release 2016.08.12

Now uses dailymotion

.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 *2016.07.03.1*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.08.12*. 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 Issues with outdated version will be rejected.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 **2016.07.03.1**
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.08.12**
 
 ### 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 2016.07.03.1
+[debug] youtube-dl version 2016.08.12
 [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: {}

.travis.yml

@@ -7,9 +7,6 @@ python:
   - "3.4"
   - "3.5"
 sudo: false
-install:
-  - bash ./devscripts/install_srelay.sh
-  - export PATH=$PATH:$(pwd)/tmp/srelay-0.4.8b6
 script: nosetests test --verbose
 notifications:
   email:

AUTHORS

@@ -177,3 +177,7 @@ Roman Tsiupa
 Artur Krysiak
 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

@@ -0,0 +1,369 @@
+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
+- [twitch:clips] Sort formats
+- [tv2] Use m3u8_native
+- [tv2:article] Fix video detection (#10188)
+- rtve (#10076)
+- [dailymotion:playlist] Optimize download archive processing (#10180)
+
+
+version 2016.07.28
+
+Fixed/improved extractors
+- shared (#10170)
+- soundcloud (#10179)
+- twitch (#9767)
+
+
+version 2016.07.26.2
+
+Fixed/improved extractors
+- smotri
+- camdemy
+- mtv
+- comedycentral
+- cmt
+- cbc
+- mgtv
+- orf
+
+
+version 2016.07.24
+
+New extractors
+- arkena (#8682)
+- lcp (#8682)
+
+Fixed/improved extractors
+- facebook (#10151)
+- dailymail
+- telegraaf
+- dcn
+- onet
+- tvp
+
+Miscellaneous
+- Support $Time$ in DASH manifests
+
+
+version 2016.07.22
+
+New extractors
+- odatv (#9285)
+
+Fixed/improved extractors
+- bbc
+- youjizz (#10131)
+- youtube (#10140)
+- pornhub (#10138)
+- eporner (#10139)
+
+
+version 2016.07.17
+
+New extractors
+- nintendo (#9986)
+- streamable (#9122)
+
+Fixed/improved extractors
+- ard (#10095)
+- mtv
+- comedycentral (#10101)
+- viki (#10098)
+- spike (#10106)
+
+Miscellaneous
+- Improved twitter player detection (#10090)
+
+
+version 2016.07.16
+
+New extractors
+- ninenow (#5181)
+
+Fixed/improved extractors
+- rtve (#10076)
+- brightcove
+- 3qsdn
+- syfy (#9087, #3820, #2388)
+- youtube (#10083)
+
+Miscellaneous
+- Fix subtitle embedding for video-only and audio-only files (#10081)
+
+
+version 2016.07.13
+
+New extractors
+- rudo
+
+Fixed/improved extractors
+- biobiochiletv
+- tvplay
+- dbtv
+- brightcove
+- tmz
+- youtube (#10059)
+- shahid (#10062)
+- vk
+- ellentv (#10067)
+
+
+version 2016.07.11
+
+New Extractors
+- roosterteeth (#9864)
+
+Fixed/improved extractors
+- miomio (#9605)
+- vuclip
+- youtube
+- vidzi (#10058)
+
+
+version 2016.07.09.2
+
+Fixed/improved extractors
+- vimeo (#1638)
+- facebook (#10048)
+- lynda (#10047)
+- animeondemand
+
+Fixed/improved features
+- Embedding subtitles no longer throws an error with problematic inputs (#9063)
+
+
+version 2016.07.09.1
+
+Fixed/improved extractors
+- youtube
+- ard
+- srmediatek (#9373)
+
+
+version 2016.07.09
+
+New extractors
+- Flipagram (#9898)
+
+Fixed/improved extractors
+- telecinco
+- toutv
+- radiocanada
+- tweakers (#9516)
+- lynda
+- nick (#7542)
+- polskieradio (#10028)
+- le
+- facebook (#9851)
+- mgtv
+- animeondemand (#10031)
+
+Fixed/improved features
+- `--postprocessor-args` and `--downloader-args` now accepts non-ASCII inputs
+  on non-Windows systems
+
+
+version 2016.07.07
+
+New extractors
+- kamcord (#10001)
+
+Fixed/improved extractors
+- spiegel (#10018)
+- metacafe (#8539, #3253)
+- onet (#9950)
+- francetv (#9955)
+- brightcove (#9965)
+- daum (#9972)
+
+
+version 2016.07.06
+
+Fixed/improved extractors
+- youtube (#10007, #10009)
+- xuite
+- stitcher
+- spiegel
+- slideshare
+- sandia
+- rtvnh
+- prosiebensat1
+- onionstudios
+
+
+version 2016.07.05
+
+Fixed/improved extractors
+- brightcove
+- yahoo (#9995)
+- pornhub (#9997)
+- iqiyi
+- kaltura (#5557)
+- la7
+- Changed features
+- Rename --cn-verfication-proxy to --geo-verification-proxy
+Miscellaneous
+- Add script for displaying downloads statistics
+
+
+version 2016.07.03.1
+
+Fixed/improved extractors
+- theplatform
+- aenetworks
+- nationalgeographic
+- hrti (#9482)
+- facebook (#5701)
+- buzzfeed (#5701)
+- rai (#8617, #9157, #9232, #8552, #8551)
+- nationalgeographic (#9991)
+- iqiyi
+
+
+version 2016.07.03
+
+New extractors
+- hrti (#9482)
+
+Fixed/improved extractors
+- vk (#9981)
+- facebook (#9938)
+- xtube (#9953, #9961)
+
+
+version 2016.07.02
+
+New extractors
+- fusion (#9958)
+
+Fixed/improved extractors
+- twitch (#9975)
+- vine (#9970)
+- periscope (#9967)
+- pornhub (#8696)
+
+
+version 2016.07.01
+
+New extractors
+- 9c9media
+- ctvnews (#2156)
+- ctv (#4077)
+
+Fixed/Improved extractors
+- rds
+- meta (#8789)
+- pornhub (#9964)
+- sixplay (#2183)
+
+New features
+- Accept quoted strings across multiple lines (#9940)

Makefile

@@ -94,7 +94,7 @@ _EXTRACTOR_FILES != find youtube_dl/extractor -iname '*.py' -and -not -iname 'la
 youtube_dl/extractor/lazy_extractors.py: devscripts/make_lazy_extractors.py devscripts/lazy_load_template.py $(_EXTRACTOR_FILES)
 	$(PYTHON) devscripts/make_lazy_extractors.py $@
 
-youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish
+youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish ChangeLog
 	@tar -czf youtube-dl.tar.gz --transform "s|^|youtube-dl/|" --owner 0 --group 0 \
 		--exclude '*.DS_Store' \
 		--exclude '*.kate-swp' \
@@ -107,7 +107,7 @@ youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-
 		--exclude 'docs/_build' \
 		-- \
 		bin devscripts test youtube_dl docs \
-		LICENSE README.md README.txt \
+		ChangeLog LICENSE README.md README.txt \
 		Makefile MANIFEST.in youtube-dl.1 youtube-dl.bash-completion \
 		youtube-dl.zsh youtube-dl.fish setup.py \
 		youtube-dl

README.md

@@ -17,7 +17,7 @@ youtube-dl - download videos from youtube.com or other video platforms
 
 To install it right away for all UNIX users (Linux, OS X, etc.), type:
 
-    sudo curl -L https://yt-dl.org/latest/youtube-dl -o /usr/local/bin/youtube-dl
+    sudo curl -L https://yt-dl.org/downloads/latest/youtube-dl -o /usr/local/bin/youtube-dl
     sudo chmod a+rx /usr/local/bin/youtube-dl
 
 If you do not have curl, you can alternatively use a recent wget:
@@ -103,9 +103,9 @@ which means you can modify it, redistribute it or use it however you like.
                                      (experimental)
     -6, --force-ipv6                 Make all connections via IPv6
                                      (experimental)
-    --cn-verification-proxy URL      Use this proxy to verify the IP address for
+    --geo-verification-proxy URL     Use this proxy to verify the IP address for
-                                     some Chinese sites. The default proxy
+                                     some geo-restricted sites. The default
-                                     specified by --proxy (or none, if the
+                                     proxy specified by --proxy (or none, if the
                                      options is not present) is used for the
                                      actual downloading. (experimental)
 
@@ -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
@@ -424,7 +432,7 @@ which means you can modify it, redistribute it or use it however you like.
 
 # CONFIGURATION
 
-You can configure youtube-dl by placing any supported command line option to a configuration file. On Linux and OS X, the system wide configuration file is located at `/etc/youtube-dl.conf` and the user wide configuration file at `~/.config/youtube-dl/config`. On Windows, the user wide configuration file locations are `%APPDATA%\youtube-dl\config.txt` or `C:\Users\<user name>\youtube-dl.conf`.
+You can configure youtube-dl by placing any supported command line option to a configuration file. On Linux and OS X, the system wide configuration file is located at `/etc/youtube-dl.conf` and the user wide configuration file at `~/.config/youtube-dl/config`. On Windows, the user wide configuration file locations are `%APPDATA%\youtube-dl\config.txt` or `C:\Users\<user name>\youtube-dl.conf`. Note that by default configuration file may not exist so you may need to create it yourself.
 
 For example, with the following configuration file youtube-dl will always extract the audio, not copy the mtime, use a proxy and save all videos under `Movies` directory in your home directory:
 ```
@@ -432,6 +440,7 @@ For example, with the following configuration file youtube-dl will always extrac
 --no-mtime
 --proxy 127.0.0.1:3128
 -o ~/Movies/%(title)s.%(ext)s
+# Lines starting with # are comments
 ```
 
 Note that options in configuration file are just the same options aka switches used in regular command line calls thus there **must be no whitespace** after `-` or `--`, e.g. `-o` or `--proxy` but not `- o` or `-- proxy`.
@@ -1195,7 +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/gh-pages/generate-download.py

@@ -15,13 +15,9 @@ data = urllib.request.urlopen(URL).read()
 with open('download.html.in', 'r', encoding='utf-8') as tmplf:
     template = tmplf.read()
 
-md5sum = hashlib.md5(data).hexdigest()
-sha1sum = hashlib.sha1(data).hexdigest()
 sha256sum = hashlib.sha256(data).hexdigest()
 template = template.replace('@PROGRAM_VERSION@', version)
 template = template.replace('@PROGRAM_URL@', URL)
-template = template.replace('@PROGRAM_MD5SUM@', md5sum)
-template = template.replace('@PROGRAM_SHA1SUM@', sha1sum)
 template = template.replace('@PROGRAM_SHA256SUM@', sha256sum)
 template = template.replace('@EXE_URL@', versions_info['versions'][version]['exe'][0])
 template = template.replace('@EXE_SHA256SUM@', versions_info['versions'][version]['exe'][1])

devscripts/install_srelay.sh

@@ -1,8 +0,0 @@
-#!/bin/bash
-
-mkdir -p tmp && cd tmp
-wget -N http://downloads.sourceforge.net/project/socks-relay/socks-relay/srelay-0.4.8/srelay-0.4.8b6.tar.gz
-tar zxvf srelay-0.4.8b6.tar.gz
-cd srelay-0.4.8b6
-./configure
-make

devscripts/prepare_manpage.py

@@ -54,17 +54,21 @@ 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())
-                split_option = option.split(' ')
+                # 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
+                    if not split_option[-1].startswith('-'):  # metavar
-                    option = ' '.join(split_option[:-1] + ['*%s*' % split_option[-1]])
+                        option = ' '.join(split_option[:-1] + ['*%s*' % split_option[-1]])
 
-                # Pandoc's definition_lists. See http://pandoc.org/README.html
+                    # Pandoc's definition_lists. See http://pandoc.org/README.html
-                # for more information.
+                    # for more information.
-                ret += '\n%s\n:   %s\n' % (option, description)
+                    ret += '\n%s\n:   %s\n' % (option, description)
-            else:
+                    continue
-                ret += line.lstrip() + '\n'
+            ret += line.lstrip() + '\n'
         else:
             ret += line + '\n'
 

devscripts/release.sh

@@ -71,9 +71,12 @@ fi
 /bin/echo -e "\n### Changing version in version.py..."
 sed -i "s/__version__ = '.*'/__version__ = '$version'/" youtube_dl/version.py
 
+/bin/echo -e "\n### Changing version in ChangeLog..."
+sed -i "s/<unreleased>/$version/" ChangeLog
+
 /bin/echo -e "\n### Committing documentation, templates and youtube_dl/version.py..."
 make README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE.md supportedsites
-git add README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE.md docs/supportedsites.md youtube_dl/version.py
+git add README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE.md docs/supportedsites.md youtube_dl/version.py ChangeLog
 git commit $gpg_sign_commits -m "release $version"
 
 /bin/echo -e "\n### Now tagging, signing and pushing..."

devscripts/show-downloads-statistics.py

@@ -0,0 +1,47 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+import itertools
+import json
+import os
+import re
+import sys
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.compat import (
+    compat_print,
+    compat_urllib_request,
+)
+from youtube_dl.utils import format_bytes
+
+
+def format_size(bytes):
+    return '%s (%d bytes)' % (format_bytes(bytes), bytes)
+
+
+total_bytes = 0
+
+for page in itertools.count(1):
+    releases = json.loads(compat_urllib_request.urlopen(
+        'https://api.github.com/repos/rg3/youtube-dl/releases?page=%s' % page
+    ).read().decode('utf-8'))
+
+    if not releases:
+        break
+
+    for release in releases:
+        compat_print(release['name'])
+        for asset in release['assets']:
+            asset_name = asset['name']
+            total_bytes += asset['download_count'] * asset['size']
+            if all(not re.match(p, asset_name) for p in (
+                    r'^youtube-dl$',
+                    r'^youtube-dl-\d{4}\.\d{2}\.\d{2}(?:\.\d+)?\.tar\.gz$',
+                    r'^youtube-dl\.exe$')):
+                continue
+            compat_print(
+                ' %s size: %s downloads: %d'
+                % (asset_name, format_size(asset['size']), asset['download_count']))
+
+compat_print('total downloads traffic: %s' % format_size(total_bytes))

docs/supportedsites.md

@@ -14,6 +14,7 @@
  - **8tracks**
  - **91porn**
  - **9gag**
+ - **9now.com.au**
  - **abc.net.au**
  - **Abc7News**
  - **abcnews**
@@ -45,6 +46,7 @@
  - **archive.org**: archive.org videos
  - **ARD**
  - **ARD:mediathek**
+ - **Arkena**
  - **arte.tv**
  - **arte.tv:+7**
  - **arte.tv:cinema**
@@ -140,7 +142,8 @@
  - **CollegeRama**
  - **ComCarCoff**
  - **ComedyCentral**
- - **ComedyCentralShows**: The Daily Show / The Colbert Report
+ - **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**
  - **Cracked**
@@ -179,6 +182,7 @@
  - **DigitallySpeaking**
  - **Digiteka**
  - **Discovery**
+ - **DiscoveryGo**
  - **Dotsub**
  - **DouyuTV**: 斗鱼
  - **DPlay**
@@ -224,6 +228,7 @@
  - **Firstpost**
  - **FiveTV**
  - **Flickr**
+ - **Flipagram**
  - **Folketinget**: Folketinget (ft.dk; Danish parliament)
  - **FootyRoom**
  - **Formula1**
@@ -244,7 +249,6 @@
  - **FunnyOrDie**
  - **Fusion**
  - **GameInformer**
- - **Gamekings**
  - **GameOne**
  - **gameone:playlist**
  - **Gamersyde**
@@ -261,7 +265,6 @@
  - **GloboArticle**
  - **GodTube**
  - **GodTV**
- - **GoldenMoustache**
  - **Golem**
  - **GoogleDrive**
  - **Goshgay**
@@ -312,6 +315,7 @@
  - **jpopsuki.tv**
  - **JWPlatform**
  - **Kaltura**
+ - **Kamcord**
  - **KanalPlay**: Kanal 5/9/11 Play
  - **Kankan**
  - **Karaoketv**
@@ -333,6 +337,8 @@
  - **kuwo:song**: 酷我音乐
  - **la7.it**
  - **Laola1Tv**
+ - **Lcp**
+ - **LcpPlay**
  - **Le**: 乐视网
  - **Learnr**
  - **Lecture2Go**
@@ -394,7 +400,6 @@
  - **MSN**
  - **MTV**
  - **mtv.de**
- - **mtviggy.com**
  - **mtvservices:embedded**
  - **MuenchenTV**: münchen.tv
  - **MusicPlayOn**
@@ -410,7 +415,8 @@
  - **MyVidster**
  - **n-tv.de**
  - **natgeo**
- - **natgeo:channel**
+ - **natgeo:episodeguide**
+ - **natgeo:video**
  - **Naver**
  - **NBA**
  - **NBC**
@@ -434,7 +440,6 @@
  - **Newstube**
  - **NextMedia**: 蘋果日報
  - **NextMediaActionNews**: 蘋果日報 - 動新聞
- - **nextmovie.com**
  - **nfb**: National Film Board of Canada
  - **nfl.com**
  - **nhl.com**
@@ -446,6 +451,7 @@
  - **niconico**: ニコニコ動画
  - **NiconicoPlaylist**
  - **NineCNineMedia**
+ - **Nintendo**
  - **njoy**: N-JOY
  - **njoy:embed**
  - **Noco**
@@ -473,9 +479,12 @@
  - **NYTimes**
  - **NYTimesArticle**
  - **ocw.mit.edu**
+ - **OdaTV**
  - **Odnoklassniki**
  - **OktoberfestTV**
  - **on.aol.com**
+ - **onet.tv**
+ - **onet.tv:channel**
  - **OnionStudios**
  - **Ooyala**
  - **OoyalaExternal**
@@ -509,6 +518,7 @@
  - **plus.google**: Google Plus
  - **pluzz.francetv.fr**
  - **podomatic**
+ - **Pokemon**
  - **PolskieRadio**
  - **PornHd**
  - **PornHub**: PornHub and Thumbzilla
@@ -550,8 +560,10 @@
  - **RICE**
  - **RingTV**
  - **RockstarGames**
+ - **RoosterTeeth**
  - **RottenTomatoes**
  - **Roxwel**
+ - **Rozhlas**
  - **RTBF**
  - **rte**: Raidió Teilifís Éireann TV
  - **rte:radio**: Raidió Teilifís Éireann radio
@@ -562,7 +574,9 @@
  - **rtve.es:alacarta**: RTVE a la carta
  - **rtve.es:infantil**: RTVE infantil
  - **rtve.es:live**: RTVE.es live streams
+ - **rtve.es:television**
  - **RTVNH**
+ - **Rudo**
  - **RUHD**
  - **RulePorn**
  - **rutube**: Rutube videos
@@ -607,6 +621,7 @@
  - **smotri:user**: Smotri.com user videos
  - **Snotr**
  - **Sohu**
+ - **SonyLIV**
  - **soundcloud**
  - **soundcloud:playlist**
  - **soundcloud:search**: Soundcloud search
@@ -637,6 +652,7 @@
  - **stanfordoc**: Stanford Open ClassRoom
  - **Steam**
  - **Stitcher**
+ - **Streamable**
  - **streamcloud.eu**
  - **StreamCZ**
  - **StreetVoice**
@@ -684,6 +700,7 @@
  - **TNAFlix**
  - **TNAFlixNetworkEmbed**
  - **toggle**
+ - **Tosh**: Tosh.0
  - **tou.tv**
  - **Toypics**: Toypics user profile
  - **ToypicsUser**: Toypics user profile
@@ -713,6 +730,7 @@
  - **tvigle**: Интернет-телевидение Tvigle.ru
  - **tvland.com**
  - **tvp**: Telewizja Polska
+ - **tvp:embed**: Telewizja Polska
  - **tvp:series**
  - **TVPlay**: TV3Play and related services
  - **Tweakers**
@@ -730,6 +748,7 @@
  - **udemy:course**
  - **UDNEmbed**: 聯合影音
  - **Unistra**
+ - **uol.com.br**
  - **Urort**: NRK P3 Urørt
  - **URPlay**
  - **USAToday**
@@ -789,8 +808,10 @@
  - **vine:user**
  - **vk**: VK
  - **vk:uservideos**: VK - User's Videos
+ - **vk:wallpost**
  - **vlive**
  - **Vodlocker**
+ - **VODPlatform**
  - **VoiceRepublic**
  - **VoxMedia**
  - **Vporn**
@@ -857,6 +878,7 @@
  - **youtube:search**: YouTube.com searches
  - **youtube:search:date**: YouTube.com searches, newest videos first
  - **youtube:search_url**: YouTube.com search URLs
+ - **youtube:shared**
  - **youtube:show**: YouTube.com (multi-season) shows
  - **youtube:subscriptions**: YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)
  - **youtube:user**: YouTube.com user videos (URL or "ytuser" keyword)

