuIE(InfoExtractor):
             'thumbnail': 're:^http://.*\.jpg',
             'duration': 5592,
         },
-        },
-        {
+    }, {
         'url': 'http://www.ntv.ru/serial/Delo_vrachey/m31760/o233916/',
         'md5': '9320cd0e23f3ea59c330dc744e06ff3b',
         'info_dict': {
@@ -73,8 +68,7 @@ class NTVRuIE(InfoExtractor):
             'thumbnail': 're:^http://.*\.jpg',
             'duration': 2590,
         },
-        },
-    ]
+    }]
 
     _VIDEO_ID_REGEXES = [
         r'<meta property="og:url" content="http://www\.ntv\.ru/video/(\d+)',
@@ -87,11 +81,21 @@ class NTVRuIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        video_id = self._html_search_regex(self._VIDEO_ID_REGEXES, webpage, 'video id')
+        video_url = self._og_search_property(
+            ('video', 'video:iframe'), webpage, default=None)
+        if video_url:
+            video_id = self._search_regex(
+                r'https?://(?:www\.)?ntv\.ru/video/(?:embed/)?(\d+)',
+                video_url, 'video id', default=None)
+
+        if not video_id:
+            video_id = self._html_search_regex(
+                self._VIDEO_ID_REGEXES, webpage, 'video id')
 
         player = self._download_xml(
             'http://www.ntv.ru/vi%s/' % video_id,
             video_id, 'Downloading video XML')
+
         title = clean_html(xpath_text(player, './data/title', 'title', fatal=True))
         description = clean_html(xpath_text(player, './data/description', 'description'))
 

youtube_dl/extractor/openload.py

@@ -1,15 +1,14 @@
 # coding: utf-8
-from __future__ import unicode_literals
+from __future__ import unicode_literals, division
 
-import re
+import math
 
 from .common import InfoExtractor
 from ..compat import compat_chr
 from ..utils import (
+    decode_png,
     determine_ext,
-    encode_base_n,
     ExtractorError,
-    mimetype2ext,
 )
 
 
@@ -41,60 +40,6 @@ class OpenloadIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    @staticmethod
-    def openload_level2_debase(m):
-        radix, num = int(m.group(1)) + 27, int(m.group(2))
-        return '"' + encode_base_n(num, radix) + '"'
-
-    @classmethod
-    def openload_level2(cls, txt):
-        # The function name is ǃ \u01c3
-        # Using escaped unicode literals does not work in Python 3.2
-        return re.sub(r'ǃ\((\d+),(\d+)\)', cls.openload_level2_debase, txt, re.UNICODE).replace('"+"', '')
-
-    # Openload uses a variant of aadecode
-    # openload_decode and related functions are originally written by
-    # vitas@matfyz.cz and released with public domain
-    # See https://github.com/rg3/youtube-dl/issues/8489
-    @classmethod
-    def openload_decode(cls, txt):
-        symbol_table = [
-            ('_', '(ﾟДﾟ) [ﾟΘﾟ]'),
-            ('a', '(ﾟДﾟ) [ﾟωﾟﾉ]'),
-            ('b', '(ﾟДﾟ) [ﾟΘﾟﾉ]'),
-            ('c', '(ﾟДﾟ) [\'c\']'),
-            ('d', '(ﾟДﾟ) [ﾟｰﾟﾉ]'),
-            ('e', '(ﾟДﾟ) [ﾟДﾟﾉ]'),
-            ('f', '(ﾟДﾟ) [1]'),
-
-            ('o', '(ﾟДﾟ) [\'o\']'),
-            ('u', '(oﾟｰﾟo)'),
-            ('c', '(ﾟДﾟ) [\'c\']'),
-
-            ('7', '((ﾟｰﾟ) + (o^_^o))'),
-            ('6', '((o^_^o) +(o^_^o) +(c^_^o))'),
-            ('5', '((ﾟｰﾟ) + (ﾟΘﾟ))'),
-            ('4', '(-~3)'),
-            ('3', '(-~-~1)'),
-            ('2', '(-~1)'),
-            ('1', '(-~0)'),
-            ('0', '((c^_^o)-(c^_^o))'),
-        ]
-        delim = '(ﾟДﾟ)[ﾟεﾟ]+'
-        ret = ''
-        for aachar in txt.split(delim):
-            for val, pat in symbol_table:
-                aachar = aachar.replace(pat, val)
-            aachar = aachar.replace('+ ', '')
-            m = re.match(r'^\d+', aachar)
-            if m:
-                ret += compat_chr(int(m.group(0), 8))
-            else:
-                m = re.match(r'^u([\da-f]+)', aachar)
-                if m:
-                    ret += compat_chr(int(m.group(1), 16))
-        return cls.openload_level2(ret)
-
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
@@ -102,29 +47,77 @@ class OpenloadIE(InfoExtractor):
         if 'File not found' in webpage:
             raise ExtractorError('File not found', expected=True)
 
-        code = self._search_regex(
-            r'</video>\s*</div>\s*<script[^>]+>[^>]+</script>\s*<script[^>]+>([^<]+)</script>',
-            webpage, 'JS code')
+        # The following extraction logic is proposed by @Belderak and @gdkchan
+        # and declared to be used freely in youtube-dl
+        # See https://github.com/rg3/youtube-dl/issues/9706
 
-        decoded = self.openload_decode(code)
+        numbers_js = self._download_webpage(
+            'https://openload.co/assets/js/obfuscator/n.js', video_id,
+            note='Downloading signature numbers')
+        signums = self._search_regex(
+            r'window\.signatureNumbers\s*=\s*[\'"](?P<data>[a-z]+)[\'"]',
+            numbers_js, 'signature numbers', group='data')
 
-        video_url = self._search_regex(
-            r'return\s+"(https?://[^"]+)"', decoded, 'video URL')
+        linkimg_uri = self._search_regex(
+            r'<img[^>]+id="linkimg"[^>]+src="([^"]+)"', webpage, 'link image')
+        linkimg = self._request_webpage(
+            linkimg_uri, video_id, note=False).read()
+
+        width, height, pixels = decode_png(linkimg)
+
+        output = ''
+        for y in range(height):
+            for x in range(width):
+                r, g, b = pixels[y][3 * x:3 * x + 3]
+                if r == 0 and g == 0 and b == 0:
+                    break
+                else:
+                    output += compat_chr(r)
+                    output += compat_chr(g)
+                    output += compat_chr(b)
+
+        img_str_length = len(output) // 200
+        img_str = [[0 for x in range(img_str_length)] for y in range(10)]
+
+        sig_str_length = len(signums) // 260
+        sig_str = [[0 for x in range(sig_str_length)] for y in range(10)]
+
+        for i in range(10):
+            for j in range(img_str_length):
+                begin = i * img_str_length * 20 + j * 20
+                img_str[i][j] = output[begin:begin + 20]
+            for j in range(sig_str_length):
+                begin = i * sig_str_length * 26 + j * 26
+                sig_str[i][j] = signums[begin:begin + 26]
+
+        parts = []
+        # TODO: find better names for str_, chr_ and sum_
+        str_ = ''
+        for i in [2, 3, 5, 7]:
+            str_ = ''
+            sum_ = float(99)
+            for j in range(len(sig_str[i])):
+                for chr_idx in range(len(img_str[i][j])):
+                    if sum_ > float(122):
+                        sum_ = float(98)
+                    chr_ = compat_chr(int(math.floor(sum_)))
+                    if sig_str[i][j][chr_idx] == chr_ and j >= len(str_):
+                        sum_ += float(2.5)
+                        str_ += img_str[i][j][chr_idx]
+            parts.append(str_.replace(',', ''))
+
+        video_url = 'https://openload.co/stream/%s~%s~%s~%s' % (parts[3], parts[1], parts[2], parts[0])
 
         title = self._og_search_title(webpage, default=None) or self._search_regex(
             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
             'title', default=None) or self._html_search_meta(
             'description', webpage, 'title', fatal=True)
 
-        ext = mimetype2ext(self._search_regex(
-            r'window\.vt\s*=\s*(["\'])(?P<mimetype>.+?)\1', decoded,
-            'mimetype', default=None, group='mimetype')) or determine_ext(
-            video_url, 'mp4')
-
         return {
             'id': video_id,
             'title': title,
-            'ext': ext,
             'thumbnail': self._og_search_thumbnail(webpage, default=None),
             'url': video_url,
+            # Seems all videos have extensions in their titles
+            'ext': determine_ext(title),
         }