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_YoutubeDL.py

@@ -335,6 +335,40 @@ class TestFormatSelection(unittest.TestCase):
             downloaded = ydl.downloaded_info_dicts[0]
             self.assertEqual(downloaded['format_id'], f1['format_id'])
 
+    def test_audio_only_extractor_format_selection(self):
+        # For extractors with incomplete formats (all formats are audio-only or
+        # video-only) best and worst should fallback to corresponding best/worst
+        # video-only or audio-only formats (as per
+        # https://github.com/rg3/youtube-dl/pull/5556)
+        formats = [
+            {'format_id': 'low', 'ext': 'mp3', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
+            {'format_id': 'high', 'ext': 'mp3', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
+        ]
+        info_dict = _make_result(formats)
+
+        ydl = YDL({'format': 'best'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'high')
+
+        ydl = YDL({'format': 'worst'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'low')
+
+    def test_format_not_available(self):
+        formats = [
+            {'format_id': 'regular', 'ext': 'mp4', 'height': 360, 'url': TEST_URL},
+            {'format_id': 'video', 'ext': 'mp4', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
+        ]
+        info_dict = _make_result(formats)
+
+        # This must fail since complete video-audio format does not match filter
+        # and extractor does not provide incomplete only formats (i.e. only
+        # video-only or audio-only).
+        ydl = YDL({'format': 'best[height>360]'})
+        self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
+
     def test_invalid_format_specs(self):
         def assert_syntax_error(format_spec):
             ydl = YDL({'format': format_spec})

test/test_all_urls.py

@@ -101,8 +101,6 @@ class TestAllURLsMatching(unittest.TestCase):
         self.assertMatch(':ytsubs', ['youtube:subscriptions'])
         self.assertMatch(':ytsubscriptions', ['youtube:subscriptions'])
         self.assertMatch(':ythistory', ['youtube:history'])
-        self.assertMatch(':thedailyshow', ['ComedyCentralShows'])
-        self.assertMatch(':tds', ['ComedyCentralShows'])
 
     def test_vimeo_matching(self):
         self.assertMatch('https://vimeo.com/channels/tributes', ['vimeo:channel'])

test/test_compat.py

@@ -88,6 +88,7 @@ class TestCompat(unittest.TestCase):
     def test_compat_shlex_split(self):
         self.assertEqual(compat_shlex_split('-option "one two"'), ['-option', 'one two'])
         self.assertEqual(compat_shlex_split('-option "one\ntwo" \n -flag'), ['-option', 'one\ntwo', '-flag'])
+        self.assertEqual(compat_shlex_split('-val 中文'), ['-val', '中文'])
 
     def test_compat_etree_fromstring(self):
         xml = '''

test/test_http.py

@@ -138,27 +138,27 @@ class TestProxy(unittest.TestCase):
         self.proxy_thread.daemon = True
         self.proxy_thread.start()
 
-        self.cn_proxy = compat_http_server.HTTPServer(
+        self.geo_proxy = compat_http_server.HTTPServer(
-            ('localhost', 0), _build_proxy_handler('cn'))
+            ('localhost', 0), _build_proxy_handler('geo'))
-        self.cn_port = http_server_port(self.cn_proxy)
+        self.geo_port = http_server_port(self.geo_proxy)
-        self.cn_proxy_thread = threading.Thread(target=self.cn_proxy.serve_forever)
+        self.geo_proxy_thread = threading.Thread(target=self.geo_proxy.serve_forever)
-        self.cn_proxy_thread.daemon = True
+        self.geo_proxy_thread.daemon = True
-        self.cn_proxy_thread.start()
+        self.geo_proxy_thread.start()
 
     def test_proxy(self):
-        cn_proxy = 'localhost:{0}'.format(self.cn_port)
+        geo_proxy = 'localhost:{0}'.format(self.geo_port)
         ydl = YoutubeDL({
             'proxy': 'localhost:{0}'.format(self.port),
-            'cn_verification_proxy': cn_proxy,
+            'geo_verification_proxy': geo_proxy,
         })
         url = 'http://foo.com/bar'
         response = ydl.urlopen(url).read().decode('utf-8')
         self.assertEqual(response, 'normal: {0}'.format(url))
 
         req = compat_urllib_request.Request(url)
-        req.add_header('Ytdl-request-proxy', cn_proxy)
+        req.add_header('Ytdl-request-proxy', geo_proxy)
         response = ydl.urlopen(req).read().decode('utf-8')
-        self.assertEqual(response, 'cn: {0}'.format(url))
+        self.assertEqual(response, 'geo: {0}'.format(url))
 
     def test_proxy_with_idn(self):
         ydl = YoutubeDL({

test/test_utils.py

@@ -33,6 +33,7 @@ from youtube_dl.utils import (
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
+    get_element_by_class,
     InAdvancePagedList,
     intlist_to_bytes,
     is_html,
@@ -41,6 +42,7 @@ from youtube_dl.utils import (
     ohdave_rsa_encrypt,
     OnDemandPagedList,
     orderedSet,
+    parse_age_limit,
     parse_duration,
     parse_filesize,
     parse_count,
@@ -80,6 +82,7 @@ from youtube_dl.utils import (
     cli_option,
     cli_valueless_option,
     cli_bool_option,
+    parse_codecs,
 )
 from youtube_dl.compat import (
     compat_chr,
@@ -306,6 +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')
@@ -405,6 +409,12 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(res_url, url)
         self.assertEqual(res_data, None)
 
+        smug_url = smuggle_url(url, {'a': 'b'})
+        smug_smug_url = smuggle_url(smug_url, {'c': 'd'})
+        res_url, res_data = unsmuggle_url(smug_smug_url)
+        self.assertEqual(res_url, url)
+        self.assertEqual(res_data, {'a': 'b', 'c': 'd'})
+
     def test_shell_quote(self):
         args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')]
         self.assertEqual(shell_quote(args), """ffmpeg -i 'ñ€ß'"'"'.mp4'""")
@@ -423,6 +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)
@@ -601,6 +625,29 @@ class TestUtil(unittest.TestCase):
             limit_length('foo bar baz asd', 12).startswith('foo bar'))
         self.assertTrue('...' in limit_length('foo bar baz asd', 12))
 
+    def test_parse_codecs(self):
+        self.assertEqual(parse_codecs(''), {})
+        self.assertEqual(parse_codecs('avc1.77.30, mp4a.40.2'), {
+            'vcodec': 'avc1.77.30',
+            'acodec': 'mp4a.40.2',
+        })
+        self.assertEqual(parse_codecs('mp4a.40.2'), {
+            'vcodec': 'none',
+            'acodec': 'mp4a.40.2',
+        })
+        self.assertEqual(parse_codecs('mp4a.40.5,avc1.42001e'), {
+            'vcodec': 'avc1.42001e',
+            'acodec': 'mp4a.40.5',
+        })
+        self.assertEqual(parse_codecs('avc3.640028'), {
+            'vcodec': 'avc3.640028',
+            'acodec': 'none',
+        })
+        self.assertEqual(parse_codecs(', h264,,newcodec,aac'), {
+            'vcodec': 'h264',
+            'acodec': 'aac',
+        })
+
     def test_escape_rfc3986(self):
         reserved = "!*'();:@&=+$,/?#[]"
         unreserved = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_.~'
@@ -985,5 +1032,13 @@ The first line
         self.assertEqual(urshift(3, 1), 1)
         self.assertEqual(urshift(-3, 1), 2147483646)
 
+    def test_get_element_by_class(self):
+        html = '''
+            <span class="foo bar">nice</span>
+        '''
+
+        self.assertEqual(get_element_by_class('foo', html), 'nice')
+        self.assertEqual(get_element_by_class('no-such-class', html), None)
+
 if __name__ == '__main__':
     unittest.main()

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('--username' in serr)
+        self.assertTrue('johnsmith' not in serr)
+        self.assertTrue('--password' in serr)
+        self.assertTrue('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('-u' in serr)
+        self.assertTrue('johnsmith' not in serr)
+        self.assertTrue('-p' in serr)
+        self.assertTrue('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('--username' in serr)
+        self.assertTrue('johnsmith' not in serr)
+        self.assertTrue('--password' in serr)
+        self.assertTrue('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('-u' in serr)
+        self.assertTrue('johnsmith' not in serr)
+        self.assertTrue('-p' in serr)
+        self.assertTrue('secret' not in serr)
+
+if __name__ == '__main__':
+    unittest.main()

youtube_dl/YoutubeDL.py

@@ -5,6 +5,7 @@ from __future__ import absolute_import, unicode_literals
 
 import collections
 import contextlib
+import copy
 import datetime
 import errno
 import fileinput
@@ -196,8 +197,8 @@ class YoutubeDL(object):
     prefer_insecure:   Use HTTP instead of HTTPS to retrieve information.
                        At the moment, this is only supported by YouTube.
     proxy:             URL of the proxy server to use
-    cn_verification_proxy:  URL of the proxy to use for IP address verification
+    geo_verification_proxy:  URL of the proxy to use for IP address verification
-                       on Chinese sites. (Experimental)
+                       on geo-restricted sites. (Experimental)
     socket_timeout:    Time to wait for unresponsive hosts, in seconds
     bidi_workaround:   Work around buggy terminals without bidirectional text
                        support, using fridibi
@@ -248,7 +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
@@ -304,6 +314,11 @@ class YoutubeDL(object):
         self.params.update(params)
         self.cache = Cache(self)
 
+        if self.params.get('cn_verification_proxy') is not None:
+            self.report_warning('--cn-verification-proxy is deprecated. Use --geo-verification-proxy instead.')
+            if self.params.get('geo_verification_proxy') is None:
+                self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
+
         if params.get('bidi_workaround', False):
             try:
                 import pty
@@ -1046,9 +1061,9 @@ class YoutubeDL(object):
             if isinstance(selector, list):
                 fs = [_build_selector_function(s) for s in selector]
 
-                def selector_function(formats):
+                def selector_function(ctx):
                     for f in fs:
-                        for format in f(formats):
+                        for format in f(ctx):
                             yield format
                 return selector_function
             elif selector.type == GROUP:
@@ -1056,17 +1071,17 @@ class YoutubeDL(object):
             elif selector.type == PICKFIRST:
                 fs = [_build_selector_function(s) for s in selector.selector]
 
-                def selector_function(formats):
+                def selector_function(ctx):
                     for f in fs:
-                        picked_formats = list(f(formats))
+                        picked_formats = list(f(ctx))
                         if picked_formats:
                             return picked_formats
                     return []
             elif selector.type == SINGLE:
                 format_spec = selector.selector
 
-                def selector_function(formats):
+                def selector_function(ctx):
-                    formats = list(formats)
+                    formats = list(ctx['formats'])
                     if not formats:
                         return
                     if format_spec == 'all':
@@ -1079,9 +1094,10 @@ class YoutubeDL(object):
                             if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
                         if audiovideo_formats:
                             yield audiovideo_formats[format_idx]
-                        # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format
+                        # for extractors with incomplete formats (audio only (soundcloud)
-                        elif (all(f.get('acodec') != 'none' for f in formats) or
+                        # or video only (imgur)) we will fallback to best/worst
-                              all(f.get('vcodec') != 'none' for f in formats)):
+                        # {video,audio}-only format
+                        elif ctx['incomplete_formats']:
                             yield formats[format_idx]
                     elif format_spec == 'bestaudio':
                         audio_formats = [
@@ -1155,17 +1171,18 @@ class YoutubeDL(object):
                     }
                 video_selector, audio_selector = map(_build_selector_function, selector.selector)
 
-                def selector_function(formats):
+                def selector_function(ctx):
-                    formats = list(formats)
+                    for pair in itertools.product(
-                    for pair in itertools.product(video_selector(formats), audio_selector(formats)):
+                            video_selector(copy.deepcopy(ctx)), audio_selector(copy.deepcopy(ctx))):
                         yield _merge(pair)
 
             filters = [self._build_format_filter(f) for f in selector.filters]
 
-            def final_selector(formats):
+            def final_selector(ctx):
+                ctx_copy = copy.deepcopy(ctx)
                 for _filter in filters:
-                    formats = list(filter(_filter, formats))
+                    ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
-                return selector_function(formats)
+                return selector_function(ctx_copy)
             return final_selector
 
         stream = io.BytesIO(format_spec.encode('utf-8'))
@@ -1372,7 +1389,34 @@ class YoutubeDL(object):
             req_format_list.append('best')
             req_format = '/'.join(req_format_list)
         format_selector = self.build_format_selector(req_format)
-        formats_to_download = list(format_selector(formats))
+
+        # While in format selection we may need to have an access to the original
+        # format set in order to calculate some metrics or do some processing.
+        # For now we need to be able to guess whether original formats provided
+        # by extractor are incomplete or not (i.e. whether extractor provides only
+        # video-only or audio-only formats) for proper formats selection for
+        # extractors with such incomplete formats (see
+        # https://github.com/rg3/youtube-dl/pull/5556).
+        # Since formats may be filtered during format selection and may not match
+        # the original formats the results may be incorrect. Thus original formats
+        # or pre-calculated metrics should be passed to format selection routines
+        # as well.
+        # We will pass a context object containing all necessary additional data
+        # instead of just formats.
+        # This fixes incorrect format selection issue (see
+        # https://github.com/rg3/youtube-dl/issues/10083).
+        incomplete_formats = (
+            # All formats are video-only or
+            all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats) or
+            # all formats are audio-only
+            all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
+
+        ctx = {
+            'formats': formats,
+            'incomplete_formats': incomplete_formats,
+        }
+
+        formats_to_download = list(format_selector(ctx))
         if not formats_to_download:
             raise ExtractorError('requested format not available',
                                  expected=True)
@@ -1559,7 +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,
@@ -382,6 +393,8 @@ def _real_main(argv=None):
         'external_downloader_args': external_downloader_args,
         'postprocessor_args': postprocessor_args,
         'cn_verification_proxy': opts.cn_verification_proxy,
+        'geo_verification_proxy': opts.geo_verification_proxy,
+
     }
 
     with YoutubeDL(ydl_opts) as ydl:

youtube_dl/compat.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import binascii
@@ -2594,15 +2595,19 @@ except ImportError:  # Python < 3.3
             return "'" + s.replace("'", "'\"'\"'") + "'"
 
 
-if sys.version_info >= (2, 7, 3):
+try:
+    args = shlex.split('中文')
+    assert (isinstance(args, list) and
+            isinstance(args[0], compat_str) and
+            args[0] == '中文')
     compat_shlex_split = shlex.split
-else:
+except (AssertionError, UnicodeEncodeError):
     # Working around shlex issue with unicode strings on some python 2
     # versions (see http://bugs.python.org/issue1548891)
     def compat_shlex_split(s, comments=False, posix=True):
         if isinstance(s, compat_str):
             s = s.encode('utf-8')
-        return shlex.split(s, comments, posix)
+        return list(map(lambda s: s.decode('utf-8'), shlex.split(s, comments, posix)))
 
 
 def compat_ord(c):

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')
+        min_sleep_interval = self.params.get('sleep_interval')
-        if 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/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:
-                    raise ExtractorError('Unable to find 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 stream and stream.get('videoPlaybackID'):
-            if not clips:
+                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/amp.py

@@ -5,6 +5,8 @@ from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     parse_iso8601,
+    mimetype2ext,
+    determine_ext,
 )
 
 
@@ -50,21 +52,25 @@ class AMPIE(InfoExtractor):
         if isinstance(media_content, dict):
             media_content = [media_content]
         for media_data in media_content:
-            media = media_data['@attributes']
+            media = media_data.get('@attributes', {})
-            media_type = media['type']
+            media_url = media.get('url')
-            if media_type in ('video/f4m', 'application/f4m+xml'):
+            if not media_url:
+                continue
+            ext = mimetype2ext(media.get('type')) or determine_ext(media_url)
+            if ext == 'f4m':
                 formats.extend(self._extract_f4m_formats(
-                    media['url'] + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
+                    media_url + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
                     video_id, f4m_id='hds', fatal=False))
-            elif media_type == 'application/x-mpegURL':
+            elif ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    media['url'], video_id, 'mp4', m3u8_id='hls', fatal=False))
+                    media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
             else:
                 formats.append({
                     'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'),
                     'url': media['url'],
                     'tbr': int_or_none(media.get('bitrate')),
                     'filesize': int_or_none(media.get('fileSize')),
+                    'ext': ext,
                 })
 
         self._sort_formats(formats)

youtube_dl/extractor/animeondemand.py

@@ -22,6 +22,7 @@ class AnimeOnDemandIE(InfoExtractor):
     _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
     _NETRC_MACHINE = 'animeondemand'
     _TESTS = [{
+        # jap, OmU
         'url': 'https://www.anime-on-demand.de/anime/161',
         'info_dict': {
             'id': '161',
@@ -30,17 +31,21 @@ class AnimeOnDemandIE(InfoExtractor):
         },
         'playlist_mincount': 4,
     }, {
-        # Film wording is used instead of Episode
+        # Film wording is used instead of Episode, ger/jap, Dub/OmU
         'url': 'https://www.anime-on-demand.de/anime/39',
         'only_matching': True,
     }, {
-        # Episodes without titles
+        # Episodes without titles, jap, OmU
         'url': 'https://www.anime-on-demand.de/anime/162',
         'only_matching': True,
     }, {
         # ger/jap, Dub/OmU, account required
         'url': 'https://www.anime-on-demand.de/anime/169',
         'only_matching': True,
+    }, {
+        # Full length film, non-series, ger/jap, Dub/OmU, account required
+        'url': 'https://www.anime-on-demand.de/anime/185',
+        'only_matching': True,
     }]
 
     def _login(self):
@@ -110,35 +115,12 @@ class AnimeOnDemandIE(InfoExtractor):
 
         entries = []
 
-        for num, episode_html in enumerate(re.findall(
+        def extract_info(html, video_id, num=None):
-                r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', webpage), 1):
+            title, description = [None] * 2
-            episodebox_title = self._search_regex(
-                (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
-                 r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
-                episode_html, 'episodebox title', default=None, group='title')
-            if not episodebox_title:
-                continue
-
-            episode_number = int(self._search_regex(
-                r'(?:Episode|Film)\s*(\d+)',
-                episodebox_title, 'episode number', default=num))
-            episode_title = self._search_regex(
-                r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
-                episodebox_title, 'episode title', default=None)
-
-            video_id = 'episode-%d' % episode_number
-
-            common_info = {
-                'id': video_id,
-                'series': anime_title,
-                'episode': episode_title,
-                'episode_number': episode_number,
-            }
-
             formats = []
 
             for input_ in re.findall(
-                    r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', episode_html):
+                    r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', html):
                 attributes = extract_attributes(input_)
                 playlist_urls = []
                 for playlist_key in ('data-playlist', 'data-otherplaylist'):
@@ -161,7 +143,7 @@ class AnimeOnDemandIE(InfoExtractor):
                         format_id_list.append(lang)
                     if kind:
                         format_id_list.append(kind)
-                    if not format_id_list:
+                    if not format_id_list and num is not None:
                         format_id_list.append(compat_str(num))
                     format_id = '-'.join(format_id_list)
                     format_note = ', '.join(filter(None, (kind, lang_note)))
@@ -215,28 +197,74 @@ class AnimeOnDemandIE(InfoExtractor):
                             })
                         formats.extend(file_formats)
 
-            if formats:
+            return {
-                self._sort_formats(formats)
+                'title': title,
+                'description': description,
+                'formats': formats,
+            }
+
+        def extract_entries(html, video_id, common_info, num=None):
+            info = extract_info(html, video_id, num)
+
+            if info['formats']:
+                self._sort_formats(info['formats'])
                 f = common_info.copy()
-                f.update({
+                f.update(info)
-                    'title': title,
-                    'description': description,
-                    'formats': formats,
-                })
                 entries.append(f)
 
-            # Extract teaser only when full episode is not available
+            # Extract teaser/trailer only when full episode is not available
-            if not formats:
+            if not info['formats']:
                 m = re.search(
-                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>Teaser<',
+                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<',
-                    episode_html)
+                    html)
                 if m:
                     f = common_info.copy()
                     f.update({
-                        'id': '%s-teaser' % f['id'],
+                        'id': '%s-%s' % (f['id'], m.group('kind').lower()),
                         'title': m.group('title'),
                         'url': compat_urlparse.urljoin(url, m.group('href')),
                     })
                     entries.append(f)
 
+        def extract_episodes(html):
+            for num, episode_html in enumerate(re.findall(
+                    r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', html), 1):
+                episodebox_title = self._search_regex(
+                    (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
+                     r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
+                    episode_html, 'episodebox title', default=None, group='title')
+                if not episodebox_title:
+                    continue
+
+                episode_number = int(self._search_regex(
+                    r'(?:Episode|Film)\s*(\d+)',
+                    episodebox_title, 'episode number', default=num))
+                episode_title = self._search_regex(
+                    r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
+                    episodebox_title, 'episode title', default=None)
+
+                video_id = 'episode-%d' % episode_number
+
+                common_info = {
+                    'id': video_id,
+                    'series': anime_title,
+                    'episode': episode_title,
+                    'episode_number': episode_number,
+                }
+
+                extract_entries(episode_html, video_id, common_info)
+
+        def extract_film(html, video_id):
+            common_info = {
+                'id': anime_id,
+                'title': anime_title,
+                'description': anime_description,
+            }
+            extract_entries(html, video_id, common_info)
+
+        extract_episodes(webpage)
+
+        if not entries:
+            extract_film(webpage, anime_id)
+
         return self.playlist_result(entries, anime_id, anime_title, anime_description)

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/' +
+        embed_url = 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id
-                     video_id + '/vt/frame')
         webpage = self._download_webpage(embed_url, video_id)
 
-        video_urls = [video_url.replace('\\/', '/') for video_url in re.findall(
+        file_list = self._parse_json(self._search_regex(
-            r'(?:fileList\[[0-9]+\]\s*=|"file"\s*:)\s*"([^"]+)"', webpage)]
+            r'fileList\s*=\s*JSON\.parse\(\'([^\']+)\'\)', webpage, 'file list'), video_id)
-        for i, video_url in enumerate(video_urls):
+        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 .jwplatform import JWPlatformBaseIE
-from ..utils import unified_strdate
+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'
+        def get_optional(metadata, field):
-        data = self._download_json(json_url, video_id)
+            return metadata.get(field, [None])[0]
 
-        def get_optional(data_dict, field):
+        metadata = self._download_json(
-            return data_dict['metadata'].get(field, [None])[0]
+            'http://archive.org/details/' + video_id, video_id, query={
-
+                'output': 'json',
-        title = get_optional(data, 'title')
+            })['metadata']
-        description = get_optional(data, 'description')
+        info.update({
-        uploader = get_optional(data, 'creator')
+            'title': get_optional(metadata, 'title') or info.get('title'),
-        upload_date = unified_strdate(get_optional(data, 'date'))
+            'description': clean_html(get_optional(metadata, 'description')),
-
+        })
-        formats = [
+        if info.get('_type') != 'playlist':
-            {
+            info.update({
-                'format': fdata['format'],
+                'uploader': get_optional(metadata, 'creator'),
-                'url': 'http://' + data['server'] + data['dir'] + fn,
+                'upload_date': unified_strdate(get_optional(metadata, 'date')),
-                'file_size': int(fdata['size']),
+            })
-            }
+        return info
-            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'),
-        }

youtube_dl/extractor/ard.py

@@ -13,13 +13,14 @@ from ..utils import (
     parse_duration,
     unified_strdate,
     xpath_text,
+    update_url_query,
 )
 from ..compat import compat_etree_fromstring
 
 
 class ARDMediathekIE(InfoExtractor):
     IE_NAME = 'ARD:mediathek'
-    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
+    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
 
     _TESTS = [{
         'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
@@ -34,6 +35,7 @@ class ARDMediathekIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
         'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
@@ -44,6 +46,7 @@ class ARDMediathekIE(InfoExtractor):
             'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
             'duration': 5252,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         # audio
         'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
@@ -55,9 +58,22 @@ class ARDMediathekIE(InfoExtractor):
             'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
             'duration': 3240,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
         'only_matching': True,
+    }, {
+        # audio
+        'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
+        'md5': '4e8f00631aac0395fee17368ac0e9867',
+        'info_dict': {
+            'id': '30796318',
+            'ext': 'mp3',
+            'title': 'Vor dem Fest',
+            'description': 'md5:c0c1c8048514deaed2a73b3a60eecacb',
+            'duration': 3287,
+        },
+        'skip': 'Video is no longer available',
     }]
 
     def _extract_media_info(self, media_info_url, webpage, video_id):
@@ -113,11 +129,14 @@ class ARDMediathekIE(InfoExtractor):
                         continue
                     if ext == 'f4m':
                         formats.extend(self._extract_f4m_formats(
-                            stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124',
+                            update_url_query(stream_url, {
-                            video_id, preference=-1, f4m_id='hds', fatal=False))
+                                'hdcore': '3.1.1',
+                                'plugin': 'aasp-3.1.1.69.124'
+                            }),
+                            video_id, f4m_id='hds', fatal=False))
                     elif ext == 'm3u8':
                         formats.extend(self._extract_m3u8_formats(
-                            stream_url, video_id, 'mp4', preference=1, m3u8_id='hls', fatal=False))
+                            stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
                     else:
                         if server and server.startswith('rtmp'):
                             f = {
@@ -231,7 +250,8 @@ class ARDIE(InfoExtractor):
             'title': 'Die Story im Ersten: Mission unter falscher Flagge',
             'upload_date': '20140804',
             'thumbnail': 're:^https?://.*\.jpg$',
-        }
+        },
+        'skip': 'HTTP Error 404: Not Found',
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/arkena.py

@@ -0,0 +1,115 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    float_or_none,
+    int_or_none,
+    mimetype2ext,
+    parse_iso8601,
+    strip_jsonp,
+)
+
+
+class ArkenaIE(InfoExtractor):
+    _VALID_URL = r'https?://play\.arkena\.com/(?:config|embed)/avp/v\d/player/media/(?P<id>[^/]+)/[^/]+/(?P<account_id>\d+)'
+    _TESTS = [{
+        'url': 'https://play.arkena.com/embed/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411',
+        'md5': 'b96f2f71b359a8ecd05ce4e1daa72365',
+        'info_dict': {
+            'id': 'b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe',
+            'ext': 'mp4',
+            'title': 'Big Buck Bunny',
+            'description': 'Royalty free test video',
+            'timestamp': 1432816365,
+            'upload_date': '20150528',
+            'is_live': False,
+        },
+    }, {
+        'url': 'https://play.arkena.com/config/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411/?callbackMethod=jQuery1111023664739129262213_1469227693893',
+        'only_matching': True,
+    }, {
+        'url': 'http://play.arkena.com/config/avp/v1/player/media/327336/darkmatter/131064/?callbackMethod=jQuery1111002221189684892677_1469227595972',
+        'only_matching': True,
+    }, {
+        'url': 'http://play.arkena.com/embed/avp/v1/player/media/327336/darkmatter/131064/',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_url(webpage):
+        # See https://support.arkena.com/display/PLAY/Ways+to+embed+your+video
+        mobj = re.search(
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//play\.arkena\.com/embed/avp/.+?)\1',
+            webpage)
+        if mobj:
+            return mobj.group('url')
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        account_id = mobj.group('account_id')
+
+        playlist = self._download_json(
+            'https://play.arkena.com/config/avp/v2/player/media/%s/0/%s/?callbackMethod=_'
+            % (video_id, account_id),
+            video_id, transform_source=strip_jsonp)['Playlist'][0]
+
+        media_info = playlist['MediaInfo']
+        title = media_info['Title']
+        media_files = playlist['MediaFiles']
+
+        is_live = False
+        formats = []
+        for kind_case, kind_formats in media_files.items():
+            kind = kind_case.lower()
+            for f in kind_formats:
+                f_url = f.get('Url')
+                if not f_url:
+                    continue
+                is_live = f.get('Live') == 'true'
+                exts = (mimetype2ext(f.get('Type')), determine_ext(f_url, None))
+                if kind == 'm3u8' or 'm3u8' in exts:
+                    formats.extend(self._extract_m3u8_formats(
+                        f_url, video_id, 'mp4',
+                        entry_protocol='m3u8' if is_live else 'm3u8_native',
+                        m3u8_id=kind, fatal=False, live=is_live))
+                elif kind == 'flash' or 'f4m' in exts:
+                    formats.extend(self._extract_f4m_formats(
+                        f_url, video_id, f4m_id=kind, fatal=False))
+                elif kind == 'dash' or 'mpd' in exts:
+                    formats.extend(self._extract_mpd_formats(
+                        f_url, video_id, mpd_id=kind, fatal=False))
+                elif kind == 'silverlight':
+                    # TODO: process when ism is supported (see
+                    # https://github.com/rg3/youtube-dl/issues/8118)
+                    continue
+                else:
+                    tbr = float_or_none(f.get('Bitrate'), 1000)
+                    formats.append({
+                        'url': f_url,
+                        'format_id': '%s-%d' % (kind, tbr) if tbr else kind,
+                        'tbr': tbr,
+                    })
+        self._sort_formats(formats)
+
+        description = media_info.get('Description')
+        video_id = media_info.get('VideoId') or video_id
+        timestamp = parse_iso8601(media_info.get('PublishDate'))
+        thumbnails = [{
+            'url': thumbnail['Url'],
+            'width': int_or_none(thumbnail.get('Size')),
+        } for thumbnail in (media_info.get('Poster') or []) if thumbnail.get('Url')]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'timestamp': timestamp,
+            'is_live': is_live,
+            'thumbnails': thumbnails,
+            'formats': formats,
+        }

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':
+            if kind in ('video', 'audio'):
-                formats.extend(self._extract_audio(media, programme_id))
+                bitrate = int_or_none(media.get('bitrate'))
-            elif kind == 'video':
+                encoding = media.get('encoding')
-                formats.extend(self._extract_video(media, programme_id))
+                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
@@ -590,6 +581,7 @@ class BBCIE(BBCCoUkIE):
             'id': '150615_telabyad_kentin_cogu',
             'ext': 'mp4',
             'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
+            'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
             'timestamp': 1434397334,
             'upload_date': '20150615',
         },
@@ -603,6 +595,7 @@ class BBCIE(BBCCoUkIE):
             'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
             'ext': 'mp4',
             'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
+            'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
             'timestamp': 1434713142,
             'upload_date': '20150619',
         },
@@ -652,6 +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',
@@ -749,7 +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')
@@ -818,8 +828,29 @@ 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:
-                            entries.append(self._extract_from_playlist_sxml(
+                            entry = None
-                                playlist.get('progressiveDownloadUrl'), playlist_id, timestamp))
+                            for key in ('streaming', 'progressiveDownload'):
+                                playlist_url = playlist.get('%sUrl' % key)
+                                if not playlist_url:
+                                    continue
+                                try:
+                                    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.
+                                    # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
+                                    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)
@@ -852,6 +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),
@@ -869,7 +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)
@@ -998,10 +1073,10 @@ class BBCCoUkPlaylistBaseIE(InfoExtractor):
 
 class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
     IE_NAME = 'bbc.co.uk:iplayer:playlist'
-    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/episodes/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
     _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
     _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
         'info_dict': {
             'id': 'b05rcz9v',
@@ -1009,7 +1084,17 @@ class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
             'description': 'French thriller serial about a missing teenager.',
         },
         'playlist_mincount': 6,
-    }
+        'skip': 'This programme is not currently available on BBC iPlayer',
+    }, {
+        # Available for over a year unlike 30 days for most other programmes
+        'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32',
+        'info_dict': {
+            'id': 'p02tcc32',
+            'title': 'Bohemian Icons',
+            'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7',
+        },
+        'playlist_mincount': 10,
+    }]
 
     def _extract_title_and_description(self, webpage):
         title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)

youtube_dl/extractor/bigflix.py

@@ -12,7 +12,7 @@ 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': 'ec76aa9b1129e2e5b301a474e54fab74',
+        'md5': 'dc1b4aebb46e3a7077ecc0d9f43f61e3',
         'info_dict': {
             'id': '16537',
             'ext': 'mp4',
@@ -26,7 +26,7 @@ class BigflixIE(InfoExtractor):
             'id': '16070',
             'ext': 'mp4',
             'title': 'Madarasapatinam',
-            'description': 'md5:63b9b8ed79189c6f0418c26d9a3452ca',
+            'description': 'md5:9f0470b26a4ba8e824c823b5d95c2f6b',
             'formats': 'mincount:2',
         },
         'params': {

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',
@@ -41,73 +41,33 @@ 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',
+            '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,
+            'upload_date': '20160529',
+            'uploader': '喜欢拉面',
+            'uploader_id': '151066',
         },
-        'playlist': [{
-            'md5': '55cdadedf3254caaa0d5d27cf20a8f9c',
-            'info_dict': {
-                'id': '4808130_part1',
-                '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': '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/biobiochiletv.py

@@ -2,11 +2,15 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import remove_end
+from ..utils import (
+    ExtractorError,
+    remove_end,
+)
+from .rudo import RudoIE
 
 
 class BioBioChileTVIE(InfoExtractor):
-    _VALID_URL = r'https?://tv\.biobiochile\.cl/notas/(?:[^/]+/)+(?P<id>[^/]+)\.shtml'
+    _VALID_URL = r'https?://(?:tv|www)\.biobiochile\.cl/(?:notas|noticias)/(?:[^/]+/)+(?P<id>[^/]+)\.shtml'
 
     _TESTS = [{
         'url': 'http://tv.biobiochile.cl/notas/2015/10/21/sobre-camaras-y-camarillas-parlamentarias.shtml',
@@ -18,6 +22,7 @@ class BioBioChileTVIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': 'Fernando Atria',
         },
+        'skip': 'URL expired and redirected to http://www.biobiochile.cl/portada/bbtv/index.html',
     }, {
         # different uploader layout
         'url': 'http://tv.biobiochile.cl/notas/2016/03/18/natalia-valdebenito-repasa-a-diputado-hasbun-paso-a-la-categoria-de-hablar-brutalidades.shtml',
@@ -32,6 +37,16 @@ class BioBioChileTVIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+        'skip': 'URL expired and redirected to http://www.biobiochile.cl/portada/bbtv/index.html',
+    }, {
+        'url': 'http://www.biobiochile.cl/noticias/bbtv/comentarios-bio-bio/2016/07/08/edecanes-del-congreso-figuras-decorativas-que-le-cuestan-muy-caro-a-los-chilenos.shtml',
+        'info_dict': {
+            'id': 'edecanes-del-congreso-figuras-decorativas-que-le-cuestan-muy-caro-a-los-chilenos',
+            'ext': 'mp4',
+            'uploader': '(none)',
+            'upload_date': '20160708',
+            'title': 'Edecanes del Congreso: Figuras decorativas que le cuestan muy caro a los chilenos',
+        },
     }, {
         'url': 'http://tv.biobiochile.cl/notas/2015/10/22/ninos-transexuales-de-quien-es-la-decision.shtml',
         'only_matching': True,
@@ -45,42 +60,22 @@ class BioBioChileTVIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
+        rudo_url = RudoIE._extract_url(webpage)
+        if not rudo_url:
+            raise ExtractorError('No videos found')
+
         title = remove_end(self._og_search_title(webpage), ' - BioBioChile TV')
 
-        file_url = self._search_regex(
-            r'loadFWPlayerVideo\([^,]+,\s*(["\'])(?P<url>.+?)\1',
-            webpage, 'file url', group='url')
-
-        base_url = self._search_regex(
-            r'file\s*:\s*(["\'])(?P<url>.+?)\1\s*\+\s*fileURL', webpage,
-            'base url', default='http://unlimited2-cl.digitalproserver.com/bbtv/',
-            group='url')
-
-        formats = self._extract_m3u8_formats(
-            '%s%s/playlist.m3u8' % (base_url, file_url), video_id, 'mp4',
-            entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
-        f = {
-            'url': '%s%s' % (base_url, file_url),
-            'format_id': 'http',
-            'protocol': 'http',
-            'preference': 1,
-        }
-        if formats:
-            f_copy = formats[-1].copy()
-            f_copy.update(f)
-            f = f_copy
-        formats.append(f)
-        self._sort_formats(formats)
-
         thumbnail = self._og_search_thumbnail(webpage)
         uploader = self._html_search_regex(
-            r'<a[^>]+href=["\']https?://busca\.biobiochile\.cl/author[^>]+>(.+?)</a>',
+            r'<a[^>]+href=["\']https?://(?:busca|www)\.biobiochile\.cl/(?:lista/)?(?:author|autor)[^>]+>(.+?)</a>',
             webpage, 'uploader', fatal=False)
 
         return {
+            '_type': 'url_transparent',
+            'url': rudo_url,
             'id': video_id,
             'title': title,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'formats': formats,
         }