youtube_dl/extractor/pbs.py

@@ -4,13 +4,13 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_HTTPError
 from ..utils import (
     ExtractorError,
     determine_ext,
     int_or_none,
     js_to_json,
     strip_jsonp,
+    strip_or_none,
     unified_strdate,
     US_RATINGS,
 )
@@ -201,7 +201,7 @@ class PBSIE(InfoExtractor):
                 'id': '2365006249',
                 'ext': 'mp4',
                 'title': 'Constitution USA with Peter Sagal - A More Perfect Union',
-                'description': 'md5:36f341ae62e251b8f5bd2b754b95a071',
+                'description': 'md5:31b664af3c65fd07fa460d306b837d00',
                 'duration': 3190,
             },
         },
@@ -212,7 +212,7 @@ class PBSIE(InfoExtractor):
                 'id': '2365297690',
                 'ext': 'mp4',
                 'title': 'FRONTLINE - Losing Iraq',
-                'description': 'md5:4d3eaa01f94e61b3e73704735f1196d9',
+                'description': 'md5:5979a4d069b157f622d02bff62fbe654',
                 'duration': 5050,
             },
         },
@@ -223,7 +223,7 @@ class PBSIE(InfoExtractor):
                 'id': '2201174722',
                 'ext': 'mp4',
                 'title': 'PBS NewsHour - Cyber Schools Gain Popularity, but Quality Questions Persist',
-                'description': 'md5:95a19f568689d09a166dff9edada3301',
+                'description': 'md5:86ab9a3d04458b876147b355788b8781',
                 'duration': 801,
             },
         },
@@ -268,7 +268,7 @@ class PBSIE(InfoExtractor):
                 'display_id': 'player',
                 'ext': 'mp4',
                 'title': 'American Experience - Death and the Civil War, Chapter 1',
-                'description': 'md5:1b80a74e0380ed2a4fb335026de1600d',
+                'description': 'md5:67fa89a9402e2ee7d08f53b920674c18',
                 'duration': 682,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
@@ -294,13 +294,13 @@ class PBSIE(InfoExtractor):
             # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see
             # https://github.com/rg3/youtube-dl/issues/7059)
             'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/',
-            'md5': '84ced42850d78f1d4650297356e95e6f',
+            'md5': '59b0ef5009f9ac8a319cc5efebcd865e',
             'info_dict': {
                 'id': '2365546844',
                 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business',
                 'ext': 'mp4',
                 'title': "A Chef's Life - Season 3, Ep. 5: Prickly Business",
-                'description': 'md5:54033c6baa1f9623607c6e2ed245888b',
+                'description': 'md5:c0ff7475a4b70261c7e58f493c2792a5',
                 'duration': 1480,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
@@ -313,7 +313,7 @@ class PBSIE(InfoExtractor):
                 'display_id': 'the-atomic-artists',
                 'ext': 'mp4',
                 'title': 'FRONTLINE - The Atomic Artists',
-                'description': 'md5:1a2481e86b32b2e12ec1905dd473e2c1',
+                'description': 'md5:f677e4520cfacb4a5ce1471e31b57800',
                 'duration': 723,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
@@ -324,7 +324,7 @@ class PBSIE(InfoExtractor):
         {
             # Serves hd only via wigget/partnerplayer page
             'url': 'http://www.pbs.org/video/2365641075/',
-            'md5': 'acfd4c400b48149a44861cb16dd305cf',
+            'md5': 'fdf907851eab57211dd589cf12006666',
             'info_dict': {
                 'id': '2365641075',
                 'ext': 'mp4',
@@ -353,11 +353,16 @@ class PBSIE(InfoExtractor):
     def _extract_webpage(self, url):
         mobj = re.match(self._VALID_URL, url)
 
+        description = None
+
         presumptive_id = mobj.group('presumptive_id')
         display_id = presumptive_id
         if presumptive_id:
             webpage = self._download_webpage(url, display_id)
 
+            description = strip_or_none(self._og_search_description(
+                webpage, default=None) or self._html_search_meta(
+                'description', webpage, default=None))
             upload_date = unified_strdate(self._search_regex(
                 r'<input type="hidden" id="air_date_[0-9]+" value="([^"]+)"',
                 webpage, 'upload date', default=None))
@@ -370,7 +375,7 @@ class PBSIE(InfoExtractor):
             for p in MULTI_PART_REGEXES:
                 tabbed_videos = re.findall(p, webpage)
                 if tabbed_videos:
-                    return tabbed_videos, presumptive_id, upload_date
+                    return tabbed_videos, presumptive_id, upload_date, description
 
             MEDIA_ID_REGEXES = [
                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
@@ -382,7 +387,7 @@ class PBSIE(InfoExtractor):
             media_id = self._search_regex(
                 MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
             if media_id:
-                return media_id, presumptive_id, upload_date
+                return media_id, presumptive_id, upload_date, description
 
             # Fronline video embedded via flp
             video_id = self._search_regex(
@@ -399,7 +404,7 @@ class PBSIE(InfoExtractor):
                     'http://www.pbs.org/wgbh/pages/frontline/.json/getdir/getdir%d.json' % prg_id,
                     presumptive_id, 'Downloading getdir JSON',
                     transform_source=strip_jsonp)
-                return getdir['mid'], presumptive_id, upload_date
+                return getdir['mid'], presumptive_id, upload_date, description
 
             for iframe in re.findall(r'(?s)<iframe(.+?)></iframe>', webpage):
                 url = self._search_regex(
@@ -423,10 +428,10 @@ class PBSIE(InfoExtractor):
             video_id = mobj.group('id')
             display_id = video_id
 
-        return video_id, display_id, None
+        return video_id, display_id, None, description
 
     def _real_extract(self, url):
-        video_id, display_id, upload_date = self._extract_webpage(url)
+        video_id, display_id, upload_date, description = self._extract_webpage(url)
 
         if isinstance(video_id, list):
             entries = [self.url_result(
@@ -448,17 +453,6 @@ class PBSIE(InfoExtractor):
                     redirects.append(redirect)
                     redirect_urls.add(redirect_url)
 
-        try:
-            video_info = self._download_json(
-                'http://player.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
-                display_id, 'Downloading video info JSON')
-            extract_redirect_urls(video_info)
-            info = video_info
-        except ExtractorError as e:
-            # videoInfo API may not work for some videos
-            if not isinstance(e.cause, compat_HTTPError) or e.cause.code != 404:
-                raise
-
         # Player pages may also serve different qualities
         for page in ('widget/partnerplayer', 'portalplayer'):
             player = self._download_webpage(
@@ -511,15 +505,19 @@ class PBSIE(InfoExtractor):
             formats))
         if http_url:
             for m3u8_format in m3u8_formats:
-                bitrate = self._search_regex(r'(\d+k)', m3u8_format['url'], 'bitrate', default=None)
-                # extract only the formats that we know that they will be available as http format.
-                # https://projects.pbs.org/confluence/display/coveapi/COVE+Video+Specifications
-                if not bitrate or bitrate not in ('400k', '800k', '1200k', '2500k'):
+                bitrate = self._search_regex(r'(\d+)k', m3u8_format['url'], 'bitrate', default=None)
+                # Lower qualities (150k and 192k) are not available as HTTP formats (see [1]),
+                # we won't try extracting them.
+                # Since summer 2016 higher quality formats (4500k and 6500k) are also available
+                # albeit they are not documented in [2].
+                # 1. https://github.com/rg3/youtube-dl/commit/cbc032c8b70a038a69259378c92b4ba97b42d491#commitcomment-17313656
+                # 2. https://projects.pbs.org/confluence/display/coveapi/COVE+Video+Specifications
+                if not bitrate or int(bitrate) < 400:
                     continue
-                f_url = re.sub(r'\d+k|baseline', bitrate, http_url)
+                f_url = re.sub(r'\d+k|baseline', bitrate + 'k', http_url)
                 # This may produce invalid links sometimes (e.g.
                 # http://www.pbs.org/wgbh/frontline/film/suicide-plan)
-                if not self._is_valid_url(f_url, display_id, 'http-%s video' % bitrate):
+                if not self._is_valid_url(f_url, display_id, 'http-%sk video' % bitrate):
                     continue
                 f = m3u8_format.copy()
                 f.update({
@@ -562,11 +560,14 @@ class PBSIE(InfoExtractor):
         if alt_title:
             info['title'] = alt_title + ' - ' + re.sub(r'^' + alt_title + '[\s\-:]+', '', info['title'])
 
+        description = info.get('description') or info.get(
+            'program', {}).get('description') or description
+
         return {
             'id': video_id,
             'display_id': display_id,
             'title': info['title'],
-            'description': info.get('description') or info.get('program', {}).get('description'),
+            'description': description,
             'thumbnail': info.get('image_url'),
             'duration': int_or_none(info.get('duration')),
             'age_limit': age_limit,

youtube_dl/extractor/pokemon.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    extract_attributes,
+    int_or_none,
+)
+
+
+class PokemonIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?pokemon\.com/[a-z]{2}(?:.*?play=(?P<id>[a-z0-9]{32})|/[^/]+/\d+_\d+-(?P<display_id>[^/?#]+))'
+    _TESTS = [{
+        'url': 'http://www.pokemon.com/us/pokemon-episodes/19_01-from-a-to-z/?play=true',
+        'md5': '9fb209ae3a569aac25de0f5afc4ee08f',
+        'info_dict': {
+            'id': 'd0436c00c3ce4071ac6cee8130ac54a1',
+            'ext': 'mp4',
+            'title': 'From A to Z!',
+            'description': 'Bonnie makes a new friend, Ash runs into an old friend, and a terrifying premonition begins to unfold!',
+            'timestamp': 1460478136,
+            'upload_date': '20160412',
+        },
+        'add_id': ['LimelightMedia']
+    }, {
+        'url': 'http://www.pokemon.com/uk/pokemon-episodes/?play=2e8b5c761f1d4a9286165d7748c1ece2',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.pokemon.com/fr/episodes-pokemon/18_09-un-hiver-inattendu/',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.pokemon.com/de/pokemon-folgen/01_20-bye-bye-smettbo/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
+        webpage = self._download_webpage(url, video_id or display_id)
+        video_data = extract_attributes(self._search_regex(
+            r'(<[^>]+data-video-id="%s"[^>]*>)' % (video_id if video_id else '[a-z0-9]{32}'),
+            webpage, 'video data element'))
+        video_id = video_data['data-video-id']
+        title = video_data['data-video-title']
+        return {
+            '_type': 'url_transparent',
+            'id': video_id,
+            'url': 'limelight:media:%s' % video_id,
+            'title': title,
+            'description': video_data.get('data-video-summary'),
+            'thumbnail': video_data.get('data-video-poster'),
+            'series': 'Pokémon',
+            'season_number': int_or_none(video_data.get('data-video-season')),
+            'episode': title,
+            'episode_number': int_or_none(video_data.get('data-video-episode')),
+            'ie_key': 'LimelightMedia',
+        }

youtube_dl/extractor/pornotube.py

@@ -3,10 +3,7 @@ from __future__ import unicode_literals
 import json
 
 from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    sanitized_Request,
-)
+from ..utils import int_or_none
 
 
 class PornotubeIE(InfoExtractor):
@@ -31,59 +28,55 @@ class PornotubeIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        # Fetch origin token
-        js_config = self._download_webpage(
-            'http://www.pornotube.com/assets/src/app/config.js', video_id,
-            note='Download JS config')
-        originAuthenticationSpaceKey = self._search_regex(
-            r"constant\('originAuthenticationSpaceKey',\s*'([^']+)'",
-            js_config, 'originAuthenticationSpaceKey')
+        token = self._download_json(
+            'https://api.aebn.net/auth/v2/origins/authenticate',
+            video_id, note='Downloading token',
+            data=json.dumps({'credentials': 'Clip Application'}).encode('utf-8'),
+            headers={
+                'Content-Type': 'application/json',
+                'Origin': 'http://www.pornotube.com',
+            })['tokenKey']
 
-        # Fetch actual token
-        token_req_data = {
-            'authenticationSpaceKey': originAuthenticationSpaceKey,
-            'credentials': 'Clip Application',
-        }
-        token_req = sanitized_Request(
-            'https://api.aebn.net/auth/v1/token/primal',
-            data=json.dumps(token_req_data).encode('utf-8'))
-        token_req.add_header('Content-Type', 'application/json')
-        token_req.add_header('Origin', 'http://www.pornotube.com')
-        token_answer = self._download_json(
-            token_req, video_id, note='Requesting primal token')
-        token = token_answer['tokenKey']
+        video_url = self._download_json(
+            'https://api.aebn.net/delivery/v1/clips/%s/MP4' % video_id,
+            video_id, note='Downloading delivery information',
+            headers={'Authorization': token})['mediaUrl']
 
-        # Get video URL
-        delivery_req = sanitized_Request(
-            'https://api.aebn.net/delivery/v1/clips/%s/MP4' % video_id)
-        delivery_req.add_header('Authorization', token)
-        delivery_info = self._download_json(
-            delivery_req, video_id, note='Downloading delivery information')
-        video_url = delivery_info['mediaUrl']
+        FIELDS = (
+            'title', 'description', 'startSecond', 'endSecond', 'publishDate',
+            'studios{name}', 'categories{name}', 'movieId', 'primaryImageNumber'
+        )
 
-        # Get additional info (title etc.)
-        info_req = sanitized_Request(
-            'https://api.aebn.net/content/v1/clips/%s?expand='
-            'title,description,primaryImageNumber,startSecond,endSecond,'
-            'movie.title,movie.MovieId,movie.boxCoverFront,movie.stars,'
-            'movie.studios,stars.name,studios.name,categories.name,'
-            'clipActive,movieActive,publishDate,orientations' % video_id)
-        info_req.add_header('Authorization', token)
         info = self._download_json(
-            info_req, video_id, note='Downloading metadata')
+            'https://api.aebn.net/content/v2/clips/%s?fields=%s'
+            % (video_id, ','.join(FIELDS)), video_id,
+            note='Downloading metadata',
+            headers={'Authorization': token})
+
+        if isinstance(info, list):
+            info = info[0]
+
+        title = info['title']
 
         timestamp = int_or_none(info.get('publishDate'), scale=1000)
         uploader = info.get('studios', [{}])[0].get('name')
-        movie_id = info['movie']['movieId']
+        movie_id = info.get('movieId')
+        primary_image_number = info.get('primaryImageNumber')
+        thumbnail = None
+        if movie_id and primary_image_number:
             thumbnail = 'http://pic.aebn.net/dis/t/%s/%s_%08d.jpg' % (
-            movie_id, movie_id, info['primaryImageNumber'])
-        categories = [c['name'] for c in info.get('categories')]
+                movie_id, movie_id, primary_image_number)
+        start = int_or_none(info.get('startSecond'))
+        end = int_or_none(info.get('endSecond'))
+        duration = end - start if start and end else None
+        categories = [c['name'] for c in info.get('categories', []) if c.get('name')]
 
         return {
             'id': video_id,
             'url': video_url,
-            'title': info['title'],
+            'title': title,
             'description': info.get('description'),
+            'duration': duration,
             'timestamp': timestamp,
             'uploader': uploader,
             'thumbnail': thumbnail,

youtube_dl/extractor/rbmaradio.py

@@ -1,55 +1,71 @@
-# encoding: utf-8
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    ExtractorError,
+    clean_html,
+    int_or_none,
+    unified_timestamp,
+    update_url_query,
 )
 
 
 class RBMARadioIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rbmaradio\.com/shows/(?P<videoID>[^/]+)$'
+    _VALID_URL = r'https?://(?:www\.)?rbmaradio\.com/shows/(?P<show_id>[^/]+)/episodes/(?P<id>[^/?#&]+)'
     _TEST = {
-        'url': 'http://www.rbmaradio.com/shows/ford-lopatin-live-at-primavera-sound-2011',
+        'url': 'https://www.rbmaradio.com/shows/main-stage/episodes/ford-lopatin-live-at-primavera-sound-2011',
         'md5': '6bc6f9bcb18994b4c983bc3bf4384d95',
         'info_dict': {
             'id': 'ford-lopatin-live-at-primavera-sound-2011',
             'ext': 'mp3',
-            'uploader_id': 'ford-lopatin',
-            'location': 'Spain',
-            'description': 'Joel Ford and Daniel ’Oneohtrix Point Never’ Lopatin fly their midified pop extravaganza to Spain. Live at Primavera Sound 2011.',
-            'uploader': 'Ford & Lopatin',
-            'title': 'Live at Primavera Sound 2011',
+            'title': 'Main Stage - Ford & Lopatin',
+            'description': 'md5:4f340fb48426423530af5a9d87bd7b91',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 2452,
+            'timestamp': 1307103164,
+            'upload_date': '20110603',
         },
     }
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('videoID')
+        mobj = re.match(self._VALID_URL, url)
+        show_id = mobj.group('show_id')
+        episode_id = mobj.group('id')
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, episode_id)
 
-        json_data = self._search_regex(r'window\.gon.*?gon\.show=(.+?);$',
-                                       webpage, 'json data', flags=re.MULTILINE)
+        episode = self._parse_json(
+            self._search_regex(
+                r'__INITIAL_STATE__\s*=\s*({.+?})\s*</script>',
+                webpage, 'json data'),
+            episode_id)['episodes'][show_id][episode_id]
 
-        try:
-            data = json.loads(json_data)
-        except ValueError as e:
-            raise ExtractorError('Invalid JSON: ' + str(e))
+        title = episode['title']
 
-        video_url = data['akamai_url'] + '&cbr=256'
+        show_title = episode.get('showTitle')
+        if show_title:
+            title = '%s - %s' % (show_title, title)
+
+        formats = [{
+            'url': update_url_query(episode['audioURL'], query={'cbr': abr}),
+            'format_id': compat_str(abr),
+            'abr': abr,
+            'vcodec': 'none',
+        } for abr in (96, 128, 256)]
+
+        description = clean_html(episode.get('longTeaser'))
+        thumbnail = self._proto_relative_url(episode.get('imageURL', {}).get('landscape'))
+        duration = int_or_none(episode.get('duration'))
+        timestamp = unified_timestamp(episode.get('publishedAt'))
 
         return {
-            'id': video_id,
-            'url': video_url,
-            'title': data['title'],
-            'description': data.get('teaser_text'),
-            'location': data.get('country_of_origin'),
-            'uploader': data.get('host', {}).get('name'),
-            'uploader_id': data.get('host', {}).get('slug'),
-            'thumbnail': data.get('image', {}).get('large_url_2x'),
-            'duration': data.get('duration'),
+            'id': episode_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'timestamp': timestamp,
+            'formats': formats,
         }