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/brightcove.py

@@ -26,6 +26,8 @@ from ..utils import (
     unescapeHTML,
     unsmuggle_url,
     update_url_query,
+    clean_html,
+    mimetype2ext,
 )
 
 
@@ -90,6 +92,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                 'description': 'md5:363109c02998fee92ec02211bd8000df',
                 'uploader': 'National Ballet of Canada',
             },
+            'skip': 'Video gone',
         },
         {
             # test flv videos served by akamaihd.net
@@ -108,7 +111,7 @@ class BrightcoveLegacyIE(InfoExtractor):
             },
         },
         {
-            # playlist test
+            # playlist with 'videoList'
             # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players
             'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL',
             'info_dict': {
@@ -117,6 +120,15 @@ class BrightcoveLegacyIE(InfoExtractor):
             },
             'playlist_mincount': 7,
         },
+        {
+            # playlist with 'playlistTab' (https://github.com/rg3/youtube-dl/issues/9965)
+            'url': 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=AQ%7E%7E,AAABXlLMdok%7E,NJ4EoMlZ4rZdx9eU1rkMVd8EaYPBBUlg',
+            'info_dict': {
+                'id': '1522758701001',
+                'title': 'Lesson 08',
+            },
+            'playlist_mincount': 10,
+        },
     ]
     FLV_VCODECS = {
         1: 'SORENSON',
@@ -298,13 +310,19 @@ class BrightcoveLegacyIE(InfoExtractor):
             info_url, player_key, 'Downloading playlist information')
 
         json_data = json.loads(playlist_info)
-        if 'videoList' not in json_data:
+        if 'videoList' in json_data:
+            playlist_info = json_data['videoList']
+            playlist_dto = playlist_info['mediaCollectionDTO']
+        elif 'playlistTabs' in json_data:
+            playlist_info = json_data['playlistTabs']
+            playlist_dto = playlist_info['lineupListDTO']['playlistDTOs'][0]
+        else:
             raise ExtractorError('Empty playlist')
-        playlist_info = json_data['videoList']
+
-        videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
+        videos = [self._extract_video_info(video_info) for video_info in playlist_dto['videoDTOs']]
 
         return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'],
-                                    playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
+                                    playlist_title=playlist_dto['displayName'])
 
     def _extract_video_info(self, video_info):
         video_id = compat_str(video_info['id'])
@@ -528,14 +546,16 @@ class BrightcoveNewIE(InfoExtractor):
         formats = []
         for source in json_data.get('sources', []):
             container = source.get('container')
-            source_type = source.get('type')
+            ext = mimetype2ext(source.get('type'))
             src = source.get('src')
-            if source_type == 'application/x-mpegURL' or container == 'M2TS':
+            if ext == 'ism':
+                continue
+            elif ext == 'm3u8' or container == 'M2TS':
                 if not src:
                     continue
                 formats.extend(self._extract_m3u8_formats(
                     src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
-            elif source_type == 'application/dash+xml':
+            elif ext == 'mpd':
                 if not src:
                     continue
                 formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False))
@@ -551,7 +571,7 @@ class BrightcoveNewIE(InfoExtractor):
                     'tbr': tbr,
                     'filesize': int_or_none(source.get('size')),
                     'container': container,
-                    'ext': container.lower(),
+                    'ext': ext or container.lower(),
                 }
                 if width == 0 and height == 0:
                     f.update({
@@ -585,6 +605,13 @@ class BrightcoveNewIE(InfoExtractor):
                         'format_id': build_format_id('rtmp'),
                     })
                 formats.append(f)
+
+        errors = json_data.get('errors')
+        if not formats and errors:
+            error = errors[0]
+            raise ExtractorError(
+                error.get('message') or error.get('error_subcode') or error['error_code'], expected=True)
+
         self._sort_formats(formats)
 
         subtitles = {}
@@ -597,7 +624,7 @@ class BrightcoveNewIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'description': json_data.get('description'),
+            'description': clean_html(json_data.get('description')),
             'thumbnail': json_data.get('thumbnail') or json_data.get('poster'),
             'duration': float_or_none(json_data.get('duration'), 1000),
             'timestamp': parse_iso8601(json_data.get('published_at')),

youtube_dl/extractor/camdemy.py

@@ -1,7 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import datetime
 import re
 
 from .common import InfoExtractor
@@ -10,8 +9,10 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
-    parse_iso8601,
+    clean_html,
+    parse_duration,
     str_to_int,
+    unified_strdate,
 )
 
 
@@ -26,14 +27,14 @@ class CamdemyIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Ch1-1 Introduction, Signals (02-23-2012)',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'description': '',
             'creator': 'ss11spring',
+            'duration': 1591,
             'upload_date': '20130114',
-            'timestamp': 1358154556,
             'view_count': int,
         }
     }, {
         # With non-empty description
+        # webpage returns "No permission or not login"
         'url': 'http://www.camdemy.com/media/13885',
         'md5': '4576a3bb2581f86c61044822adbd1249',
         'info_dict': {
@@ -41,64 +42,71 @@ class CamdemyIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'EverCam + Camdemy QuickStart',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'description': 'md5:050b62f71ed62928f8a35f1a41e186c9',
+            'description': 'md5:2a9f989c2b153a2342acee579c6e7db6',
             'creator': 'evercam',
-            'upload_date': '20140620',
+            'duration': 318,
-            'timestamp': 1403271569,
         }
     }, {
-        # External source
+        # External source (YouTube)
         'url': 'http://www.camdemy.com/media/14842',
-        'md5': '50e1c3c3aa233d3d7b7daa2fa10b1cf7',
         'info_dict': {
             'id': '2vsYQzNIsJo',
             'ext': 'mp4',
+            'title': 'Excel 2013 Tutorial - How to add Password Protection',
+            'description': 'Excel 2013 Tutorial for Beginners - How to add Password Protection',
             'upload_date': '20130211',
             'uploader': 'Hun Kim',
-            'description': 'Excel 2013 Tutorial for Beginners - How to add Password Protection',
             'uploader_id': 'hunkimtutorials',
-            'title': 'Excel 2013 Tutorial - How to add Password Protection',
+        },
-        }
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        page = self._download_webpage(url, video_id)
+
+        webpage = self._download_webpage(url, video_id)
 
         src_from = self._html_search_regex(
-            r"<div class='srcFrom'>Source: <a title='([^']+)'", page,
+            r"class=['\"]srcFrom['\"][^>]*>Sources?(?:\s+from)?\s*:\s*<a[^>]+(?:href|title)=(['\"])(?P<url>(?:(?!\1).)+)\1",
-            'external source', default=None)
+            webpage, 'external source', default=None, group='url')
         if src_from:
             return self.url_result(src_from)
 
         oembed_obj = self._download_json(
             'http://www.camdemy.com/oembed/?format=json&url=' + url, video_id)
 
+        title = oembed_obj['title']
         thumb_url = oembed_obj['thumbnail_url']
         video_folder = compat_urlparse.urljoin(thumb_url, 'video/')
         file_list_doc = self._download_xml(
             compat_urlparse.urljoin(video_folder, 'fileList.xml'),
-            video_id, 'Filelist XML')
+            video_id, 'Downloading filelist XML')
         file_name = file_list_doc.find('./video/item/fileName').text
         video_url = compat_urlparse.urljoin(video_folder, file_name)
 
-        timestamp = parse_iso8601(self._html_search_regex(
+        # Some URLs return "No permission or not login" in a webpage despite being
-            r"<div class='title'>Posted\s*:</div>\s*<div class='value'>([^<>]+)<",
+        # freely available via oembed JSON URL (e.g. http://www.camdemy.com/media/13885)
-            page, 'creation time', fatal=False),
+        upload_date = unified_strdate(self._search_regex(
-            delimiter=' ', timezone=datetime.timedelta(hours=8))
+            r'>published on ([^<]+)<', webpage,
-        view_count = str_to_int(self._html_search_regex(
+            'upload date', default=None))
-            r"<div class='title'>Views\s*:</div>\s*<div class='value'>([^<>]+)<",
+        view_count = str_to_int(self._search_regex(
-            page, 'view count', fatal=False))
+            r'role=["\']viewCnt["\'][^>]*>([\d,.]+) views',
+            webpage, 'view count', default=None))
+        description = self._html_search_meta(
+            'description', webpage, default=None) or clean_html(
+            oembed_obj.get('description'))
 
         return {
             'id': video_id,
             'url': video_url,
-            'title': oembed_obj['title'],
+            'title': title,
             'thumbnail': thumb_url,
-            'description': self._html_search_meta('description', page),
+            'description': description,
-            'creator': oembed_obj['author_name'],
+            'creator': oembed_obj.get('author_name'),
-            'duration': oembed_obj['duration'],
+            'duration': parse_duration(oembed_obj.get('duration')),
-            'timestamp': timestamp,
+            'upload_date': upload_date,
             'view_count': view_count,
         }
 

youtube_dl/extractor/cbc.py

@@ -4,9 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     js_to_json,
     smuggle_url,
+    try_get,
 )
 
 
@@ -25,8 +27,22 @@ class CBCIE(InfoExtractor):
             'upload_date': '20160203',
             'uploader': 'CBCC-NEW',
         },
+        'skip': 'Geo-restricted to Canada',
     }, {
-        # with clipId
+        # with clipId, feed available via tpfeed.cbc.ca and feed.theplatform.com
+        'url': 'http://www.cbc.ca/22minutes/videos/22-minutes-update/22-minutes-update-episode-4',
+        'md5': '162adfa070274b144f4fdc3c3b8207db',
+        'info_dict': {
+            'id': '2414435309',
+            'ext': 'mp4',
+            'title': '22 Minutes Update: What Not To Wear Quebec',
+            'description': "This week's latest Canadian top political story is What Not To Wear Quebec.",
+            'upload_date': '20131025',
+            'uploader': 'CBCC-NEW',
+            'timestamp': 1382717907,
+        },
+    }, {
+        # with clipId, feed only available via tpfeed.cbc.ca
         'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live',
         'md5': '0274a90b51a9b4971fe005c63f592f12',
         'info_dict': {
@@ -64,6 +80,7 @@ class CBCIE(InfoExtractor):
                 'uploader': 'CBCC-NEW',
             },
         }],
+        'skip': 'Geo-restricted to Canada',
     }]
 
     @classmethod
@@ -81,9 +98,15 @@ class CBCIE(InfoExtractor):
             media_id = player_info.get('mediaId')
             if not media_id:
                 clip_id = player_info['clipId']
-                media_id = self._download_json(
+                feed = self._download_json(
-                    'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
+                    'http://tpfeed.cbc.ca/f/ExhSPC/vms_5akSXx4Ng_Zn?byCustomValue={:mpsReleases}{%s}' % clip_id,
-                    clip_id)['entries'][0]['id'].split('/')[-1]
+                    clip_id, fatal=False)
+                if feed:
+                    media_id = try_get(feed, lambda x: x['entries'][0]['guid'], compat_str)
+                if not media_id:
+                    media_id = self._download_json(
+                        'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
+                        clip_id)['entries'][0]['id'].split('/')[-1]
             return self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
         else:
             entries = [self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id) for media_id in re.findall(r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"', webpage)]
@@ -104,6 +127,7 @@ class CBCPlayerIE(InfoExtractor):
             'upload_date': '20160210',
             'uploader': 'CBCC-NEW',
         },
+        'skip': 'Geo-restricted to Canada',
     }, {
         # Redirected from http://www.cbc.ca/player/AudioMobile/All%20in%20a%20Weekend%20Montreal/ID/2657632011/
         'url': 'http://www.cbc.ca/player/play/2657631896',

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
+        timestamp = unified_timestamp(time_str)
-        if time_str:
-            timestamp = calendar.timegm(datetime.datetime.strptime(
-                time_str, '%b %d, %Y %I:%M %p').timetuple())
 
         info_dict.update({
             'display_id': display_id,

youtube_dl/extractor/cbsnews.py

@@ -26,6 +26,7 @@ class CBSNewsIE(CBSBaseIE):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'Subscribers only',
         },
         {
             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
@@ -69,7 +70,7 @@ class CBSNewsLiveVideoIE(InfoExtractor):
     IE_DESC = 'CBS News Live Videos'
     _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
         'info_dict': {
             'id': 'clinton-sanders-prepare-to-face-off-in-nh',
@@ -77,7 +78,15 @@ class CBSNewsLiveVideoIE(InfoExtractor):
             'title': 'Clinton, Sanders Prepare To Face Off In NH',
             'duration': 334,
         },
-    }
+        'skip': 'Video gone, redirected to http://www.cbsnews.com/live/',
+    }, {
+        'url': 'http://www.cbsnews.com/live/video/video-shows-intense-paragliding-accident/',
+        'info_dict': {
+            'id': 'video-shows-intense-paragliding-accident',
+            'ext': 'flv',
+            'title': 'Video Shows Intense Paragliding Accident',
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

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 (
+from ..utils import parse_duration
-    parse_duration,
-    int_or_none,
-)
 
 
 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',
+        'url': 'http://chirb.it/be2abG',
-        'md5': '9847b0dad6ac3e074568bf2cfb197de8',
         'info_dict': {
-            'id': 'PrIPv5',
+            'id': 'be2abG',
             'ext': 'mp3',
-            'title': 'Фасадстрой',
+            'title': 'md5:f542ea253f5255240be4da375c6a5d7e',
-            'duration': 52,
+            'description': 'md5:f24a4e22a71763e32da5fed59e47c770',
-            'view_count': int,
+            'duration': 306,
-            'comment_count': int,
+        },
+        '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(
+        data_fd = self._search_regex(
-            r'"setFile"\s*,\s*"([^"]+)"', webpage, 'audio url')
+            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')
+            r'class=["\']chirbit-title["\'][^>]*>([^<]+)', webpage, 'title')
-        duration = parse_duration(self._html_search_meta(
+        description = self._search_regex(
-            'duration', webpage, 'duration', fatal=False))
+            r'<h3>Description</h3>\s*<pre[^>]*>([^<]+)</pre>',
-        view_count = int_or_none(self._search_regex(
+            webpage, 'description', default=None)
-            r'itemprop="playCount"\s*>(\d+)', webpage,
+        duration = parse_duration(self._search_regex(
-            'listen count', fatal=False))
+            r'class=["\']c-length["\'][^>]*>([^<]+)',
-        comment_count = int_or_none(self._search_regex(
+            webpage, 'duration', fatal=False))
-            r'>(\d+) Comments?:', webpage,
-            'comment count', 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/cliphunter.py

@@ -23,7 +23,7 @@ class CliphunterIE(InfoExtractor):
         (?P<id>[0-9]+)/
         (?P<seo>.+?)(?:$|[#\?])
     '''
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
         'md5': 'b7c9bbd4eb3a226ab91093714dcaa480',
         'info_dict': {
@@ -32,8 +32,19 @@ class CliphunterIE(InfoExtractor):
             'title': 'Fun Jynx Maze solo',
             'thumbnail': 're:^https?://.*\.jpg$',
             'age_limit': 18,
-        }
+        },
-    }
+        'skip': 'Video gone',
+    }, {
+        'url': 'http://www.cliphunter.com/w/2019449/ShesNew__My_booty_girlfriend_Victoria_Paradices_pussy_filled_with_jizz',
+        'md5': '55a723c67bfc6da6b0cfa00d55da8a27',
+        'info_dict': {
+            'id': '2019449',
+            'ext': 'mp4',
+            'title': 'ShesNew - My booty girlfriend, Victoria Paradice\'s pussy filled with jizz',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'age_limit': 18,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/cliprs.py

@@ -1,16 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
+from .onet import OnetBaseIE
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    int_or_none,
-    parse_iso8601,
-)
 
 
-class ClipRsIE(InfoExtractor):
+class ClipRsIE(OnetBaseIE):
     _VALID_URL = r'https?://(?:www\.)?clip\.rs/(?P<id>[^/]+)/\d+'
     _TEST = {
         'url': 'http://www.clip.rs/premijera-frajle-predstavljaju-novi-spot-za-pesmu-moli-me-moli/3732',
@@ -27,64 +21,13 @@ class ClipRsIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        display_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, display_id)
 
-        video_id = self._search_regex(
+        mvp_id = self._search_mvp_id(webpage)
-            r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
 
-        response = self._download_json(
+        info_dict = self._extract_from_id(mvp_id, webpage)
-            'http://qi.ckm.onetapi.pl/', video_id,
+        info_dict['display_id'] = display_id
-            query={
-                'body[id]': video_id,
-                'body[jsonrpc]': '2.0',
-                'body[method]': 'get_asset_detail',
-                'body[params][ID_Publikacji]': video_id,
-                'body[params][Service]': 'www.onet.pl',
-                'content-type': 'application/jsonp',
-                'x-onet-app': 'player.front.onetapi.pl',
-            })
 
-        error = response.get('error')
+        return info_dict
-        if error:
-            raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, error['message']), expected=True)
-
-        video = response['result'].get('0')
-
-        formats = []
-        for _, formats_dict in video['formats'].items():
-            if not isinstance(formats_dict, dict):
-                continue
-            for format_id, format_list in formats_dict.items():
-                if not isinstance(format_list, list):
-                    continue
-                for f in format_list:
-                    if not f.get('url'):
-                        continue
-                    formats.append({
-                        'url': f['url'],
-                        'format_id': format_id,
-                        'height': int_or_none(f.get('vertical_resolution')),
-                        'width': int_or_none(f.get('horizontal_resolution')),
-                        'abr': float_or_none(f.get('audio_bitrate')),
-                        'vbr': float_or_none(f.get('video_bitrate')),
-                    })
-        self._sort_formats(formats)
-
-        meta = video.get('meta', {})
-
-        title = self._og_search_title(webpage, default=None) or meta['title']
-        description = self._og_search_description(webpage, default=None) or meta.get('description')
-        duration = meta.get('length') or meta.get('lenght')
-        timestamp = parse_iso8601(meta.get('addDate'), ' ')
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'duration': duration,
-            'timestamp': timestamp,
-            'formats': formats,
-        }

youtube_dl/extractor/cloudy.py

@@ -6,7 +6,6 @@ import re
 from .common import InfoExtractor
 from ..compat import (
     compat_parse_qs,
-    compat_urllib_parse_urlencode,
     compat_HTTPError,
 )
 from ..utils import (
@@ -17,37 +16,26 @@ from ..utils import (
 
 
 class CloudyIE(InfoExtractor):
-    _IE_DESC = 'cloudy.ec and videoraj.ch'
+    _IE_DESC = 'cloudy.ec'
     _VALID_URL = r'''(?x)
-        https?://(?:www\.)?(?P<host>cloudy\.ec|videoraj\.(?:ch|to))/
+        https?://(?:www\.)?cloudy\.ec/
         (?:v/|embed\.php\?id=)
         (?P<id>[A-Za-z0-9]+)
         '''
-    _EMBED_URL = 'http://www.%s/embed.php?id=%s'
+    _EMBED_URL = 'http://www.cloudy.ec/embed.php?id=%s'
-    _API_URL = 'http://www.%s/api/player.api.php?%s'
+    _API_URL = 'http://www.cloudy.ec/api/player.api.php'
     _MAX_TRIES = 2
-    _TESTS = [
+    _TEST = {
-        {
+        'url': 'https://www.cloudy.ec/v/af511e2527aac',
-            'url': 'https://www.cloudy.ec/v/af511e2527aac',
+        'md5': '5cb253ace826a42f35b4740539bedf07',
-            'md5': '5cb253ace826a42f35b4740539bedf07',
+        'info_dict': {
-            'info_dict': {
+            'id': 'af511e2527aac',
-                'id': 'af511e2527aac',
+            'ext': 'flv',
-                'ext': 'flv',
+            'title': 'Funny Cats and Animals Compilation june 2013',
-                'title': 'Funny Cats and Animals Compilation june 2013',
-            }
-        },
-        {
-            'url': 'http://www.videoraj.to/v/47f399fd8bb60',
-            'md5': '7d0f8799d91efd4eda26587421c3c3b0',
-            'info_dict': {
-                'id': '47f399fd8bb60',
-                'ext': 'flv',
-                'title': 'Burning a New iPhone 5 with Gasoline - Will it Survive?',
-            }
         }
-    ]
+    }
 