youtube_dl/extractor/rozhlas.py

@@ -0,0 +1,50 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    remove_start,
+)
+
+
+class RozhlasIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?prehravac\.rozhlas\.cz/audio/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://prehravac.rozhlas.cz/audio/3421320',
+        'md5': '504c902dbc9e9a1fd50326eccf02a7e2',
+        'info_dict': {
+            'id': '3421320',
+            'ext': 'mp3',
+            'title': 'Echo Pavla Klusáka (30.06.2015 21:00)',
+            'description': 'Osmdesátiny Terryho Rileyho jsou skvělou příležitostí proletět se elektronickými i akustickými díly zakladatatele minimalismu, který je aktivní už přes padesát let'
+        }
+    }, {
+        'url': 'http://prehravac.rozhlas.cz/audio/3421320/embed',
+        'skip_download': True,
+    }]
+
+    def _real_extract(self, url):
+        audio_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://prehravac.rozhlas.cz/audio/%s' % audio_id, audio_id)
+
+        title = self._html_search_regex(
+            r'<h3>(.+?)</h3>\s*<p[^>]*>.*?</p>\s*<div[^>]+id=["\']player-track',
+            webpage, 'title', default=None) or remove_start(
+            self._og_search_title(webpage), 'Radio Wave - ')
+        description = self._html_search_regex(
+            r'<p[^>]+title=(["\'])(?P<url>(?:(?!\1).)+)\1[^>]*>.*?</p>\s*<div[^>]+id=["\']player-track',
+            webpage, 'description', fatal=False, group='url')
+        duration = int_or_none(self._search_regex(
+            r'data-duration=["\'](\d+)', webpage, 'duration', default=None))
+
+        return {
+            'id': audio_id,
+            'url': 'http://media.rozhlas.cz/_audio/%s.mp3' % audio_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'vcodec': 'none',
+        }

youtube_dl/extractor/rtlnl.py

@@ -14,7 +14,7 @@ class RtlNlIE(InfoExtractor):
     _VALID_URL = r'''(?x)
         https?://(?:www\.)?
         (?:
-            rtlxl\.nl/\#!/[^/]+/|
+            rtlxl\.nl/[^\#]*\#!/[^/]+/|
             rtl\.nl/system/videoplayer/(?:[^/]+/)+(?:video_)?embed\.html\b.+?\buuid=
         )
         (?P<id>[0-9a-f-]+)'''
@@ -67,6 +67,9 @@ class RtlNlIE(InfoExtractor):
     }, {
         'url': 'http://www.rtl.nl/system/videoplayer/derden/embed.html#!/uuid=bb0353b0-d6a4-1dad-90e9-18fe75b8d1f0',
         'only_matching': True,
+    }, {
+        'url': 'http://rtlxl.nl/?_ga=1.204735956.572365465.1466978370#!/rtl-nieuws-132237/3c487912-023b-49ac-903e-2c5d79f8410f',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/safari.py

@@ -75,7 +75,7 @@ class SafariBaseIE(InfoExtractor):
 class SafariIE(SafariBaseIE):
     IE_NAME = 'safari'
     IE_DESC = 'safaribooksonline.com online video'
-    _VALID_URL = r'https?://(?:www\.)?safaribooksonline\.com/library/view/[^/]+/(?P<course_id>[^/]+)/(?P<part>part\d+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?safaribooksonline\.com/library/view/[^/]+/(?P<course_id>[^/]+)/(?P<part>[^/?#&]+)\.html'
 
     _TESTS = [{
         'url': 'https://www.safaribooksonline.com/library/view/hadoop-fundamentals-livelessons/9780133392838/part00.html',
@@ -92,6 +92,9 @@ class SafariIE(SafariBaseIE):
         # non-digits in course id
         'url': 'https://www.safaribooksonline.com/library/view/create-a-nodejs/100000006A0210/part00.html',
         'only_matching': True,
+    }, {
+        'url': 'https://www.safaribooksonline.com/library/view/learning-path-red/9780134664057/RHCE_Introduction.html',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -132,12 +135,15 @@ class SafariIE(SafariBaseIE):
 
 class SafariApiIE(SafariBaseIE):
     IE_NAME = 'safari:api'
-    _VALID_URL = r'https?://(?:www\.)?safaribooksonline\.com/api/v1/book/(?P<course_id>[^/]+)/chapter(?:-content)?/(?P<part>part\d+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?safaribooksonline\.com/api/v1/book/(?P<course_id>[^/]+)/chapter(?:-content)?/(?P<part>[^/?#&]+)\.html'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'https://www.safaribooksonline.com/api/v1/book/9780133392838/chapter/part00.html',
         'only_matching': True,
-    }
+    }, {
+        'url': 'https://www.safaribooksonline.com/api/v1/book/9780134664057/chapter/RHCE_Introduction.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/sohu.py

@@ -14,10 +14,10 @@ from ..utils import ExtractorError
 class SohuIE(InfoExtractor):
     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
 
+    # Sohu videos give different MD5 sums on Travis CI and my machine
     _TESTS = [{
         'note': 'This video is available only in Mainland China',
         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
-        'md5': '29175c8cadd8b5cc4055001e85d6b372',
         'info_dict': {
             'id': '382479172',
             'ext': 'mp4',
@@ -26,7 +26,6 @@ class SohuIE(InfoExtractor):
         'skip': 'On available in China',
     }, {
         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
-        'md5': '699060e75cf58858dd47fb9c03c42cfb',
         'info_dict': {
             'id': '409385080',
             'ext': 'mp4',
@@ -34,7 +33,6 @@ class SohuIE(InfoExtractor):
         }
     }, {
         'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
-        'md5': '9bf34be48f2f4dadcb226c74127e203c',
         'info_dict': {
             'id': '78693464',
             'ext': 'mp4',
@@ -48,7 +46,6 @@ class SohuIE(InfoExtractor):
             'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
         },
         'playlist': [{
-            'md5': 'bdbfb8f39924725e6589c146bc1883ad',
             'info_dict': {
                 'id': '78910339_part1',
                 'ext': 'mp4',
@@ -56,7 +53,6 @@ class SohuIE(InfoExtractor):
                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
             }
         }, {
-            'md5': '3e1f46aaeb95354fd10e7fca9fc1804e',
             'info_dict': {
                 'id': '78910339_part2',
                 'ext': 'mp4',
@@ -64,7 +60,6 @@ class SohuIE(InfoExtractor):
                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
             }
         }, {
-            'md5': '8407e634175fdac706766481b9443450',
             'info_dict': {
                 'id': '78910339_part3',
                 'ext': 'mp4',

youtube_dl/extractor/sonyliv.py

@@ -0,0 +1,34 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class SonyLIVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?sonyliv\.com/details/[^/]+/(?P<id>\d+)'
+    _TESTS = [{
+        'url': "http://www.sonyliv.com/details/episodes/5024612095001/Ep.-1---Achaari-Cheese-Toast---Bachelor's-Delight",
+        'info_dict': {
+            'title': "Ep. 1 - Achaari Cheese Toast - Bachelor's Delight",
+            'id': '5024612095001',
+            'ext': 'mp4',
+            'upload_date': '20160707',
+            'description': 'md5:7f28509a148d5be9d0782b4d5106410d',
+            'uploader_id': '4338955589001',
+            'timestamp': 1467870968,
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': ['BrightcoveNew'],
+    }, {
+        'url': 'http://www.sonyliv.com/details/full%20movie/4951168986001/Sei-Raat-(Bangla)',
+        'only_matching': True,
+    }]
+
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/4338955589001/default_default/index.html?videoId=%s'
+
+    def _real_extract(self, url):
+        brightcove_id = self._match_id(url)
+        return self.url_result(
+            self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew', brightcove_id)

youtube_dl/extractor/sunporno.py

@@ -12,25 +12,29 @@ from ..utils import (
 
 
 class SunPornoIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?sunporno\.com/videos/(?P<id>\d+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:(?:www\.)?sunporno\.com/videos|embeds\.sunporno\.com/embed)/(?P<id>\d+)'
+    _TESTS = [{
         'url': 'http://www.sunporno.com/videos/807778/',
-        'md5': '6457d3c165fd6de062b99ef6c2ff4c86',
+        'md5': '507887e29033502f29dba69affeebfc9',
         'info_dict': {
             'id': '807778',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'md5:0a400058e8105d39e35c35e7c5184164',
             'description': 'md5:a31241990e1bd3a64e72ae99afb325fb',
             'thumbnail': 're:^https?://.*\.jpg$',
             'duration': 302,
             'age_limit': 18,
         }
-    }
+    }, {
+        'url': 'http://embeds.sunporno.com/embed/807778',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(
+            'http://www.sunporno.com/videos/%s' % video_id, video_id)
 
         title = self._html_search_regex(
             r'<title>([^<]+)</title>', webpage, 'title')
@@ -40,7 +44,8 @@ class SunPornoIE(InfoExtractor):
             r'poster="([^"]+)"', webpage, 'thumbnail', fatal=False)
 
         duration = parse_duration(self._search_regex(
-            r'itemprop="duration">\s*(\d+:\d+)\s*<',
+            (r'itemprop="duration"[^>]*>\s*(\d+:\d+)\s*<',
+             r'>Duration:\s*<span[^>]+>\s*(\d+:\d+)\s*<'),
             webpage, 'duration', fatal=False))
 
         view_count = int_or_none(self._html_search_regex(
@@ -48,7 +53,7 @@ class SunPornoIE(InfoExtractor):
             webpage, 'view count', fatal=False))
         comment_count = int_or_none(self._html_search_regex(
             r'(\d+)</b> Comments?',
-            webpage, 'comment count', fatal=False))
+            webpage, 'comment count', fatal=False, default=None))
 
         formats = []
         quality = qualities(['mp4', 'flv'])