-    def _extract_video(self, video_host, video_id, file_key, error_url=None, try_num=0):
+    def _extract_video(self, video_id, file_key, error_url=None, try_num=0):
 
         if try_num > self._MAX_TRIES - 1:
             raise ExtractorError('Unable to extract video URL', expected=True)
@@ -64,9 +52,8 @@ class CloudyIE(InfoExtractor):
                 'errorUrl': error_url,
             })
 
-        data_url = self._API_URL % (video_host, compat_urllib_parse_urlencode(form))
         player_data = self._download_webpage(
-            data_url, video_id, 'Downloading player data')
+            self._API_URL, video_id, 'Downloading player data', query=form)
         data = compat_parse_qs(player_data)
 
         try_num += 1
@@ -88,7 +75,7 @@ class CloudyIE(InfoExtractor):
             except ExtractorError as e:
                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in [404, 410]:
                     self.report_warning('Invalid video URL, requesting another', video_id)
-                    return self._extract_video(video_host, video_id, file_key, video_url, try_num)
+                    return self._extract_video(video_id, file_key, video_url, try_num)
 
         return {
             'id': video_id,
@@ -98,14 +85,13 @@ class CloudyIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_host = mobj.group('host')
         video_id = mobj.group('id')
 
-        url = self._EMBED_URL % (video_host, video_id)
+        url = self._EMBED_URL % video_id
         webpage = self._download_webpage(url, video_id)
 
         file_key = self._search_regex(
             [r'key\s*:\s*"([^"]+)"', r'filekey\s*=\s*"([^"]+)"'],
             webpage, 'file_key')
 
-        return self._extract_video(video_host, video_id, file_key)
+        return self._extract_video(video_id, file_key)

youtube_dl/extractor/cmt.py

@@ -1,5 +1,7 @@
 from __future__ import unicode_literals
+
 from .mtv import MTVIE
+from ..utils import ExtractorError
 
 
 class CMTIE(MTVIE):
@@ -16,7 +18,27 @@ class CMTIE(MTVIE):
             'title': 'Garth Brooks - "The Call (featuring Trisha Yearwood)"',
             'description': 'Blame It All On My Roots',
         },
+        'skip': 'Video not available',
+    }, {
+        'url': 'http://www.cmt.com/videos/misc/1504699/still-the-king-ep-109-in-3-minutes.jhtml#id=1739908',
+        'md5': 'e61a801ca4a183a466c08bd98dccbb1c',
+        'info_dict': {
+            'id': '1504699',
+            'ext': 'mp4',
+            'title': 'Still The King Ep. 109 in 3 Minutes',
+            'description': 'Relive or catch up with Still The King by watching this recap of season 1, episode 9. New episodes Sundays 9/8c.',
+            'timestamp': 1469421000.0,
+            'upload_date': '20160725',
+        },
     }, {
         'url': 'http://www.cmt.com/shows/party-down-south/party-down-south-ep-407-gone-girl/1738172/playlist/#id=1738172',
         'only_matching': True,
     }]
+
+    @classmethod
+    def _transform_rtmp_url(cls, rtmp_video_url):
+        if 'error_not_available.swf' in rtmp_video_url:
+            raise ExtractorError(
+                '%s said: video is not available' % cls.IE_NAME, expected=True)
+
+        return super(CMTIE, cls)._transform_rtmp_url(rtmp_video_url)

youtube_dl/extractor/comedycentral.py

@@ -1,17 +1,7 @@
 from __future__ import unicode_literals
 
-import re
-
 from .mtv import MTVServicesInfoExtractor
-from ..compat import (
+from .common import InfoExtractor
-    compat_str,
-    compat_urllib_parse_urlencode,
-)
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    unified_strdate,
-)
 
 
 class ComedyCentralIE(MTVServicesInfoExtractor):
@@ -26,8 +16,10 @@ class ComedyCentralIE(MTVServicesInfoExtractor):
         'info_dict': {
             'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354',
             'ext': 'mp4',
-            'title': 'CC:Stand-Up|Greg Fitzsimmons: Life on Stage|Uncensored - Too Good of a Mother',
+            'title': 'CC:Stand-Up|August 18, 2013|1|0101|Uncensored - Too Good of a Mother',
             'description': 'After a certain point, breastfeeding becomes c**kblocking.',
+            'timestamp': 1376798400,
+            'upload_date': '20130818',
         },
     }, {
         'url': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/interviews/6yx39d/exclusive-rand-paul-extended-interview',
@@ -35,241 +27,92 @@ class ComedyCentralIE(MTVServicesInfoExtractor):
     }]
 
 
-class ComedyCentralShowsIE(MTVServicesInfoExtractor):
+class ToshIE(MTVServicesInfoExtractor):
-    IE_DESC = 'The Daily Show / The Colbert Report'
+    IE_DESC = 'Tosh.0'
-    # urls can be abbreviations like :thedailyshow
+    _VALID_URL = r'^https?://tosh\.cc\.com/video-(?:clips|collections)/[^/]+/(?P<videotitle>[^/?#]+)'
-    # urls for episodes like:
+    _FEED_URL = 'http://tosh.cc.com/feeds/mrss'
-    # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
+
-    #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
-    #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
-    _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow)
-                      |https?://(:www\.)?
-                          (?P<showname>thedailyshow|thecolbertreport|tosh)\.(?:cc\.)?com/
-                         ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)|
-                          (?P<clip>
-                              (?:(?:guests/[^/]+|videos|video-(?:clips|playlists)|special-editions|news-team/[^/]+)/[^/]+/(?P<videotitle>[^/?#]+))
-                              |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
-                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
-                          )|
-                          (?P<interview>
-                              extended-interviews/(?P<interID>[0-9a-z]+)/
-                              (?:playlist_tds_extended_)?(?P<interview_title>[^/?#]*?)
-                              (?:/[^/?#]?|[?#]|$))))
-                     '''
     _TESTS = [{
-        'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
-        'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
-        'info_dict': {
-            'id': 'ab9ab3e7-5a98-4dbe-8b21-551dc0523d55',
-            'ext': 'mp4',
-            'upload_date': '20121213',
-            'description': 'Kristen Stewart learns to let loose in "On the Road."',
-            'uploader': 'thedailyshow',
-            'title': 'thedailyshow kristen-stewart part 1',
-        }
-    }, {
-        'url': 'http://thedailyshow.cc.com/extended-interviews/b6364d/sarah-chayes-extended-interview',
-        'info_dict': {
-            'id': 'sarah-chayes-extended-interview',
-            'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
-            'title': 'thedailyshow Sarah Chayes Extended Interview',
-        },
-        'playlist': [
-            {
-                'info_dict': {
-                    'id': '0baad492-cbec-4ec1-9e50-ad91c291127f',
-                    'ext': 'mp4',
-                    'upload_date': '20150129',
-                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
-                    'uploader': 'thedailyshow',
-                    'title': 'thedailyshow sarah-chayes-extended-interview part 1',
-                },
-            },
-            {
-                'info_dict': {
-                    'id': '1e4fb91b-8ce7-4277-bd7c-98c9f1bbd283',
-                    'ext': 'mp4',
-                    'upload_date': '20150129',
-                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
-                    'uploader': 'thedailyshow',
-                    'title': 'thedailyshow sarah-chayes-extended-interview part 2',
-                },
-            },
-        ],
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
-        'only_matching': True,
-    }, {
-        'url': 'http://thecolbertreport.cc.com/videos/29w6fx/-realhumanpraise-for-fox-news',
-        'only_matching': True,
-    }, {
-        'url': 'http://thecolbertreport.cc.com/videos/gh6urb/neil-degrasse-tyson-pt--1?xrs=eml_col_031114',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/episodes/sy7yv0/april-8--2014---denis-leary',
-        'only_matching': True,
-    }, {
-        'url': 'http://thecolbertreport.cc.com/episodes/8ase07/april-8--2014---jane-goodall',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/video-playlists/npde3s/the-daily-show-19088-highlights',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/video-playlists/t6d9sg/the-daily-show-20038-highlights/be3cwo',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/special-editions/2l8fdb/special-edition---a-look-back-at-food',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/news-team/michael-che/7wnfel/we-need-to-talk-about-israel',
-        'only_matching': True,
-    }, {
         'url': 'http://tosh.cc.com/video-clips/68g93d/twitter-users-share-summer-plans',
+        'info_dict': {
+            'description': 'Tosh asked fans to share their summer plans.',
+            'title': 'Twitter Users Share Summer Plans',
+        },
+        'playlist': [{
+            'md5': 'f269e88114c1805bb6d7653fecea9e06',
+            'info_dict': {
+                'id': '90498ec2-ed00-11e0-aca6-0026b9414f30',
+                'ext': 'mp4',
+                'title': 'Tosh.0|June 9, 2077|2|211|Twitter Users Share Summer Plans',
+                'description': 'Tosh asked fans to share their summer plans.',
+                'thumbnail': 're:^https?://.*\.jpg',
+                # It's really reported to be published on year 2077
+                'upload_date': '20770610',
+                'timestamp': 3390510600,
+                'subtitles': {
+                    'en': 'mincount:3',
+                },
+            },
+        }]
+    }, {
+        'url': 'http://tosh.cc.com/video-collections/x2iz7k/just-plain-foul/m5q4fp',
         'only_matching': True,
     }]
 
-    _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
+    @classmethod
+    def _transform_rtmp_url(cls, rtmp_video_url):
+        new_urls = super(ToshIE, cls)._transform_rtmp_url(rtmp_video_url)
+        new_urls['rtmp'] = rtmp_video_url.replace('viacomccstrm', 'viacommtvstrm')
+        return new_urls
 
-    _video_extensions = {
+
-        '3500': 'mp4',
+class ComedyCentralTVIE(MTVServicesInfoExtractor):
-        '2200': 'mp4',
+    _VALID_URL = r'https?://(?:www\.)?comedycentral\.tv/(?:staffeln|shows)/(?P<id>[^/?#&]+)'
-        '1700': 'mp4',
+    _TESTS = [{
-        '1200': 'mp4',
+        'url': 'http://www.comedycentral.tv/staffeln/7436-the-mindy-project-staffel-4',
-        '750': 'mp4',
+        'info_dict': {
-        '400': 'mp4',
+            'id': 'local_playlist-f99b626bdfe13568579a',
-    }
+            'ext': 'flv',
-    _video_dimensions = {
+            'title': 'Episode_the-mindy-project_shows_season-4_episode-3_full-episode_part1',
-        '3500': (1280, 720),
+        },
-        '2200': (960, 540),
+        'params': {
-        '1700': (768, 432),
+            # rtmp download
-        '1200': (640, 360),
+            'skip_download': True,
-        '750': (512, 288),
+        },
-        '400': (384, 216),
+    }, {
-    }
+        'url': 'http://www.comedycentral.tv/shows/1074-workaholics',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.comedycentral.tv/shows/1727-the-mindy-project/bonus',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
 
-        if mobj.group('shortname'):
+        webpage = self._download_webpage(url, video_id)
-            return self.url_result('http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes')
 
-        if mobj.group('clip'):
+        mrss_url = self._search_regex(
-            if mobj.group('videotitle'):
+            r'data-mrss=(["\'])(?P<url>(?:(?!\1).)+)\1',
-                epTitle = mobj.group('videotitle')
+            webpage, 'mrss url', group='url')
-            elif mobj.group('showname') == 'thedailyshow':
-                epTitle = mobj.group('tdstitle')
-            else:
-                epTitle = mobj.group('cntitle')
-            dlNewest = False
-        elif mobj.group('interview'):
-            epTitle = mobj.group('interview_title')
-            dlNewest = False
-        else:
-            dlNewest = not mobj.group('episode')
-            if dlNewest:
-                epTitle = mobj.group('showname')
-            else:
-                epTitle = mobj.group('episode')
-        show_name = mobj.group('showname')
 
-        webpage, htmlHandle = self._download_webpage_handle(url, epTitle)
+        return self._get_videos_info_from_url(mrss_url, video_id)
-        if dlNewest:
-            url = htmlHandle.geturl()
-            mobj = re.match(self._VALID_URL, url, re.VERBOSE)
-            if mobj is None:
-                raise ExtractorError('Invalid redirected URL: ' + url)
-            if mobj.group('episode') == '':
-                raise ExtractorError('Redirected URL is still not specific: ' + url)
-            epTitle = (mobj.group('episode') or mobj.group('videotitle')).rpartition('/')[-1]
 
-        mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
-        if len(mMovieParams) == 0:
-            # The Colbert Report embeds the information in a without
-            # a URL prefix; so extract the alternate reference
-            # and then add the URL prefix manually.
 
-            altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video|playlist).*?:.*?)"', webpage)
+class ComedyCentralShortnameIE(InfoExtractor):
-            if len(altMovieParams) == 0:
+    _VALID_URL = r'^:(?P<id>tds|thedailyshow)$'
-                raise ExtractorError('unable to find Flash URL in webpage ' + url)
+    _TESTS = [{
-            else:
+        'url': ':tds',
-                mMovieParams = [('http://media.mtvnservices.com/' + altMovieParams[0], altMovieParams[0])]
+        'only_matching': True,
+    }, {
+        'url': ':thedailyshow',
+        'only_matching': True,
+    }]
 
-        uri = mMovieParams[0][1]
+    def _real_extract(self, url):
-        # Correct cc.com in uri
+        video_id = self._match_id(url)
-        uri = re.sub(r'(episode:[^.]+)(\.cc)?\.com', r'\1.com', uri)
+        shortcut_map = {
-
+            'tds': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes',
-        index_url = 'http://%s.cc.com/feeds/mrss?%s' % (show_name, compat_urllib_parse_urlencode({'uri': uri}))
+            'thedailyshow': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes',
-        idoc = self._download_xml(
-            index_url, epTitle,
-            'Downloading show index', 'Unable to download episode index')
-
-        title = idoc.find('./channel/title').text
-        description = idoc.find('./channel/description').text
-
-        entries = []
-        item_els = idoc.findall('.//item')
-        for part_num, itemEl in enumerate(item_els):
-            upload_date = unified_strdate(itemEl.findall('./pubDate')[0].text)
-            thumbnail = itemEl.find('.//{http://search.yahoo.com/mrss/}thumbnail').attrib.get('url')
-
-            content = itemEl.find('.//{http://search.yahoo.com/mrss/}content')
-            duration = float_or_none(content.attrib.get('duration'))
-            mediagen_url = content.attrib['url']
-            guid = itemEl.find('./guid').text.rpartition(':')[-1]
-
-            cdoc = self._download_xml(
-                mediagen_url, epTitle,
-                'Downloading configuration for segment %d / %d' % (part_num + 1, len(item_els)))
-
-            turls = []
-            for rendition in cdoc.findall('.//rendition'):
-                finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
-                turls.append(finfo)
-
-            formats = []
-            for format, rtmp_video_url in turls:
-                w, h = self._video_dimensions.get(format, (None, None))
-                formats.append({
-                    'format_id': 'vhttp-%s' % format,
-                    'url': self._transform_rtmp_url(rtmp_video_url),
-                    'ext': self._video_extensions.get(format, 'mp4'),
-                    'height': h,
-                    'width': w,
-                })
-                formats.append({
-                    'format_id': 'rtmp-%s' % format,
-                    'url': rtmp_video_url.replace('viacomccstrm', 'viacommtvstrm'),
-                    'ext': self._video_extensions.get(format, 'mp4'),
-                    'height': h,
-                    'width': w,
-                })
-                self._sort_formats(formats)
-
-            subtitles = self._extract_subtitles(cdoc, guid)
-
-            virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
-            entries.append({
-                'id': guid,
-                'title': virtual_id,
-                'formats': formats,
-                'uploader': show_name,
-                'upload_date': upload_date,
-                'duration': duration,
-                'thumbnail': thumbnail,
-                'description': description,
-                'subtitles': subtitles,
-            })
-
-        return {
-            '_type': 'playlist',
-            'id': epTitle,
-            'entries': entries,
-            'title': show_name + ' ' + title,
-            'description': description,
         }
+        return self.url_result(shortcut_map[video_id])

youtube_dl/extractor/common.py

@@ -44,6 +44,7 @@ from ..utils import (
     sanitized_Request,
     unescapeHTML,
     unified_strdate,
+    unified_timestamp,
     url_basename,
     xpath_element,
     xpath_text,
@@ -54,6 +55,8 @@ from ..utils import (
     update_Request,
     update_url_query,
     parse_m3u8_attributes,
+    extract_attributes,
+    parse_codecs,
 )
 
 
@@ -161,6 +164,7 @@ class InfoExtractor(object):
                         * "height" (optional, int)
                         * "resolution" (optional, string "{width}x{height"},
                                         deprecated)
+                        * "filesize" (optional, int)
     thumbnail:      Full URL to a video thumbnail image.
     description:    Full video description.
     uploader:       Full name of the video uploader.
@@ -723,9 +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
+            name = 'OpenGraph %s' % prop[0]
-        escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs)
+        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)
@@ -803,40 +812,66 @@ class InfoExtractor(object):
         return self._html_search_meta('twitter:player', html,
                                       'twitter card player')
 