youtube_dl/extractor/tapely.py

@@ -1,109 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    clean_html,
-    ExtractorError,
-    float_or_none,
-    parse_iso8601,
-    sanitized_Request,
-)
-
-
-class TapelyIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?(?:tape\.ly|tapely\.com)/(?P<id>[A-Za-z0-9\-_]+)(?:/(?P<songnr>\d+))?'
-    _API_URL = 'http://tape.ly/showtape?id={0:}'
-    _S3_SONG_URL = 'http://mytape.s3.amazonaws.com/{0:}'
-    _SOUNDCLOUD_SONG_URL = 'http://api.soundcloud.com{0:}'
-    _TESTS = [
-        {
-            'url': 'http://tape.ly/my-grief-as-told-by-water',
-            'info_dict': {
-                'id': 23952,
-                'title': 'my grief as told by water',
-                'thumbnail': 're:^https?://.*\.png$',
-                'uploader_id': 16484,
-                'timestamp': 1411848286,
-                'description': 'For Robin and Ponkers, whom the tides of life have taken out to sea.',
-            },
-            'playlist_count': 13,
-        },
-        {
-            'url': 'http://tape.ly/my-grief-as-told-by-water/1',
-            'md5': '79031f459fdec6530663b854cbc5715c',
-            'info_dict': {
-                'id': 258464,
-                'title': 'Dreaming Awake  (My Brightest Diamond)',
-                'ext': 'm4a',
-            },
-        },
-        {
-            'url': 'https://tapely.com/my-grief-as-told-by-water',
-            'only_matching': True,
-        },
-    ]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('id')
-
-        playlist_url = self._API_URL.format(display_id)
-        request = sanitized_Request(playlist_url)
-        request.add_header('X-Requested-With', 'XMLHttpRequest')
-        request.add_header('Accept', 'application/json')
-        request.add_header('Referer', url)
-
-        playlist = self._download_json(request, display_id)
-
-        tape = playlist['tape']
-
-        entries = []
-        for s in tape['songs']:
-            song = s['song']
-            entry = {
-                'id': song['id'],
-                'duration': float_or_none(song.get('songduration'), 1000),
-                'title': song['title'],
-            }
-            if song['source'] == 'S3':
-                entry.update({
-                    'url': self._S3_SONG_URL.format(song['filename']),
-                })
-                entries.append(entry)
-            elif song['source'] == 'YT':
-                self.to_screen('YouTube video detected')
-                yt_id = song['filename'].replace('/youtube/', '')
-                entry.update(self.url_result(yt_id, 'Youtube', video_id=yt_id))
-                entries.append(entry)
-            elif song['source'] == 'SC':
-                self.to_screen('SoundCloud song detected')
-                sc_url = self._SOUNDCLOUD_SONG_URL.format(song['filename'])
-                entry.update(self.url_result(sc_url, 'Soundcloud'))
-                entries.append(entry)
-            else:
-                self.report_warning('Unknown song source: %s' % song['source'])
-
-        if mobj.group('songnr'):
-            songnr = int(mobj.group('songnr')) - 1
-            try:
-                return entries[songnr]
-            except IndexError:
-                raise ExtractorError(
-                    'No song with index: %s' % mobj.group('songnr'),
-                    expected=True)
-
-        return {
-            '_type': 'playlist',
-            'id': tape['id'],
-            'display_id': display_id,
-            'title': tape['name'],
-            'entries': entries,
-            'thumbnail': tape.get('image_url'),
-            'description': clean_html(tape.get('subtext')),
-            'like_count': tape.get('likescount'),
-            'uploader_id': tape.get('user_id'),
-            'timestamp': parse_iso8601(tape.get('published_at')),
-        }

youtube_dl/extractor/tnaflix.py

@@ -118,8 +118,12 @@ class TNAFlixNetworkBaseIE(InfoExtractor):
             xpath_text(cfg_xml, './startThumb', 'thumbnail'), 'http:')
         thumbnails = self._extract_thumbnails(cfg_xml)
 
+        title = None
+        if self._TITLE_REGEX:
             title = self._html_search_regex(
-            self._TITLE_REGEX, webpage, 'title') if self._TITLE_REGEX else self._og_search_title(webpage)
+                self._TITLE_REGEX, webpage, 'title', default=None)
+        if not title:
+            title = self._og_search_title(webpage)
 
         age_limit = self._rta_search(webpage) or 18
 
@@ -189,9 +193,9 @@ class TNAFlixNetworkEmbedIE(TNAFlixNetworkBaseIE):
 class TNAFlixIE(TNAFlixNetworkBaseIE):
     _VALID_URL = r'https?://(?:www\.)?tnaflix\.com/[^/]+/(?P<display_id>[^/]+)/video(?P<id>\d+)'
 
-    _TITLE_REGEX = r'<title>(.+?) - TNAFlix Porn Videos</title>'
-    _DESCRIPTION_REGEX = r'<meta[^>]+name="description"[^>]+content="([^"]+)"'
-    _UPLOADER_REGEX = r'<i>\s*Verified Member\s*</i>\s*<h1>(.+?)</h1>'
+    _TITLE_REGEX = r'<title>(.+?) - (?:TNAFlix Porn Videos|TNAFlix\.com)</title>'
+    _DESCRIPTION_REGEX = r'(?s)>Description:</[^>]+>(.+?)<'
+    _UPLOADER_REGEX = r'<i>\s*Verified Member\s*</i>\s*<h\d+>(.+?)<'
     _CATEGORIES_REGEX = r'(?s)<span[^>]*>Categories:</span>(.+?)</div>'
 
     _TESTS = [{

youtube_dl/extractor/tvp.py

@@ -24,6 +24,7 @@ class TVPIE(InfoExtractor):
             'id': '194536',
             'ext': 'mp4',
             'title': 'Czas honoru, I seria – odc. 13',
+            'description': 'md5:76649d2014f65c99477be17f23a4dead',
         },
     }, {
         'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
@@ -32,6 +33,16 @@ class TVPIE(InfoExtractor):
             'id': '17916176',
             'ext': 'mp4',
             'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
+            'description': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
+        },
+    }, {
+        # page id is not the same as video id(#7799)
+        'url': 'http://vod.tvp.pl/22704887/08122015-1500',
+        'md5': 'cf6a4705dfd1489aef8deb168d6ba742',
+        'info_dict': {
+            'id': '22680786',
+            'ext': 'mp4',
+            'title': 'Wiadomości, 08.12.2015, 15:00',
         },
     }, {
         'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
@@ -53,6 +64,39 @@ class TVPIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+        webpage = self._download_webpage(url, page_id)
+        video_id = self._search_regex([
+            r'<iframe[^>]+src="[^"]*?object_id=(\d+)',
+            "object_id\s*:\s*'(\d+)'"], webpage, 'video id')
+        return {
+            '_type': 'url_transparent',
+            'url': 'tvp:' + video_id,
+            'description': self._og_search_description(webpage, default=None),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'ie_key': 'TVPEmbed',
+        }
+
+
+class TVPEmbedIE(InfoExtractor):
+    IE_NAME = 'tvp:embed'
+    IE_DESC = 'Telewizja Polska'
+    _VALID_URL = r'(?:tvp:|https?://[^/]+\.tvp\.(?:pl|info)/sess/tvplayer\.php\?.*?object_id=)(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'http://www.tvp.pl/sess/tvplayer.php?object_id=22670268',
+        'md5': '8c9cd59d16edabf39331f93bf8a766c7',
+        'info_dict': {
+            'id': '22670268',
+            'ext': 'mp4',
+            'title': 'Panorama, 07.12.2015, 15:40',
+        },
+    }, {
+        'url': 'tvp:22670268',
+        'only_matching': True,
+    }]
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 

youtube_dl/extractor/tvplay.py

@@ -4,13 +4,18 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_HTTPError,
+    compat_str,
+    compat_urlparse,
+)
 from ..utils import (
+    determine_ext,
+    ExtractorError,
+    int_or_none,
     parse_iso8601,
     qualities,
-    determine_ext,
     update_url_query,
-    int_or_none,
 )
 
 
@@ -34,6 +39,9 @@ class TVPlayIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Kādi ir īri? - Viņas melo labāk',
                 'description': 'Baiba apsmej īrus, kādi tie ir un ko viņi dara.',
+                'series': 'Viņas melo labāk',
+                'season': '2.sezona',
+                'season_number': 2,
                 'duration': 25,
                 'timestamp': 1406097056,
                 'upload_date': '20140723',
@@ -46,6 +54,10 @@ class TVPlayIE(InfoExtractor):
                 'ext': 'flv',
                 'title': 'Moterys meluoja geriau',
                 'description': 'md5:9aec0fc68e2cbc992d2a140bd41fa89e',
+                'series': 'Moterys meluoja geriau',
+                'episode_number': 47,
+                'season': '1 sezonas',
+                'season_number': 1,
                 'duration': 1330,
                 'timestamp': 1403769181,
                 'upload_date': '20140626',
@@ -196,12 +208,15 @@ class TVPlayIE(InfoExtractor):
 
         title = video['title']
 
-        if video.get('is_geo_blocked'):
-            self.report_warning(
-                'This content might not be available in your country due to copyright reasons')
-
+        try:
             streams = self._download_json(
-            'http://playapi.mtgx.tv/v1/videos/stream/%s' % video_id, video_id, 'Downloading streams JSON')
+                'http://playapi.mtgx.tv/v1/videos/stream/%s' % video_id,
+                video_id, 'Downloading streams JSON')
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                msg = self._parse_json(e.cause.read().decode('utf-8'), video_id)
+                raise ExtractorError(msg['msg'], expected=True)
+            raise
 
         quality = qualities(['hls', 'medium', 'high'])
         formats = []
@@ -226,7 +241,8 @@ class TVPlayIE(InfoExtractor):
                     'ext': ext,
                 }
                 if video_url.startswith('rtmp'):
-                    m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', video_url)
+                    m = re.search(
+                        r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', video_url)
                     if not m:
                         continue
                     fmt.update({
@@ -240,15 +256,41 @@ class TVPlayIE(InfoExtractor):
                         'url': video_url,
                     })
                 formats.append(fmt)
+
+        if not formats and video.get('is_geo_blocked'):
+            self.raise_geo_restricted(
+                'This content might not be available in your country due to copyright reasons')
+
         self._sort_formats(formats)
 
+        # TODO: webvtt in m3u8
+        subtitles = {}
+        sami_path = video.get('sami_path')
+        if sami_path:
+            lang = self._search_regex(
+                r'_([a-z]{2})\.xml', sami_path, 'lang',
+                default=compat_urlparse.urlparse(url).netloc.rsplit('.', 1)[-1])
+            subtitles[lang] = [{
+                'url': sami_path,
+            }]
+
+        series = video.get('format_title')
+        episode_number = int_or_none(video.get('format_position', {}).get('episode'))
+        season = video.get('_embedded', {}).get('season', {}).get('title')
+        season_number = int_or_none(video.get('format_position', {}).get('season'))
+
         return {
             'id': video_id,
             'title': title,
             'description': video.get('description'),
+            'series': series,
+            'episode_number': episode_number,
+            'season': season,
+            'season_number': season_number,
             'duration': int_or_none(video.get('duration')),
             'timestamp': parse_iso8601(video.get('created_at')),
             'view_count': int_or_none(video.get('views', {}).get('total')),
             'age_limit': int_or_none(video.get('age_limit', 0)),
             'formats': formats,
+            'subtitles': subtitles,
         }

youtube_dl/extractor/twentyfourvideo.py

@@ -12,10 +12,9 @@ from ..utils import (
 
 class TwentyFourVideoIE(InfoExtractor):
     IE_NAME = '24video'
-    _VALID_URL = r'https?://(?:www\.)?24video\.net/(?:video/(?:view|xml)/|player/new24_play\.swf\?id=)(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?24video\.(?:net|me|xxx)/(?:video/(?:view|xml)/|player/new24_play\.swf\?id=)(?P<id>\d+)'
 
-    _TESTS = [
-        {
+    _TESTS = [{
         'url': 'http://www.24video.net/video/view/1044982',
         'md5': 'e09fc0901d9eaeedac872f154931deeb',
         'info_dict': {
@@ -32,12 +31,13 @@ class TwentyFourVideoIE(InfoExtractor):
             'like_count': int,
             'dislike_count': int,
         },
-        },
-        {
+    }, {
         'url': 'http://www.24video.net/player/new24_play.swf?id=1044982',
         'only_matching': True,
-        }
-    ]
+    }, {
+        'url': 'http://www.24video.me/video/view/1044982',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -64,7 +64,7 @@ class TwentyFourVideoIE(InfoExtractor):
             r'<span class="video-views">(\d+) просмотр',
             webpage, 'view count', fatal=False))
         comment_count = int_or_none(self._html_search_regex(
-            r'<div class="comments-title" id="comments-count">(\d+) комментари',
+            r'<a[^>]+href="#tab-comments"[^>]*>(\d+) комментари',
             webpage, 'comment count', fatal=False))
 
         # Sets some cookies

youtube_dl/extractor/uol.py

@@ -0,0 +1,128 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    int_or_none,
+    parse_duration,
+    update_url_query,
+    str_or_none,
+)
+
+
+class UOLIE(InfoExtractor):
+    IE_NAME = 'uol.com.br'
+    _VALID_URL = r'https?://(?:.+?\.)?uol\.com\.br/.*?(?:(?:mediaId|v)=|view/(?:[a-z0-9]+/)?|video(?:=|/(?:\d{4}/\d{2}/\d{2}/)?))(?P<id>\d+|[\w-]+-[A-Z0-9]+)'
+    _TESTS = [{
+        'url': 'http://player.mais.uol.com.br/player_video_v3.swf?mediaId=15951931',
+        'md5': '25291da27dc45e0afb5718a8603d3816',
+        'info_dict': {
+            'id': '15951931',
+            'ext': 'mp4',
+            'title': 'Miss simpatia é encontrada morta',
+            'description': 'md5:3f8c11a0c0556d66daf7e5b45ef823b2',
+        }
+    }, {
+        'url': 'http://tvuol.uol.com.br/video/incendio-destroi-uma-das-maiores-casas-noturnas-de-londres-04024E9A3268D4C95326',
+        'md5': 'e41a2fb7b7398a3a46b6af37b15c00c9',
+        'info_dict': {
+            'id': '15954259',
+            'ext': 'mp4',
+            'title': 'Incêndio destrói uma das maiores casas noturnas de Londres',
+            'description': 'Em Londres, um incêndio destruiu uma das maiores boates da cidade. Não há informações sobre vítimas.',
+        }
+    }, {
+        'url': 'http://mais.uol.com.br/static/uolplayer/index.html?mediaId=15951931',
+        'only_matching': True,
+    }, {
+        'url': 'http://mais.uol.com.br/view/15954259',
+        'only_matching': True,
+    }, {
+        'url': 'http://noticias.band.uol.com.br/brasilurgente/video/2016/08/05/15951931/miss-simpatia-e-encontrada-morta.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://videos.band.uol.com.br/programa.asp?e=noticias&pr=brasil-urgente&v=15951931&t=Policia-desmonte-base-do-PCC-na-Cracolandia',
+        'only_matching': True,
+    }, {
+        'url': 'http://mais.uol.com.br/view/cphaa0gl2x8r/incendio-destroi-uma-das-maiores-casas-noturnas-de-londres-04024E9A3268D4C95326',
+        'only_matching': True,
+    }, {
+        'url': 'http://noticias.uol.com.br//videos/assistir.htm?video=rafaela-silva-inspira-criancas-no-judo-04024D983968D4C95326',
+        'only_matching': True,
+    }, {
+        'url': 'http://mais.uol.com.br/view/e0qbgxid79uv/15275470',
+        'only_matching': True,
+    }]
+
+    _FORMATS = {
+        '2': {
+            'width': 640,
+            'height': 360,
+        },
+        '5': {
+            'width': 1080,
+            'height': 720,
+        },
+        '6': {
+            'width': 426,
+            'height': 240,
+        },
+        '7': {
+            'width': 1920,
+            'height': 1080,
+        },
+        '8': {
+            'width': 192,
+            'height': 144,
+        },
+        '9': {
+            'width': 568,
+            'height': 320,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        if not video_id.isdigit():
+            embed_page = self._download_webpage('https://jsuol.com.br/c/tv/uol/embed/?params=[embed,%s]' % video_id, video_id)
+            video_id = self._search_regex(r'mediaId=(\d+)', embed_page, 'media id')
+        video_data = self._download_json(
+            'http://mais.uol.com.br/apiuol/v3/player/getMedia/%s.json' % video_id,
+            video_id)['item']
+        title = video_data['title']
+
+        query = {
+            'ver': video_data.get('numRevision', 2),
+            'r': 'http://mais.uol.com.br',
+        }
+        formats = []
+        for f in video_data.get('formats', []):
+            f_url = f.get('url') or f.get('secureUrl')
+            if not f_url:
+                continue
+            format_id = str_or_none(f.get('id'))
+            fmt = {
+                'format_id': format_id,
+                'url': update_url_query(f_url, query),
+            }
+            fmt.update(self._FORMATS.get(format_id, {}))
+            formats.append(fmt)
+        self._sort_formats(formats)
+
+        tags = []
+        for tag in video_data.get('tags', []):
+            tag_description = tag.get('description')
+            if not tag_description:
+                continue
+            tags.append(tag_description)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': clean_html(video_data.get('desMedia')),
+            'thumbnail': video_data.get('thumbnail'),
+            'duration': int_or_none(video_data.get('durationSeconds')) or parse_duration(video_data.get('duration')),
+            'tags': tags,
+            'formats': formats,
+        }

youtube_dl/extractor/vbox7.py

@@ -2,17 +2,20 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
-from ..utils import (
-    ExtractorError,
-    sanitized_Request,
-    urlencode_postdata,
-)
+from ..utils import urlencode_postdata
 
 
 class Vbox7IE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?vbox7\.com/play:(?P<id>[^/]+)'
-    _TEST = {
+    _TESTS = [{
+        'url': 'http://vbox7.com/play:0946fff23c',
+        'md5': 'a60f9ab3a3a2f013ef9a967d5f7be5bf',
+        'info_dict': {
+            'id': '0946fff23c',
+            'ext': 'mp4',
+            'title': 'Борисов: Притеснен съм за бъдещето на България',
+        },
+    }, {
         'url': 'http://vbox7.com/play:249bb972c2',
         'md5': '99f65c0c9ef9b682b97313e052734c3f',
         'info_dict': {
@@ -20,43 +23,38 @@ class Vbox7IE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Смях! Чудо - чист за секунди - Скрита камера',
         },
-    }
+        'skip': 'georestricted',
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        # need to get the page 3 times for the correct jsSecretToken cookie
-        # which is necessary for the correct title
-        def get_session_id():
-            redirect_page = self._download_webpage(url, video_id)
-            session_id_url = self._search_regex(
-                r'var\s*url\s*=\s*\'([^\']+)\';', redirect_page,
-                'session id url')
-            self._download_webpage(
-                compat_urlparse.urljoin(url, session_id_url), video_id,
-                'Getting session id')
+        webpage = self._download_webpage(url, video_id)
 