-    def _search_json_ld(self, html, video_id, **kwargs):
+    def _search_json_ld(self, html, video_id, expected_type=None, **kwargs):
         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 default if default is not NO_DEFAULT else {}
-        return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True))
+        # 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):
+    def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None):
         if isinstance(json_ld, compat_str):
             json_ld = self._parse_json(json_ld, video_id, fatal=fatal)
         if not json_ld:
             return {}
         info = {}
-        if json_ld.get('@context') == 'http://schema.org':
+        if not isinstance(json_ld, (list, tuple, dict)):
-            item_type = json_ld.get('@type')
+            return info
-            if item_type == 'TVEpisode':
+        if isinstance(json_ld, dict):
-                info.update({
+            json_ld = [json_ld]
-                    'episode': unescapeHTML(json_ld.get('name')),
+        for e in json_ld:
-                    'episode_number': int_or_none(json_ld.get('episodeNumber')),
+            if e.get('@context') == 'http://schema.org':
-                    'description': unescapeHTML(json_ld.get('description')),
+                item_type = e.get('@type')
-                })
+                if expected_type is not None and expected_type != item_type:
-                part_of_season = json_ld.get('partOfSeason')
+                    return info
-                if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason':
+                if item_type == 'TVEpisode':
-                    info['season_number'] = int_or_none(part_of_season.get('seasonNumber'))
+                    info.update({
-                part_of_series = json_ld.get('partOfSeries')
+                        'episode': unescapeHTML(e.get('name')),
-                if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries':
+                        'episode_number': int_or_none(e.get('episodeNumber')),
-                    info['series'] = unescapeHTML(part_of_series.get('name'))
+                        'description': unescapeHTML(e.get('description')),
-            elif item_type == 'Article':
+                    })
-                info.update({
+                    part_of_season = e.get('partOfSeason')
-                    'timestamp': parse_iso8601(json_ld.get('datePublished')),
+                    if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason':
-                    'title': unescapeHTML(json_ld.get('headline')),
+                        info['season_number'] = int_or_none(part_of_season.get('seasonNumber'))
-                    'description': unescapeHTML(json_ld.get('articleBody')),
+                    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(e.get('datePublished')),
+                        'title': unescapeHTML(e.get('headline')),
+                        'description': unescapeHTML(e.get('articleBody')),
+                    })
+                elif item_type == 'VideoObject':
+                    info.update({
+                        '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
@@ -890,7 +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
@@ -1107,7 +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',
         }
@@ -1186,6 +1222,7 @@ class InfoExtractor(object):
                     'url': format_url(line.strip()),
                     'tbr': tbr,
                     'ext': ext,
+                    'fps': float_or_none(last_info.get('FRAME-RATE')),
                     'protocol': entry_protocol,
                     'preference': preference,
                 }
@@ -1194,24 +1231,17 @@ class InfoExtractor(object):
                     width_str, height_str = resolution.split('x')
                     f['width'] = int(width_str)
                     f['height'] = int(height_str)
-                codecs = last_info.get('CODECS')
+                # Unified Streaming Platform
-                if codecs:
+                mobj = re.search(
-                    vcodec, acodec = [None] * 2
+                    r'audio.*?(?:%3D|=)(\d+)(?:-video.*?(?:%3D|=)(\d+))?', f['url'])
-                    va_codecs = codecs.split(',')
+                if mobj:
-                    if len(va_codecs) == 1:
+                    abr, vbr = mobj.groups()
-                        # Audio only entries usually come with single codec and
+                    abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000)
-                        # no resolution. For more robustness we also check it to
-                        # be mp4 audio.
-                        if not resolution and va_codecs[0].startswith('mp4a'):
-                            vcodec, acodec = 'none', va_codecs[0]
-                        else:
-                            vcodec = va_codecs[0]
-                    else:
-                        vcodec, acodec = va_codecs[:2]
                     f.update({
-                        'acodec': acodec,
+                        'vbr': vbr,
-                        'vcodec': vcodec,
+                        'abr': abr,
                     })
+                f.update(parse_codecs(last_info.get('CODECS')))
                 if last_media is not None:
                     f['m3u8_media'] = last_media
                     last_media = None
@@ -1466,6 +1496,13 @@ class InfoExtractor(object):
             compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url, formats_dict=formats_dict)
 
     def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}):
+        """
+        Parse formats from MPD manifest.
+        References:
+         1. MPEG-DASH Standard, ISO/IEC 23009-1:2014(E),
+            http://standards.iso.org/ittf/PubliclyAvailableStandards/c065274_ISO_IEC_23009-1_2014.zip
+         2. https://en.wikipedia.org/wiki/Dynamic_Adaptive_Streaming_over_HTTP
+        """
         if mpd_doc.get('type') == 'dynamic':
             return []
 
@@ -1498,8 +1535,16 @@ class InfoExtractor(object):
                         s_e = segment_timeline.findall(_add_ns('S'))
                         if s_e:
                             ms_info['total_number'] = 0
+                            ms_info['s'] = []
                             for s in s_e:
-                                ms_info['total_number'] += 1 + int(s.get('r', '0'))
+                                r = int(s.get('r', 0))
+                                ms_info['total_number'] += 1 + r
+                                ms_info['s'].append({
+                                    't': int(s.get('t', 0)),
+                                    # @d is mandatory (see [1, 5.3.9.6.2, Table 17, page 60])
+                                    'd': int(s.attrib['d']),
+                                    'r': r,
+                                })
                     else:
                         timescale = segment_template.get('timescale')
                         if timescale:
@@ -1536,7 +1581,7 @@ class InfoExtractor(object):
                         continue
                     representation_attrib = adaptation_set.attrib.copy()
                     representation_attrib.update(representation.attrib)
-                    # According to page 41 of ISO/IEC 29001-1:2014, @mimeType is mandatory
+                    # According to [1, 5.3.7.2, Table 9, page 41], @mimeType is mandatory
                     mime_type = representation_attrib['mimeType']
                     content_type = mime_type.split('/')[0]
                     if content_type == 'text':
@@ -1580,16 +1625,40 @@ class InfoExtractor(object):
                                 representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration))
                             media_template = representation_ms_info['media_template']
                             media_template = media_template.replace('$RepresentationID$', representation_id)
-                            media_template = re.sub(r'\$(Number|Bandwidth)\$', r'%(\1)d', media_template)
+                            media_template = re.sub(r'\$(Number|Bandwidth|Time)\$', r'%(\1)d', media_template)
-                            media_template = re.sub(r'\$(Number|Bandwidth)%([^$]+)\$', r'%(\1)\2', media_template)
+                            media_template = re.sub(r'\$(Number|Bandwidth|Time)%([^$]+)\$', r'%(\1)\2', media_template)
                             media_template.replace('$$', '$')
-                            representation_ms_info['segment_urls'] = [
+
-                                media_template % {
+                            # As per [1, 5.3.9.4.4, Table 16, page 55] $Number$ and $Time$
-                                    'Number': segment_number,
+                            # can't be used at the same time
-                                    'Bandwidth': representation_attrib.get('bandwidth')}
+                            if '%(Number' in media_template:
-                                for segment_number in range(
+                                representation_ms_info['segment_urls'] = [
-                                    representation_ms_info['start_number'],
+                                    media_template % {
-                                    representation_ms_info['total_number'] + representation_ms_info['start_number'])]
+                                        'Number': segment_number,
+                                        'Bandwidth': representation_attrib.get('bandwidth'),
+                                    }
+                                    for segment_number in range(
+                                        representation_ms_info['start_number'],
+                                        representation_ms_info['total_number'] + representation_ms_info['start_number'])]
+                            else:
+                                representation_ms_info['segment_urls'] = []
+                                segment_time = 0
+
+                                def add_segment_url():
+                                    representation_ms_info['segment_urls'].append(
+                                        media_template % {
+                                            'Time': segment_time,
+                                            'Bandwidth': representation_attrib.get('bandwidth'),
+                                        }
+                                    )
+
+                                for num, s in enumerate(representation_ms_info['s']):
+                                    segment_time = s.get('t') or segment_time
+                                    add_segment_url()
+                                    for r in range(s.get('r', 0)):
+                                        segment_time += s['d']
+                                        add_segment_url()
+                                    segment_time += s['d']
                         if 'segment_urls' in representation_ms_info:
                             f.update({
                                 'segment_urls': representation_ms_info['segment_urls'],
@@ -1616,6 +1685,62 @@ class InfoExtractor(object):
                         self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats
 
+    def _parse_html5_media_entries(self, base_url, webpage):
+        def absolute_url(video_url):
+            return compat_urlparse.urljoin(base_url, video_url)
+
+        def parse_content_type(content_type):
+            if not content_type:
+                return {}
+            ctr = re.search(r'(?P<mimetype>[^/]+/[^;]+)(?:;\s*codecs="?(?P<codecs>[^"]+))?', content_type)
+            if ctr:
+                mimetype, codecs = ctr.groups()
+                f = parse_codecs(codecs)
+                f['ext'] = mimetype2ext(mimetype)
+                return f
+            return {}
+
+        entries = []
+        for media_tag, media_type, media_content in re.findall(r'(?s)(<(?P<tag>video|audio)[^>]*>)(.*?)</(?P=tag)>', webpage):
+            media_info = {
+                'formats': [],
+                'subtitles': {},
+            }
+            media_attributes = extract_attributes(media_tag)
+            src = media_attributes.get('src')
+            if src:
+                media_info['formats'].append({
+                    'url': absolute_url(src),
+                    'vcodec': 'none' if media_type == 'audio' else None,
+                })
+            media_info['thumbnail'] = media_attributes.get('poster')
+            if media_content:
+                for source_tag in re.findall(r'<source[^>]+>', media_content):
+                    source_attributes = extract_attributes(source_tag)
+                    src = source_attributes.get('src')
+                    if not src:
+                        continue
+                    f = parse_content_type(source_attributes.get('type'))
+                    f.update({
+                        'url': absolute_url(src),
+                        'vcodec': 'none' if media_type == 'audio' else None,
+                    })
+                    media_info['formats'].append(f)
+                for track_tag in re.findall(r'<track[^>]+>', media_content):
+                    track_attributes = extract_attributes(track_tag)
+                    kind = track_attributes.get('kind')
+                    if not kind or kind == 'subtitles':
+                        src = track_attributes.get('src')
+                        if not src:
+                            continue
+                        lang = track_attributes.get('srclang') or track_attributes.get('lang') or track_attributes.get('label')
+                        media_info['subtitles'].setdefault(lang, []).append({
+                            'url': absolute_url(src),
+                        })
+            if media_info['formats']:
+                entries.append(media_info)
+        return entries
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()
@@ -1676,7 +1801,7 @@ class InfoExtractor(object):
 
         any_restricted = False
         for tc in self.get_testcases(include_onlymatching=False):
-            if 'playlist' in tc:
+            if tc.get('playlist', []):
                 tc = tc['playlist'][0]
             is_restricted = age_restricted(
                 tc.get('info_dict', {}).get('age_limit'), age_limit)
@@ -1729,6 +1854,13 @@ class InfoExtractor(object):
     def _mark_watched(self, *args, **kwargs):
         raise NotImplementedError('This method must be implemented by subclasses')
 
+    def geo_verification_headers(self):
+        headers = {}
+        geo_verification_proxy = self._downloader.params.get('geo_verification_proxy')
+        if geo_verification_proxy:
+            headers['Ytdl-request-proxy'] = geo_verification_proxy
+        return headers
+
 
 class SearchInfoExtractor(InfoExtractor):
     """

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>',
+        title = self._html_search_regex(
-                                        webpage, 'series title', flags=re.DOTALL)
+            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/.+?)["\?]',
+        m_paths = re.finditer(
-                              webpage, flags=re.DOTALL)
+            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':
+        query = {}
-            description = self._html_search_regex(
+        params = self._search_regex(
-                [
+            r'(?s)var params = {(.+?)}[;,]', webpage, 'player params', default=None)
-                    r'<div class="cne-video-description">(.+?)</div>',
+        if params:
-                    r'<div class="video-post-content">(.+?)</div>',
+            query.update({
-                ],
+                'videoId': self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id'),
-                webpage, 'description', fatal=False, flags=re.DOTALL)
+                'playerId': self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id'),
+                'target': self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target'),
+            })
         else:
-            description = None
+            params = extract_attributes(self._search_regex(
-        params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
+                r'(<[^>]+data-js="video-player"[^>]+>)',
-                                    'player params', flags=re.DOTALL)
+                webpage, 'player params element'))
-        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
+            query.update({
-        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id')
+                'videoId': params['data-video'],
-        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target')
+                'playerId': params['data-player'],
-        data = compat_urllib_parse_urlencode({'videoId': video_id,
+                'target': params['id'],
-                                              'playerId': player_id,
+            })
-                                              'target': target,
+        video_id = query['videoId']
-                                              })
+        video_info = None
-        base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
+        info_page = self._download_webpage(
-                                           webpage, 'base info url',
+            'http://player.cnevids.com/player/video.js',
-                                           default='http://player.cnevids.com/player/loader.js?')
+            video_id, 'Downloading video info', query=query, fatal=False)
-        info_url = base_info_url + data
+        if info_page:
-        info_page = self._download_webpage(info_url, video_id,
+            video_info = self._parse_json(self._search_regex(
-                                           'Downloading video info')
+                r'loadCallback\(({.+})\)', info_page, 'video info'), video_id)['video']
-        video_info = self._search_regex(r'var\s+video\s*=\s*({.+?});', info_page, 'video info')
+        else:
-        video_info = self._parse_json(video_info, video_id)
+            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 = [{
+        formats = []
-            'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
+        for fdata in video_info.get('sources', [{}])[0]:
-            'url': fdata['src'],
+            src = fdata.get('src')
-            'ext': fdata['type'].split('/')[-1],
+            if not src:
-            'quality': 1 if fdata['quality'] == 'high' else 0,
+                continue
-        } for fdata in video_info['sources'][0]]
+            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'],
+            'title': title,
-            'thumbnail': video_info['poster_frame'],
+            'thumbnail': video_info.get('poster_frame'),
-            'description': description,
+            '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, url_type, item_id = re.match(self._VALID_URL, url).groups()
-        site = mobj.group('site')
-        url_type = mobj.group('type')
-        item_id = mobj.group('id')
 
         # Convert JS embed to regular embed
         if url_type == 'embedjs':

youtube_dl/extractor/cspan.py

@@ -51,8 +51,11 @@ class CSpanIE(InfoExtractor):
         'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
         'info_dict': {
             'id': 'judiciary031715',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Immigration Reforms Needed to Protect Skilled American Workers',
+        },
+        'params': {
+            'skip_download': True,  # m3u8 downloads
         }
     }]
 

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',
+        'md5': 'e4726b2ccd70ba2c319865e28f0a91d1',
-        'add_ie': ['Youtube'],
         '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):
+        news_id = self._hidden_inputs(page).get('get_id')
-            feed_url = self._html_search_regex(
+
-                r'(http://news\.cts\.com\.tw/action/mp4feed\.php\?news_id=\d+)',
+        if news_id:
-                page, 'feed url')
+            mp4_feed = self._download_json(
-            video_url = self._download_webpage(
+                'http://news.cts.com.tw/action/test_mp4feed.php',
-                feed_url, news_id, note='Fetching feed')
+                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',
+                r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url')
-                default=None)
-            if not youtube_url:
-                raise ExtractorError('The news includes no videos!', expected=True)
 
-            return {
+            return self.url_result(youtube_url, ie='Youtube')
-                '_type': 'url',
-                'url': youtube_url,
-                'ie_key': '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')
+            r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time', fatal=False)
-        # Transform into ISO 8601 format with timezone info
+        timestamp = None
-        datetime_str = datetime_str.replace('/', '-') + ':00+0800'
+        if datetime_str:
-        timestamp = parse_iso8601(datetime_str, delimiter=' ')
+            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(
+        video_data = None
-            'http://metaframe.digitalsmiths.tv/v2/CWtv/assets/%s/partner/132?format=json' % video_id, video_id)
+        formats = []
-
+        for partner in (154, 213):
-        formats = self._extract_m3u8_formats(
+            vdata = self._download_json(
-            video_data['videos']['variantplaylist']['uri'], video_id, 'mp4')
+                '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/dailymail.py

@@ -5,19 +5,20 @@ from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     determine_protocol,
+    unescapeHTML,
 )
 
 
 class DailyMailIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)'
     _TEST = {
-        'url': 'http://www.dailymail.co.uk/video/sciencetech/video-1288527/Turn-video-impressionist-masterpiece.html',
+        'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
-        'md5': '2f639d446394f53f3a33658b518b6615',
+        'md5': 'f6129624562251f628296c3a9ffde124',
         'info_dict': {
-            'id': '1288527',
+            'id': '1295863',
             'ext': 'mp4',
-            'title': 'Turn any video into an impressionist masterpiece',
+            'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
-            'description': 'md5:88ddbcb504367987b2708bb38677c9d2',
+            'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
         }
     }
 
@@ -26,7 +27,7 @@ class DailyMailIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         video_data = self._parse_json(self._search_regex(
             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
-        title = video_data['title']
+        title = unescapeHTML(video_data['title'])
         video_sources = self._download_json(video_data.get(
             'sources', {}).get('url') or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id, video_id)
 
@@ -55,7 +56,7 @@ class DailyMailIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'description': video_data.get('descr'),
+            'description': unescapeHTML(video_data.get('descr')),
             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
             'formats': formats,
         }

youtube_dl/extractor/dailymotion.py

@@ -16,6 +16,7 @@ from ..utils import (
     sanitized_Request,
     str_to_int,
     unescapeHTML,
+    mimetype2ext,
 )
 
 
@@ -111,6 +112,13 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         }
     ]
 
+    @staticmethod
+    def _extract_urls(webpage):
+        # Look for embedded Dailymotion player
+        matches = re.findall(
+            r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
+        return list(map(lambda m: unescapeHTML(m[1]), matches))
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -153,18 +161,19 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                     type_ = media.get('type')
                     if type_ == 'application/vnd.lumberjack.manifest':
                         continue
-                    ext = determine_ext(media_url)
+                    ext = mimetype2ext(type_) or determine_ext(media_url)
-                    if type_ == 'application/x-mpegURL' or ext == 'm3u8':
+                    if ext == 'm3u8':
                         formats.extend(self._extract_m3u8_formats(
                             media_url, video_id, 'mp4', preference=-1,
                             m3u8_id='hls', fatal=False))
-                    elif type_ == 'application/f4m' or ext == 'f4m':
+                    elif ext == 'f4m':
                         formats.extend(self._extract_f4m_formats(
                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
                     else:
                         f = {
                             'url': media_url,
                             'format_id': 'http-%s' % quality,
+                            'ext': ext,
                         }
                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
                         if m:
@@ -322,7 +331,9 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 
             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
                 if video_id not in video_ids:
-                    yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
+                    yield self.url_result(
+                        'http://www.dailymotion.com/video/%s' % video_id,
+                        DailymotionIE.ie_key(), video_id)
                     video_ids.add(video_id)
 
             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:

youtube_dl/extractor/daum.py

@@ -66,22 +66,32 @@ class DaumIE(InfoExtractor):
             'view_count': int,
             'comment_count': int,
         },
+    }, {
+        # Requires dte_type=WEB (#9972)
+        'url': 'http://tvpot.daum.net/v/s3794Uf1NZeZ1qMpGpeqeRU',
+        'md5': 'a8917742069a4dd442516b86e7d66529',
+        'info_dict': {
+            'id': 's3794Uf1NZeZ1qMpGpeqeRU',
+            'ext': 'mp4',
+            'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny) [쇼! 음악중심] 508회 20160611',
+            'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\n\n[쇼! 음악중심] 20160611, 507회',
+            'upload_date': '20160611',
+        },
     }]
 
     def _real_extract(self, url):
         video_id = compat_urllib_parse_unquote(self._match_id(url))
-        query = compat_urllib_parse_urlencode({'vid': video_id})
         movie_data = self._download_json(
-            'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json?' + query,
+            'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json',
-            video_id, 'Downloading video formats info')
+            video_id, 'Downloading video formats info', query={'vid': video_id, 'dte_type': 'WEB'})
 