-        get_session_id()
-        get_session_id()
+        title = self._html_search_regex(
+            r'<title>(.*)</title>', webpage, 'title').split('/')[0].strip()
 
-        webpage = self._download_webpage(url, video_id,
-                                         'Downloading redirect page')
+        video_url = self._search_regex(
+            r'src\s*:\s*(["\'])(?P<url>.+?.mp4.*?)\1',
+            webpage, 'video url', default=None, group='url')
 
-        title = self._html_search_regex(r'<title>(.*)</title>',
-                                        webpage, 'title').split('/')[0].strip()
+        thumbnail_url = self._og_search_thumbnail(webpage)
 
-        info_url = 'http://vbox7.com/play/magare.do'
-        data = urlencode_postdata({'as3': '1', 'vid': video_id})
-        info_request = sanitized_Request(info_url, data)
-        info_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        info_response = self._download_webpage(info_request, video_id, 'Downloading info webpage')
-        if info_response is None:
-            raise ExtractorError('Unable to extract the media url')
-        (final_url, thumbnail_url) = map(lambda x: x.split('=')[1], info_response.split('&'))
+        if not video_url:
+            info_response = self._download_webpage(
+                'http://vbox7.com/play/magare.do', video_id,
+                'Downloading info webpage',
+                data=urlencode_postdata({'as3': '1', 'vid': video_id}),
+                headers={'Content-Type': 'application/x-www-form-urlencoded'})
+            final_url, thumbnail_url = map(
+                lambda x: x.split('=')[1], info_response.split('&'))
+
+        if '/na.mp4' in video_url:
+            self.raise_geo_restricted()
 
         return {
             'id': video_id,
-            'url': final_url,
+            'url': self._proto_relative_url(video_url, 'http:'),
             'title': title,
             'thumbnail': thumbnail_url,
         }

youtube_dl/extractor/vodplatform.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import unescapeHTML
+
+
+class VODPlatformIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?vod-platform\.net/embed/(?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',
+        'md5': '1db2b7249ce383d6be96499006e951fc',
+        'info_dict': {
+            'id': 'RufMcytHDolTH1MuKHY9Fw',
+            'ext': 'mp4',
+            'title': 'LBCi News_ النصرة في ضيافة الـ "سي.أن.أن"',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = unescapeHTML(self._og_search_title(webpage))
+        hidden_inputs = self._hidden_inputs(webpage)
+
+        base_url = self._search_regex(
+            '(.*/)(?:playlist.m3u8|manifest.mpd)',
+            hidden_inputs.get('HiddenmyhHlsLink') or hidden_inputs['HiddenmyDashLink'],
+            'base url')
+        formats = self._extract_m3u8_formats(
+            base_url + 'playlist.m3u8', video_id, 'mp4',
+            'm3u8_native', m3u8_id='hls', fatal=False)
+        formats.extend(self._extract_mpd_formats(
+            base_url + 'manifest.mpd', video_id,
+            mpd_id='dash', fatal=False))
+        rtmp_formats = self._extract_smil_formats(
+            base_url + 'jwplayer.smil', video_id, fatal=False)
+        for rtmp_format in rtmp_formats:
+            rtsp_format = rtmp_format.copy()
+            rtsp_format['url'] = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
+            del rtsp_format['play_path']
+            del rtsp_format['ext']
+            rtsp_format.update({
+                'url': rtsp_format['url'].replace('rtmp://', 'rtsp://'),
+                'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
+                'protocol': 'rtsp',
+            })
+            formats.extend([rtmp_format, rtsp_format])
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': hidden_inputs.get('HiddenThumbnail') or self._og_search_thumbnail(webpage),
+            'formats': formats,
+        }

youtube_dl/extractor/wat.py

@@ -9,6 +9,7 @@ from ..utils import (
     ExtractorError,
     unified_strdate,
     HEADRequest,
+    int_or_none,
 )
 
 
@@ -30,34 +31,43 @@ class WatIE(InfoExtractor):
         },
         {
             'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html',
-            'md5': 'fbc84e4378165278e743956d9c1bf16b',
+            'md5': '34bdfa5ca9fd3c7eb88601b635b0424c',
             'info_dict': {
                 'id': '11713075',
                 'ext': 'mp4',
                 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)',
-                'description': 'md5:b7a849cf16a2b733d9cd10c52906dee3',
                 'upload_date': '20140816',
-                'duration': 2910,
             },
-            'skip': "Ce contenu n'est pas disponible pour l'instant.",
+            'expected_warnings': ["Ce contenu n'est pas disponible pour l'instant."],
         },
     ]
 
+    _FORMATS = (
+        (200, 416, 234),
+        (400, 480, 270),
+        (600, 640, 360),
+        (1200, 640, 360),
+        (1800, 960, 540),
+        (2500, 1280, 720),
+    )
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
         video_id = video_id if video_id.isdigit() and len(video_id) > 6 else compat_str(int(video_id, 36))
 
         # 'contentv4' is used in the website, but it also returns the related
         # videos, we don't need them
-        video_info = self._download_json(
-            'http://www.wat.tv/interface/contentv3/' + video_id, video_id)['media']
+        video_data = self._download_json(
+            'http://www.wat.tv/interface/contentv4s/' + video_id, video_id)
+        video_info = video_data['media']
 
         error_desc = video_info.get('error_desc')
         if error_desc:
-            raise ExtractorError(
-                '%s returned error: %s' % (self.IE_NAME, error_desc), expected=True)
+            self.report_warning(
+                '%s returned error: %s' % (self.IE_NAME, error_desc))
 
         chapters = video_info['chapters']
+        if chapters:
             first_chapter = chapters[0]
 
             def video_id_for_chapter(chapter):
@@ -69,9 +79,10 @@ class WatIE(InfoExtractor):
                 return self.playlist_result(entries, video_id, video_info['title'])
             # Otherwise we can continue and extract just one part, we have to use
             # the video id for getting the video url
+        else:
+            first_chapter = video_info
 
-        date_diffusion = first_chapter.get('date_diffusion')
-        upload_date = unified_strdate(date_diffusion) if date_diffusion else None
+        title = first_chapter['title']
 
         def extract_url(path_template, url_type):
             req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id)
@@ -83,10 +94,10 @@ class WatIE(InfoExtractor):
                     expected=True)
             return red_url
 
-        m3u8_url = extract_url('ipad/%s.m3u8', 'm3u8')
-        http_url = extract_url('android5/%s.mp4', 'http')
-
         formats = []
+        try:
+            http_url = extract_url('android5/%s.mp4', 'http')
+            m3u8_url = extract_url('ipad/%s.m3u8', 'm3u8')
             m3u8_formats = self._extract_m3u8_formats(
                 m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
             formats.extend(m3u8_formats)
@@ -97,22 +108,47 @@ class WatIE(InfoExtractor):
                 vbr, abr = m3u8_format.get('vbr'), m3u8_format.get('abr')
                 if not vbr or not abr:
                     continue
+                format_id = m3u8_format['format_id'].replace('hls', 'http')
+                fmt_url = re.sub(r'%s-\d+00-\d+' % video_id, '%s-%d00-%d' % (video_id, round(vbr / 100), round(abr)), http_url)
+                if self._is_valid_url(fmt_url, video_id, format_id):
                     f = m3u8_format.copy()
                     f.update({
-                'url': re.sub(r'%s-\d+00-\d+' % video_id, '%s-%d00-%d' % (video_id, round(vbr / 100), round(abr)), http_url),
-                'format_id': f['format_id'].replace('hls', 'http'),
+                        'url': fmt_url,
+                        'format_id': format_id,
                         'protocol': 'http',
                     })
                     formats.append(f)
             self._sort_formats(formats)
+        except ExtractorError:
+            abr = 64
+            for vbr, width, height in self._FORMATS:
+                tbr = vbr + abr
+                format_id = 'http-%s' % tbr
+                fmt_url = 'http://dnl.adv.tf1.fr/2/USP-0x0/%s/%s/%s/ssm/%s-%s-64k.mp4' % (video_id[-4:-2], video_id[-2:], video_id, video_id, vbr)
+                if self._is_valid_url(fmt_url, video_id, format_id):
+                    formats.append({
+                        'format_id': format_id,
+                        'url': fmt_url,
+                        'vbr': vbr,
+                        'abr': abr,
+                        'width': width,
+                        'height': height,
+                    })
+
+        date_diffusion = first_chapter.get('date_diffusion') or video_data.get('configv4', {}).get('estatS4')
+        upload_date = unified_strdate(date_diffusion) if date_diffusion else None
+        duration = None
+        files = video_info['files']
+        if files:
+            duration = int_or_none(files[0].get('duration'))
 
         return {
             'id': video_id,
-            'title': first_chapter['title'],
-            'thumbnail': first_chapter['preview'],
-            'description': first_chapter['description'],
-            'view_count': video_info['views'],
+            'title': title,
+            'thumbnail': first_chapter.get('preview'),
+            'description': first_chapter.get('description'),
+            'view_count': int_or_none(video_info.get('views')),
             'upload_date': upload_date,
-            'duration': video_info['files'][0]['duration'],
+            'duration': duration,
             'formats': formats,
         }

youtube_dl/extractor/yandexmusic.py

@@ -75,6 +75,12 @@ class YandexMusicTrackIE(YandexMusicBaseIE):
             % storage_dir,
             track_id, 'Downloading track location JSON')
 