         # For urls like http://m.tvpot.daum.net/v/65139429, where the video_id is really a clipid
         if not movie_data.get('output_list', {}).get('output_list') and re.match(r'^\d+$', video_id):
             return self.url_result('http://tvpot.daum.net/clip/ClipView.do?clipid=%s' % video_id)
 
         info = self._download_xml(
-            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
+            'http://tvpot.daum.net/clip/ClipInfoXml.do', video_id,
-            'Downloading video info')
+            'Downloading video info', query={'vid': video_id})
 
         formats = []
         for format_el in movie_data['output_list']['output_list']:

youtube_dl/extractor/dbtv.py

@@ -4,78 +4,47 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
-from ..utils import (
-    float_or_none,
-    int_or_none,
-    clean_html,
-)
 
 
 class DBTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:(?:lazyplayer|player)/)?(?P<id>[0-9]+)(?:#(?P<display_id>.+))?'
+    _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:[^/]+/)?(?P<id>[0-9]+)(?:#(?P<display_id>.+))?'
     _TESTS = [{
         'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
-        'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc',
+        'md5': '2e24f67936517b143a234b4cadf792ec',
         'info_dict': {
-            'id': '33100',
+            'id': '3649835190001',
             'display_id': 'Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
             'ext': 'mp4',
             'title': 'Skulle teste ut fornøyelsespark, men kollegaen var bare opptatt av bikinikroppen',
             'description': 'md5:1504a54606c4dde3e4e61fc97aa857e0',
-            'thumbnail': 're:https?://.*\.jpg$',
+            'thumbnail': 're:https?://.*\.jpg',
-            'timestamp': 1404039863.438,
+            'timestamp': 1404039863,
             'upload_date': '20140629',
             'duration': 69.544,
-            'view_count': int,
+            'uploader_id': '1027729757001',
-            'categories': list,
+        },
-        }
+        'add_ie': ['BrightcoveNew']
     }, {
         'url': 'http://dbtv.no/3649835190001',
         'only_matching': True,
     }, {
         'url': 'http://www.dbtv.no/lazyplayer/4631135248001',
         'only_matching': True,
+    }, {
+        'url': 'http://dbtv.no/vice/5000634109001',
+        'only_matching': True,
+    }, {
+        'url': 'http://dbtv.no/filmtrailer/3359293614001',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id') or video_id
-
-        data = self._download_json(
-            'http://api.dbtv.no/discovery/%s' % video_id, display_id)
-
-        video = data['playlist'][0]
-
-        formats = [{
-            'url': f['URL'],
-            'vcodec': f.get('container'),
-            'width': int_or_none(f.get('width')),
-            'height': int_or_none(f.get('height')),
-            'vbr': float_or_none(f.get('rate'), 1000),
-            'filesize': int_or_none(f.get('size')),
-        } for f in video['renditions'] if 'URL' in f]
-
-        if not formats:
-            for url_key, format_id in [('URL', 'mp4'), ('HLSURL', 'hls')]:
-                if url_key in video:
-                    formats.append({
-                        'url': video[url_key],
-                        'format_id': format_id,
-                    })
-
-        self._sort_formats(formats)
 
         return {
-            'id': compat_str(video['id']),
+            '_type': 'url_transparent',
+            'url': 'http://players.brightcove.net/1027729757001/default_default/index.html?videoId=%s' % video_id,
+            'id': video_id,
             'display_id': display_id,
-            'title': video['title'],
+            'ie_key': 'BrightcoveNew',
-            'description': clean_html(video['desc']),
-            'thumbnail': video.get('splash') or video.get('thumb'),
-            'timestamp': float_or_none(video.get('publishedAt'), 1000),
-            'duration': float_or_none(video.get('length'), 1000),
-            'view_count': int_or_none(video.get('views')),
-            'categories': video.get('tags'),
-            'formats': formats,
         }

youtube_dl/extractor/dcn.py

@@ -62,11 +62,9 @@ class DCNBaseIE(InfoExtractor):
                 r'file\s*:\s*"https?(://[^"]+)/playlist.m3u8',
                 r'<a[^>]+href="rtsp(://[^"]+)"'
             ], webpage, 'format url')
-        # TODO: Current DASH formats are broken - $Time$ pattern in
+        formats.extend(self._extract_mpd_formats(
-        # <SegmentTemplate> not implemented yet
+            format_url_base + '/manifest.mpd',
-        # formats.extend(self._extract_mpd_formats(
+            video_id, mpd_id='dash', fatal=False))
-        #     format_url_base + '/manifest.mpd',
-        #     video_id, mpd_id='dash', fatal=False))
         formats.extend(self._extract_m3u8_formats(
             format_url_base + '/playlist.m3u8', video_id, 'mp4',
             m3u8_entry_protocol, m3u8_id='hls', fatal=False))

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/dreisat.py

@@ -17,8 +17,12 @@ class DreiSatIE(ZDFIE):
                 'ext': 'mp4',
                 'title': 'Waidmannsheil',
                 'description': 'md5:cce00ca1d70e21425e72c86a98a56817',
-                'uploader': '3sat',
+                'uploader': 'SCHWEIZWEIT',
+                'uploader_id': '100000210',
                 'upload_date': '20140913'
+            },
+            'params': {
+                'skip_download': True,  # m3u8 downloads
             }
         },
         {

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)
+            r'<div[^>]+class="categories_list">(.+?)</div>',
-        categories = [] if not cats_str else re.findall(r'<a title="([^"]+)"', cats_str)
+            webpage, 'categories', fatal=False)
+        categories = [] if not cats_str else re.findall(
+            r'<a title="([^"]+)"', cats_str)
 
         return {
             'id': video_id,

youtube_dl/extractor/ellentv.py

@@ -6,12 +6,13 @@ import json
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    NO_DEFAULT,
 )
 
 
 class EllenTVIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?(?:ellentv|ellentube)\.com/videos/(?P<id>[a-z0-9_-]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.ellentv.com/videos/0-ipq1gsai/',
         'md5': '4294cf98bc165f218aaa0b89e0fd8042',
         'info_dict': {
@@ -22,24 +23,47 @@ class EllenTVIE(InfoExtractor):
             'timestamp': 1428035648,
             'upload_date': '20150403',
             'uploader_id': 'batchUser',
-        }
+        },
-    }
+    }, {
+        # not available via http://widgets.ellentube.com/
+        'url': 'http://www.ellentv.com/videos/1-szkgu2m2/',
+        'info_dict': {
+            'id': '1_szkgu2m2',
+            'ext': 'flv',
+            'title': "Ellen's Amazingly Talented Audience",
+            'description': 'md5:86ff1e376ff0d717d7171590e273f0a5',
+            'timestamp': 1255140900,
+            'upload_date': '20091010',
+            'uploader_id': 'ellenkaltura@gmail.com',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(
+        URLS = ('http://widgets.ellentube.com/videos/%s' % video_id, url)
-            'http://widgets.ellentube.com/videos/%s' % video_id,
-            video_id)
 
-        partner_id = self._search_regex(
+        for num, url_ in enumerate(URLS, 1):
-            r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner id')
+            webpage = self._download_webpage(
+                url_, video_id, fatal=num == len(URLS))
 
-        kaltura_id = self._search_regex(
+            default = NO_DEFAULT if num == len(URLS) else None
-            [r'id="kaltura_player_([^"]+)"',
+
-             r"_wb_entry_id\s*:\s*'([^']+)",
+            partner_id = self._search_regex(
-             r'data-kaltura-entry-id="([^"]+)'],
+                r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner id',
-            webpage, 'kaltura id')
+                default=default)
+
+            kaltura_id = self._search_regex(
+                [r'id="kaltura_player_([^"]+)"',
+                 r"_wb_entry_id\s*:\s*'([^']+)",
+                 r'data-kaltura-entry-id="([^"]+)'],
+                webpage, 'kaltura id', default=default)
+
+            if partner_id and kaltura_id:
+                break
 
         return self.url_result('kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura')
 

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/eporner.py

@@ -4,19 +4,23 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
+    encode_base_n,
+    ExtractorError,
+    int_or_none,
     parse_duration,
     str_to_int,
 )
 
 
 class EpornerIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\w+)/(?P<display_id>[\w-]+)'
+    _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\w+)(?:/(?P<display_id>[\w-]+))?'
     _TESTS = [{
         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
         'md5': '39d486f046212d8e1b911c52ab4691f8',
         'info_dict': {
-            'id': '95008',
+            'id': 'qlDUmNsj6VS',
             'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
             'ext': 'mp4',
             'title': 'Infamous Tiffany Teen Strip Tease Video',
@@ -28,34 +32,72 @@ class EpornerIE(InfoExtractor):
         # New (May 2016) URL layout
         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
         'only_matching': True,
+    }, {
+        'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        display_id = mobj.group('display_id')
+        display_id = mobj.group('display_id') or video_id
 
-        webpage = self._download_webpage(url, display_id)
+        webpage, urlh = self._download_webpage_handle(url, display_id)
-        title = self._html_search_regex(
-            r'<title>(.*?) - EPORNER', webpage, 'title')
 
-        redirect_url = 'http://www.eporner.com/config5/%s' % video_id
+        video_id = self._match_id(compat_str(urlh.geturl()))
-        player_code = self._download_webpage(
-            redirect_url, display_id, note='Downloading player config')
 
-        sources = self._search_regex(
+        hash = self._search_regex(
-            r'(?s)sources\s*:\s*\[\s*({.+?})\s*\]', player_code, 'sources')
+            r'hash\s*:\s*["\']([\da-f]{32})', webpage, 'hash')
+
+        title = self._og_search_title(webpage, default=None) or self._html_search_regex(
+            r'<title>(.+?) - EPORNER', webpage, 'title')
+
+        # Reverse engineered from vjs.js
+        def calc_hash(s):
+            return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
+
+        video = self._download_json(
+            'http://www.eporner.com/xhr/video/%s' % video_id,
+            display_id, note='Downloading video JSON',
+            query={
+                'hash': calc_hash(hash),
+                'device': 'generic',
+                'domain': 'www.eporner.com',
+                'fallback': 'false',
+            })
+
+        if video.get('available') is False:
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
+
+        sources = video['sources']
 
         formats = []
-        for video_url, format_id in re.findall(r'file\s*:\s*"([^"]+)",\s*label\s*:\s*"([^"]+)"', sources):
+        for kind, formats_dict in sources.items():
-            fmt = {
+            if not isinstance(formats_dict, dict):
-                'url': video_url,
+                continue
-                'format_id': format_id,
+            for format_id, format_dict in formats_dict.items():
-            }
+                if not isinstance(format_dict, dict):
-            m = re.search(r'^(\d+)', format_id)
+                    continue
-            if m:
+                src = format_dict.get('src')
-                fmt['height'] = int(m.group(1))
+                if not isinstance(src, compat_str) or not src.startswith('http'):
-            formats.append(fmt)
+                    continue
+                if kind == 'hls':
+                    formats.extend(self._extract_m3u8_formats(
+                        src, display_id, 'mp4', entry_protocol='m3u8_native',
+                        m3u8_id=kind, fatal=False))
+                else:
+                    height = int_or_none(self._search_regex(
+                        r'(\d+)[pP]', format_id, 'height', default=None))
+                    fps = int_or_none(self._search_regex(
+                        r'(\d+)fps', format_id, 'fps', default=None))
+
+                    formats.append({
+                        'url': src,
+                        'format_id': format_id,
+                        'height': height,
+                        'fps': fps,
+                    })
         self._sort_formats(formats)
 
         duration = parse_duration(self._html_search_meta('duration', webpage))

youtube_dl/extractor/extractors.py

@@ -44,6 +44,7 @@ from .appletrailers import (
     AppleTrailersSectionIE,
 )
 from .archiveorg import ArchiveOrgIE
+from .arkena import ArkenaIE
 from .ard import (
     ARDIE,
     ARDMediathekIE,
@@ -139,9 +140,9 @@ from .chirbit import (
     ChirbitProfileIE,
 )
 from .cinchcast import CinchcastIE
-from .cliprs import ClipRsIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
+from .cliprs import ClipRsIE
 from .clipsyndicate import ClipsyndicateIE
 from .closertotruth import CloserToTruthIE
 from .cloudy import CloudyIE
@@ -156,7 +157,12 @@ from .cnn import (
 )
 from .coub import CoubIE
 from .collegerama import CollegeRamaIE
-from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
+from .comedycentral import (
+    ComedyCentralIE,
+    ComedyCentralShortnameIE,
+    ComedyCentralTVIE,
+    ToshIE,
+)
 from .comcarcoff import ComCarCoffIE
 from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
 from .commonprotocols import RtmpIE
@@ -215,6 +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 (
@@ -256,6 +263,7 @@ from .fivemin import FiveMinIE
 from .fivetv import FiveTVIE
 from .fktv import FKTVIE
 from .flickr import FlickrIE
+from .flipagram import FlipagramIE
 from .folketinget import FolketingetIE
 from .footyroom import FootyRoomIE
 from .formula1 import Formula1IE
@@ -283,7 +291,6 @@ 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,
@@ -304,7 +311,6 @@ 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
@@ -368,6 +374,7 @@ from .jove import JoveIE
 from .jwplatform import JWPlatformIE
 from .jpopsukitv import JpopsukiIE
 from .kaltura import KalturaIE
+from .kamcord import KamcordIE
 from .kanalplay import KanalPlayIE
 from .kankan import KankanIE
 from .karaoketv import KaraoketvIE
@@ -391,6 +398,10 @@ from .kuwo import (
 )
 from .la7 import LA7IE
 from .laola1tv import Laola1TvIE
+from .lcp import (
+    LcpPlayIE,
+    LcpIE,
+)
 from .learnr import LearnrIE
 from .lecture2go import Lecture2GoIE
 from .lemonde import LemondeIE
@@ -469,7 +480,6 @@ from .msn import MSNIE
 from .mtv import (
     MTVIE,
     MTVServicesEmbeddedIE,
-    MTVIggyIE,
     MTVDEIE,
 )
 from .muenchentv import MuenchenTVIE
@@ -481,8 +491,9 @@ 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
@@ -519,7 +530,6 @@ from .nextmedia import (
     NextMediaActionNewsIE,
     AppleDailyIE,
 )
-from .nextmovie import NextMovieIE
 from .nfb import NFBIE
 from .nfl import NFLIE
 from .nhl import (
@@ -535,6 +545,8 @@ from .nick import (
 from .niconico import NiconicoIE, NiconicoPlaylistIE
 from .ninecninemedia import NineCNineMediaIE
 from .ninegag import NineGagIE
+from .ninenow import NineNowIE
+from .nintendo import NintendoIE
 from .noco import NocoIE
 from .normalboots import NormalbootsIE
 from .nosvideo import NosVideoIE
@@ -579,8 +591,13 @@ from .nytimes import (
     NYTimesArticleIE,
 )
 from .nuvid import NuvidIE
+from .odatv import OdaTVIE
 from .odnoklassniki import OdnoklassnikiIE
 from .oktoberfesttv import OktoberfestTVIE
+from .onet import (
+    OnetIE,
+    OnetChannelIE,
+)
 from .onionstudios import OnionStudiosIE
 from .ooyala import (
     OoyalaIE,
@@ -619,6 +636,7 @@ from .pluralsight import (
     PluralsightCourseIE,
 )
 from .podomatic import PodomaticIE
+from .pokemon import PokemonIE
 from .polskieradio import PolskieRadioIE
 from .porn91 import Porn91IE
 from .pornhd import PornHdIE
@@ -674,16 +692,19 @@ from .rice import RICEIE
 from .ringtv import RingTVIE
 from .ro220 import Ro220IE
 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
 from .rtl2 import RTL2IE
 from .rtp import RTPIE
 from .rts import RTSIE
-from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE
+from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE, RTVELiveIE, RTVETelevisionIE
 from .rtvnh import RTVNHIE
+from .rudo import RudoIE
 from .ruhd import RUHDIE
 from .ruleporn import RulePornIE
 from .rutube import (
@@ -734,6 +755,7 @@ from .smotri import (
 )
 from .snotr import SnotrIE
 from .sohu import SohuIE
+from .sonyliv import SonyLIVIE
 from .soundcloud import (
     SoundcloudIE,
     SoundcloudSetIE,
@@ -773,6 +795,7 @@ from .srmediathek import SRMediathekIE
 from .ssa import SSAIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .steam import SteamIE
+from .streamable import StreamableIE
 from .streamcloud import StreamcloudIE
 from .streamcz import StreamCZIE
 from .streetvoice import StreetVoiceIE
@@ -872,6 +895,7 @@ from .tvc import (
 from .tvigle import TvigleIE
 from .tvland import TVLandIE
 from .tvp import (
+    TVPEmbedIE,
     TVPIE,
     TVPSeriesIE,
 )
@@ -904,6 +928,7 @@ 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
@@ -981,9 +1006,11 @@ from .viki import (
 from .vk import (
     VKIE,
     VKUserVideosIE,
+    VKWallPostIE,
 )
 from .vlive import VLiveIE
 from .vodlocker import VodlockerIE
+from .vodplatform import VODPlatformIE
 from .voicerepublic import VoiceRepublicIE
 from .voxmedia import VoxMediaIE
 from .vporn import VpornIE
@@ -1066,6 +1093,7 @@ from .youtube import (
     YoutubeSearchDateIE,
     YoutubeSearchIE,
     YoutubeSearchURLIE,
+    YoutubeSharedVideoIE,
     YoutubeShowIE,
     YoutubeSubscriptionsIE,
     YoutubeTruncatedIDIE,

youtube_dl/extractor/facebook.py

@@ -27,7 +27,7 @@ class FacebookIE(InfoExtractor):
     _VALID_URL = r'''(?x)
                 (?:
                     https?://
-                        (?:\w+\.)?facebook\.com/
+                        (?:[\w-]+\.)?facebook\.com/
                         (?:[^#]*?\#!/)?
                         (?:
                             (?:
@@ -127,6 +127,9 @@ class FacebookIE(InfoExtractor):
     }, {
         'url': 'https://www.facebook.com/groups/164828000315060/permalink/764967300301124/',
         'only_matching': True,
+    }, {
+        'url': 'https://zh-hk.facebook.com/peoplespower/videos/1135894589806027/',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -219,12 +222,25 @@ class FacebookIE(InfoExtractor):
 