+        # Each string is now wrapped in a list, this is probably only temporarily thus
+        # supporting both scenarios (see https://github.com/rg3/youtube-dl/issues/10193)
+        for k, v in data.items():
+            if v and isinstance(v, list):
+                data[k] = v[0]
+
         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
         storage = storage_dir.split('.')
 

youtube_dl/options.py

@@ -2,6 +2,7 @@ from __future__ import unicode_literals
 
 import os.path
 import optparse
+import re
 import sys
 
 from .downloader.external import list_external_downloaders
@@ -93,8 +94,18 @@ def parseOpts(overrideArguments=None):
         setattr(parser.values, option.dest, value.split(','))
 
     def _hide_login_info(opts):
-        opts = list(opts)
-        for private_opt in ['-p', '--password', '-u', '--username', '--video-password']:
+        PRIVATE_OPTS = ['-p', '--password', '-u', '--username', '--video-password']
+        eqre = re.compile('^(?P<key>' + ('|'.join(re.escape(po) for po in PRIVATE_OPTS)) + ')=.+$')
+
+        def _scrub_eq(o):
+            m = eqre.match(o)
+            if m:
+                return m.group('key') + '=PRIVATE'
+            else:
+                return o
+
+        opts = list(map(_scrub_eq, opts))
+        for private_opt in PRIVATE_OPTS:
             try:
                 i = opts.index(private_opt)
                 opts[i + 1] = 'PRIVATE'
@@ -488,9 +499,20 @@ def parseOpts(overrideArguments=None):
         dest='bidi_workaround', action='store_true',
         help='Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
     workarounds.add_option(
-        '--sleep-interval', metavar='SECONDS',
+        '--sleep-interval', '--min-sleep-interval', metavar='SECONDS',
         dest='sleep_interval', type=float,
-        help='Number of seconds to sleep before each download.')
+        help=(
+            'Number of seconds to sleep before each download when used alone '
+            'or a lower bound of a range for randomized sleep before each download '
+            '(minimum possible number of seconds to sleep) when used along with '
+            '--max-sleep-interval.'))
+    workarounds.add_option(
+        '--max-sleep-interval', metavar='SECONDS',
+        dest='max_sleep_interval', type=float,
+        help=(
+            'Upper bound of a range for randomized sleep before each download '
+            '(maximum possible number of seconds to sleep). Must only be used '
+            'along with --min-sleep-interval.'))
 
     verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options')
     verbosity.add_option(

youtube_dl/postprocessor/metadatafromtitle.py

@@ -3,11 +3,6 @@ from __future__ import unicode_literals
 import re
 
 from .common import PostProcessor
-from ..utils import PostProcessingError
-
-
-class MetadataFromTitlePPError(PostProcessingError):
-    pass
 
 
 class MetadataFromTitlePP(PostProcessor):
@@ -38,7 +33,8 @@ class MetadataFromTitlePP(PostProcessor):
         title = info['title']
         match = re.match(self._titleregex, title)
         if match is None:
-            raise MetadataFromTitlePPError('Could not interpret title of video as "%s"' % self._titleformat)
+            self._downloader.to_screen('[fromtitle] Could not interpret title of video as "%s"' % self._titleformat)
+            return [], info
         for attribute, value in match.groupdict().items():
             value = match.group(attribute)
             info[attribute] = value

youtube_dl/utils.py

@@ -47,6 +47,7 @@ from .compat import (
     compat_socket_create_connection,
     compat_str,
     compat_struct_pack,
+    compat_struct_unpack,
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_parse_urlencode,
@@ -121,6 +122,7 @@ DATE_FORMATS = (
     '%Y %m %d',
     '%Y-%m-%d',
     '%Y/%m/%d',
+    '%Y/%m/%d %H:%M',
     '%Y/%m/%d %H:%M:%S',
     '%Y-%m-%d %H:%M:%S',
     '%Y-%m-%d %H:%M:%S.%f',
@@ -1101,7 +1103,7 @@ def unified_timestamp(date_str, day_first=True):
 
     date_str = date_str.replace(',', ' ')
 
-    pm_delta = datetime.timedelta(hours=12 if re.search(r'(?i)PM', date_str) else 0)
+    pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0
     timezone, date_str = extract_timezone(date_str)
 
     # Remove AM/PM + timezone
@@ -1109,13 +1111,13 @@ def unified_timestamp(date_str, day_first=True):
 
     for expression in date_formats(day_first):
         try:
-            dt = datetime.datetime.strptime(date_str, expression) - timezone + pm_delta
+            dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)
             return calendar.timegm(dt.timetuple())
         except ValueError:
             pass
     timetuple = email.utils.parsedate_tz(date_str)
     if timetuple:
-        return calendar.timegm(timetuple.timetuple())
+        return calendar.timegm(timetuple) + pm_delta * 3600
 
 
 def determine_ext(url, default_ext='unknown_video'):
@@ -1983,11 +1985,27 @@ US_RATINGS = {
 }
 
 
+TV_PARENTAL_GUIDELINES = {
+    'TV-Y': 0,
+    'TV-Y7': 7,
+    'TV-G': 0,
+    'TV-PG': 0,
+    'TV-14': 14,
+    'TV-MA': 17,
+}
+
+
 def parse_age_limit(s):
-    if s is None:
+    if type(s) == int:
+        return s if 0 <= s <= 21 else None
+    if not isinstance(s, compat_basestring):
         return None
     m = re.match(r'^(?P<age>\d{1,2})\+?$', s)
-    return int(m.group('age')) if m else US_RATINGS.get(s)
+    if m:
+        return int(m.group('age'))
+    if s in US_RATINGS:
+        return US_RATINGS[s]
+    return TV_PARENTAL_GUIDELINES.get(s)
 
 
 def strip_jsonp(code):
@@ -2392,6 +2410,8 @@ def dfxp2srt(dfxp_data):
 
 def cli_option(params, command_option, param):
     param = params.get(param)
+    if param:
+        param = compat_str(param)
     return [command_option, param] if param is not None else []
 
 
@@ -2969,3 +2989,110 @@ def parse_m3u8_attributes(attrib):
 
 def urshift(val, n):
     return val >> n if val >= 0 else (val + 0x100000000) >> n
+
+
+# Based on png2str() written by @gdkchan and improved by @yokrysty
+# Originally posted at https://github.com/rg3/youtube-dl/issues/9706
+def decode_png(png_data):
+    # Reference: https://www.w3.org/TR/PNG/
+    header = png_data[8:]
+
+    if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR':
+        raise IOError('Not a valid PNG file.')
+
+    int_map = {1: '>B', 2: '>H', 4: '>I'}
+    unpack_integer = lambda x: compat_struct_unpack(int_map[len(x)], x)[0]
+
+    chunks = []
+
+    while header:
+        length = unpack_integer(header[:4])
+        header = header[4:]
+
+        chunk_type = header[:4]
+        header = header[4:]
+
+        chunk_data = header[:length]
+        header = header[length:]
+
+        header = header[4:]  # Skip CRC
+
+        chunks.append({
+            'type': chunk_type,
+            'length': length,
+            'data': chunk_data
+        })
+
+    ihdr = chunks[0]['data']
+
+    width = unpack_integer(ihdr[:4])
+    height = unpack_integer(ihdr[4:8])
+
+    idat = b''
+
+    for chunk in chunks:
+        if chunk['type'] == b'IDAT':
+            idat += chunk['data']
+
+    if not idat:
+        raise IOError('Unable to read PNG data.')
+
+    decompressed_data = bytearray(zlib.decompress(idat))
+
+    stride = width * 3
+    pixels = []
+
+    def _get_pixel(idx):
+        x = idx % stride
+        y = idx // stride
+        return pixels[y][x]
+
+    for y in range(height):
+        basePos = y * (1 + stride)
+        filter_type = decompressed_data[basePos]
+
+        current_row = []
+
+        pixels.append(current_row)
+
+        for x in range(stride):
+            color = decompressed_data[1 + basePos + x]
+            basex = y * stride + x
+            left = 0
+            up = 0
+
+            if x > 2:
+                left = _get_pixel(basex - 3)
+            if y > 0:
+                up = _get_pixel(basex - stride)
+
+            if filter_type == 1:  # Sub
+                color = (color + left) & 0xff
+            elif filter_type == 2:  # Up
+                color = (color + up) & 0xff
+            elif filter_type == 3:  # Average
+                color = (color + ((left + up) >> 1)) & 0xff
+            elif filter_type == 4:  # Paeth
+                a = left
+                b = up
+                c = 0
+
+                if x > 2 and y > 0:
+                    c = _get_pixel(basex - stride - 3)
+
+                p = a + b - c
+
+                pa = abs(p - a)
+                pb = abs(p - b)
+                pc = abs(p - c)
+
+                if pa <= pb and pa <= pc:
+                    color = (color + a) & 0xff
+                elif pb <= pc:
+                    color = (color + b) & 0xff
+                else:
+                    color = (color + c) & 0xff
+
+            current_row.append(color)
+
+    return width, height, pixels

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.07.30'
+__version__ = '2016.08.13'