         BEFORE = '{swf.addParam(param[0], param[1]);});'
         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
-        m = re.search(re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER), webpage)
+        PATTERN = re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER)
-        if m:
+
-            swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"')
+        for m in re.findall(PATTERN, webpage):
+            swf_params = m.replace('\\\\', '\\').replace('\\"', '"')
             data = dict(json.loads(swf_params))
             params_raw = compat_urllib_parse_unquote(data['params'])
-            video_data = json.loads(params_raw)['video_data']
+            video_data_candidate = json.loads(params_raw)['video_data']
+            for _, f in video_data_candidate.items():
+                if not f:
+                    continue
+                if isinstance(f, dict):
+                    f = [f]
+                if not isinstance(f, list):
+                    continue
+                if f[0].get('video_id') == video_id:
+                    video_data = video_data_candidate
+                    break
+            if video_data:
+                break
 
         def video_data_list2dict(video_data):
             ret = {}

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 = self._match_id(url)
-        video_id = mobj.group('id')
+        return self.url_result('aol-video:%s' % video_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,
-        }

youtube_dl/extractor/flipagram.py

@@ -0,0 +1,115 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    float_or_none,
+    try_get,
+    unified_timestamp,
+)
+
+
+class FlipagramIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?flipagram\.com/f/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://flipagram.com/f/nyvTSJMKId',
+        'md5': '888dcf08b7ea671381f00fab74692755',
+        'info_dict': {
+            'id': 'nyvTSJMKId',
+            'ext': 'mp4',
+            'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
+            'description': 'md5:d55e32edc55261cae96a41fa85ff630e',
+            'duration': 35.571,
+            'timestamp': 1461244995,
+            'upload_date': '20160421',
+            'uploader': 'kitty juria',
+            'uploader_id': 'sjuria101',
+            'creator': 'kitty juria',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+            'comments': list,
+            'formats': 'mincount:2',
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        video_data = self._parse_json(
+            self._search_regex(
+                r'window\.reactH2O\s*=\s*({.+});', webpage, 'video data'),
+            video_id)
+
+        flipagram = video_data['flipagram']
+        video = flipagram['video']
+
+        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')
+
+        formats = [{
+            'url': video['url'],
+            'width': int_or_none(video.get('width')),
+            'height': int_or_none(video.get('height')),
+            'filesize': int_or_none(video_data.get('size')),
+        }]
+
+        preview_url = try_get(
+            flipagram, lambda x: x['music']['track']['previewUrl'], compat_str)
+        if preview_url:
+            formats.append({
+                'url': preview_url,
+                'ext': 'm4a',
+                'vcodec': 'none',
+            })
+
+        self._sort_formats(formats)
+
+        counts = flipagram.get('counts', {})
+        user = flipagram.get('user', {})
+        video_data = flipagram.get('video', {})
+
+        thumbnails = [{
+            'url': self._proto_relative_url(cover['url']),
+            'width': int_or_none(cover.get('width')),
+            'height': int_or_none(cover.get('height')),
+            'filesize': int_or_none(cover.get('size')),
+        } for cover in flipagram.get('covers', []) if cover.get('url')]
+
+        # Note that this only retrieves comments that are initally loaded.
+        # For videos with large amounts of comments, most won't be retrieved.
+        comments = []
+        for comment in video_data.get('comments', {}).get(video_id, {}).get('items', []):
+            text = comment.get('comment')
+            if not text or not isinstance(text, list):
+                continue
+            comments.append({
+                'author': comment.get('user', {}).get('name'),
+                'author_id': comment.get('user', {}).get('username'),
+                'id': comment.get('id'),
+                'text': text[0],
+                'timestamp': unified_timestamp(comment.get('created')),
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'duration': float_or_none(flipagram.get('duration'), 1000),
+            'thumbnails': thumbnails,
+            'timestamp': unified_timestamp(flipagram.get('iso8601Created')),
+            'uploader': user.get('name'),
+            'uploader_id': user.get('username'),
+            'creator': user.get('name'),
+            'view_count': int_or_none(counts.get('plays')),
+            'like_count': int_or_none(counts.get('likes')),
+            'repost_count': int_or_none(counts.get('reflips')),
+            'comment_count': int_or_none(counts.get('comments')),
+            'comments': comments,
+            'formats': formats,
+        }

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'
+    _VALID_URL = r'https?://(?:www\.)?formula1\.com/(?:content/fom-website/)?en/video/\d{4}/\d{1,2}/(?P<id>.+?)\.html'
-    _TEST = {
+    _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/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/francetv.py

@@ -14,7 +14,10 @@ from ..utils import (
     parse_duration,
     determine_ext,
 )
-from .dailymotion import DailymotionCloudIE
+from .dailymotion import (
+    DailymotionIE,
+    DailymotionCloudIE,
+)
 
 
 class FranceTVBaseInfoExtractor(InfoExtractor):
@@ -128,7 +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',
@@ -188,6 +191,24 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        # Dailymotion embed
+        'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html',
+        'md5': 'ee7f1828f25a648addc90cb2687b1f12',
+        'info_dict': {
+            'id': 'x4iiko0',
+            'ext': 'mp4',
+            'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
+            'description': 'Au lendemain de la victoire du "oui" au référendum sur l\'aéroport de Notre-Dame-des-Landes, l\'ancienne ministre écologiste est l\'invitée de Patrick Cohen. Plus d\'info : https://www.franceinter.fr/emissions/le-7-9/le-7-9-27-juin-2016',
+            'timestamp': 1467011958,
+            'upload_date': '20160627',
+            'uploader': 'France Inter',
+            '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):
@@ -197,7 +218,13 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
 
         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
         if dmcloud_url:
-            return self.url_result(dmcloud_url, 'DailymotionCloud')
+            return self.url_result(dmcloud_url, DailymotionCloudIE.ie_key())
+
+        dailymotion_urls = DailymotionIE._extract_urls(webpage)
+        if dailymotion_urls:
+            return self.playlist_result([
+                self.url_result(dailymotion_url, DailymotionIE.ie_key())
+                for dailymotion_url in dailymotion_urls])
 
         video_id, catalogue = self._search_regex(
             (r'id-video=([^@]+@[^"]+)',

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/gamespot.py

@@ -28,10 +28,13 @@ class GameSpotIE(OnceIE):
         'url': 'http://www.gamespot.com/videos/the-witcher-3-wild-hunt-xbox-one-now-playing/2300-6424837/',
         'info_dict': {
             'id': 'gs-2300-6424837',
-            'ext': 'flv',
+            'ext': 'mp4',
-            'title': 'The Witcher 3: Wild Hunt [Xbox ONE]  - Now Playing',
+            'title': 'Now Playing - The Witcher 3: Wild Hunt',
             'description': 'Join us as we take a look at the early hours of The Witcher 3: Wild Hunt and more.',
         },
+        'params': {
+            'skip_download': True,  # m3u8 downloads
+        },
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/generic.py

@@ -49,7 +49,10 @@ from .pornhub import PornHubIE
 from .xhamster import XHamsterEmbedIE
 from .tnaflix import TNAFlixNetworkEmbedIE
 from .vimeo import VimeoIE
-from .dailymotion import DailymotionCloudIE
+from .dailymotion import (
+    DailymotionIE,
+    DailymotionCloudIE,
+)
 from .onionstudios import OnionStudiosIE
 from .viewlift import ViewLiftEmbedIE
 from .screenwavemedia import ScreenwaveMediaIE
@@ -59,6 +62,7 @@ from .videomore import VideomoreIE
 from .googledrive import GoogleDriveIE
 from .jwplatform import JWPlatformIE
 from .digiteka import DigitekaIE
+from .arkena import ArkenaIE
 from .instagram import InstagramIE
 from .liveleak import LiveLeakIE
 from .threeqsdn import ThreeQSDNIE
@@ -67,6 +71,7 @@ from .vessel import VesselIE
 from .kaltura import KalturaIE
 from .eagleplatform import EaglePlatformIE
 from .facebook import FacebookIE
+from .soundcloud import SoundcloudIE
 
 
 class GenericIE(InfoExtractor):
@@ -470,7 +475,7 @@ class GenericIE(InfoExtractor):
             'url': 'http://www.vestifinance.ru/articles/25753',
             'info_dict': {
                 'id': '25753',
-                'title': 'Вести Экономика ― Прямые трансляции с Форума-выставки "Госзаказ-2013"',
+                'title': 'Прямые трансляции с Форума-выставки "Госзаказ-2013"',
             },
             'playlist': [{
                 'info_dict': {
@@ -637,6 +642,8 @@ class GenericIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Key and Peele|October 10, 2012|2|203|Liam Neesons - Uncensored',
                 'description': 'Two valets share their love for movie star Liam Neesons.',
+                'timestamp': 1349922600,
+                'upload_date': '20121011',
             },
         },
         # YouTube embed via <data-embed-url="">
@@ -778,6 +785,15 @@ class GenericIE(InfoExtractor):
                 'upload_date': '20141029',
             }
         },
+        # Soundcloud multiple embeds
+        {
+            'url': 'http://www.guitarplayer.com/lessons/1014/legato-workout-one-hour-to-more-fluid-performance---tab/52809',
+            'info_dict': {
+                'id': '52809',
+                'title': 'Guitar Essentials: Legato Workout—One-Hour to Fluid Performance  | TAB + AUDIO',
+            },
+            'playlist_mincount': 7,
+        },
         # Livestream embed
         {
             'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
@@ -853,6 +869,7 @@ class GenericIE(InfoExtractor):
                 'description': 'md5:601cb790edd05908957dae8aaa866465',
                 'upload_date': '20150220',
             },
+            'skip': 'All The Daily Show URLs now redirect to http://www.cc.com/shows/',
         },
         # jwplayer YouTube
         {
@@ -1246,6 +1263,20 @@ class GenericIE(InfoExtractor):
                 'uploader': 'www.hudl.com',
             },
         },
+        # twitter:player:stream embed
+        {
+            'url': 'http://www.rtl.be/info/video/589263.aspx?CategoryID=288',
+            'info_dict': {
+                'id': 'master',
+                'ext': 'mp4',
+                'title': 'Une nouvelle espèce de dinosaure découverte en Argentine',
+                'uploader': 'www.rtl.be',
+            },
+            'params': {
+                # m3u8 downloads
+                'skip_download': True,
+            },
+        },
         # twitter:player embed
         {
             'url': 'http://www.theatlantic.com/video/index/484130/what-do-black-holes-sound-like/',
@@ -1295,6 +1326,70 @@ class GenericIE(InfoExtractor):
                 'uploader': 'cylus cyrus',
             },
         },
+        {
+            # video stored on custom kaltura server
+            'url': 'http://www.expansion.com/multimedia/videos.html?media=EQcM30NHIPv',
+            'md5': '537617d06e64dfed891fa1593c4b30cc',
+            'info_dict': {
+                'id': '0_1iotm5bh',
+                'ext': 'mp4',
+                'title': 'Elecciones británicas: 5 lecciones para Rajoy',
+                'description': 'md5:435a89d68b9760b92ce67ed227055f16',
+                'uploader_id': 'videos.expansion@el-mundo.net',
+                'upload_date': '20150429',
+                'timestamp': 1430303472,
+            },
+            'add_ie': ['Kaltura'],
+        },
+        {
+            # Non-standard Vimeo embed
+            'url': 'https://openclassrooms.com/courses/understanding-the-web',
+            'md5': '64d86f1c7d369afd9a78b38cbb88d80a',
+            'info_dict': {
+                'id': '148867247',
+                'ext': 'mp4',
+                'title': 'Understanding the web - Teaser',
+                'description': 'This is "Understanding the web - Teaser" by openclassrooms on Vimeo, the home for high quality videos and the people who love them.',
+                'upload_date': '20151214',
+                'uploader': 'OpenClassrooms',
+                'uploader_id': 'openclassrooms',
+            },
+            'add_ie': ['Vimeo'],
+        },
+        {
+            'url': 'https://support.arkena.com/display/PLAY/Ways+to+embed+your+video',
+            'md5': 'b96f2f71b359a8ecd05ce4e1daa72365',
+            'info_dict': {
+                'id': 'b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe',
+                'ext': 'mp4',
+                'title': 'Big Buck Bunny',
+                'description': 'Royalty free test video',
+                'timestamp': 1432816365,
+                'upload_date': '20150528',
+                'is_live': False,
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'add_ie': [ArkenaIE.ie_key()],
+        },
+        # {
+        #     # TODO: find another test
+        #     # http://schema.org/VideoObject
+        #     'url': 'https://flipagram.com/f/nyvTSJMKId',
+        #     'md5': '888dcf08b7ea671381f00fab74692755',
+        #     'info_dict': {
+        #         'id': 'nyvTSJMKId',
+        #         'ext': 'mp4',
+        #         'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
+        #         'description': '#love for cats.',
+        #         'timestamp': 1461244995,
+        #         'upload_date': '20160421',
+        #     },
+        #     'params': {
+        #         'force_generic_extractor': True,
+        #     },
+        # }
     ]
 
     def report_following_redirect(self, new_url):
@@ -1658,12 +1753,9 @@ class GenericIE(InfoExtractor):
         if matches:
             return _playlist_from_matches(matches, lambda m: m[-1])
 
-        # Look for embedded Dailymotion player
+        matches = DailymotionIE._extract_urls(webpage)
-        matches = re.findall(
-            r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
         if matches:
-            return _playlist_from_matches(
+            return _playlist_from_matches(matches)
-                matches, lambda m: unescapeHTML(m[1]))
 
         # Look for embedded Dailymotion playlist player (#3822)
         m = re.search(
@@ -1917,12 +2009,9 @@ class GenericIE(InfoExtractor):
             return self.url_result(myvi_url)
 
         # Look for embedded soundcloud player
-        mobj = re.search(
+        soundcloud_urls = SoundcloudIE._extract_urls(webpage)
-            r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
+        if soundcloud_urls:
-            webpage)
+            return _playlist_from_matches(soundcloud_urls, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
-        if mobj is not None:
-            url = unescapeHTML(mobj.group('url'))
-            return self.url_result(url)
 
         # Look for embedded mtvservices player
         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
@@ -2085,6 +2174,11 @@ class GenericIE(InfoExtractor):
         if digiteka_url:
             return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key())
 
+        # Look for Arkena embeds
+        arkena_url = ArkenaIE._extract_url(webpage)
+        if arkena_url:
+            return self.url_result(arkena_url, ArkenaIE.ie_key())
+
         # Look for Limelight embeds
         mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
         if mobj:
@@ -2113,6 +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:
@@ -2137,10 +2239,18 @@ class GenericIE(InfoExtractor):
                 'uploader': video_uploader,
             }
 
-        # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser
+        # Looking for http://schema.org/VideoObject
-        embed_url = self._html_search_meta('twitter:player', webpage, default=None)
+        json_ld = self._search_json_ld(
-        if embed_url:
+            webpage, video_id, default={}, expected_type='VideoObject')
-            return self.url_result(embed_url)
+        if json_ld.get('url'):
+            info_dict.update({
+                'title': video_title or info_dict['title'],
+                'description': video_description,
+                'thumbnail': video_thumbnail,
+                'age_limit': age_limit
+            })
+            info_dict.update(json_ld)
+            return info_dict
 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
@@ -2185,6 +2295,9 @@ class GenericIE(InfoExtractor):
                 r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
         if not found:
             # Try to find twitter cards info
+            # twitter:player:stream should be checked before twitter:player since
+            # it is expected to contain a raw stream (see
+            # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
             found = filter_video(re.findall(
                 r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
         if not found:
@@ -2218,6 +2331,15 @@ class GenericIE(InfoExtractor):
                     '_type': 'url',
                     'url': new_url,
                 }
+
+        if not found:
+            # twitter:player is a https URL to iframe player that may or may not
+            # be supported by youtube-dl thus this is checked the very last (see
+            # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser)
+            embed_url = self._html_search_meta('twitter:player', webpage, default=None)
+            if embed_url:
+                return self.url_result(embed_url)
+
         if not found:
             raise UnsupportedError(url)
 

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/instagram.py

@@ -36,7 +36,6 @@ class InstagramIE(InfoExtractor):
         'info_dict': {
             'id': 'BA-pQFBG8HZ',
             'ext': 'mp4',
-            'uploader_id': 'britneyspears',
             'title': 'Video by britneyspears',
             'thumbnail': 're:^https?://.*\.jpg',
             'timestamp': 1453760977,

youtube_dl/extractor/iqiyi.py

@@ -165,7 +165,7 @@ class IqiyiIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
-        'md5': '5b0591f55961117155430b5d544fdb01',
+        # MD5 checksum differs on my machine and Travis CI
         'info_dict': {
             'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
             'ext': 'mp4',
@@ -293,14 +293,10 @@ class IqiyiIE(InfoExtractor):
             't': tm,
         }
 
-        headers = {}
-        cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-        if cn_verification_proxy:
-            headers['Ytdl-request-proxy'] = cn_verification_proxy
         return self._download_json(
             'http://cache.m.iqiyi.com/jp/tmts/%s/%s/' % (tvid, video_id),
             video_id, transform_source=lambda s: remove_start(s, 'var tvInfoJs='),
-            query=params, headers=headers)
+            query=params, headers=self.geo_verification_headers())
 
     def _extract_playlist(self, webpage):
         PAGE_SIZE = 50

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,
 )
 
 
@@ -28,74 +30,84 @@ 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:
+                video_data['sources'] = [video_data]
 
-        # JWPlayer backward compatibility: flattened sources
+            formats = []
-        # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35
+            for source in video_data['sources']:
-        if 'sources' not in video_data:
+                source_url = self._proto_relative_url(source['file'])
-            video_data['sources'] = [video_data]
+                if base_url:
-
+                    source_url = compat_urlparse.urljoin(base_url, source_url)
-        formats = []
+                source_type = source.get('type') or ''
-        for source in video_data['sources']:
+                ext = mimetype2ext(source_type) or determine_ext(source_url)
-            source_url = self._proto_relative_url(source['file'])
+                if source_type == 'hls' or ext == 'm3u8':
-            source_type = source.get('type') or ''
+                    formats.extend(self._extract_m3u8_formats(
-            if source_type in ('application/vnd.apple.mpegurl', 'hls') or determine_ext(source_url) == 'm3u8':
+                        source_url, video_id, 'mp4', 'm3u8_native', m3u8_id=m3u8_id, fatal=False))
-                formats.extend(self._extract_m3u8_formats(
+                # https://github.com/jwplayer/jwplayer/blob/master/src/js/providers/default.js#L67
-                    source_url, video_id, 'mp4', 'm3u8_native', m3u8_id=m3u8_id, fatal=False))
+                elif source_type.startswith('audio') or ext in ('oga', 'aac', 'mp3', 'mpeg', 'vorbis'):
-            elif source_type.startswith('audio'):
+                    formats.append({
-                formats.append({
+                        'url': source_url,
-                    'url': source_url,
+                        'vcodec': 'none',
-                    'vcodec': 'none',
+                        'ext': ext,
-                })
-            else:
-                a_format = {
-                    'url': source_url,
-                    'width': int_or_none(source.get('width')),
-                    'height': int_or_none(source.get('height')),
-                }
-                if source_url.startswith('rtmp'):
-                    a_format['ext'] = 'flv',
-
-                    # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as
-                    # of jwplayer.flash.swf
-                    rtmp_url_parts = re.split(
-                        r'((?:mp4|mp3|flv):)', source_url, 1)
-                    if len(rtmp_url_parts) == 3:
-                        rtmp_url, prefix, play_path = rtmp_url_parts
-                        a_format.update({
-                            'url': rtmp_url,
-                            'play_path': prefix + play_path,
-                        })
-                    if rtmp_params:
-                        a_format.update(rtmp_params)
-                formats.append(a_format)
-        self._sort_formats(formats)
-
-        subtitles = {}
-        tracks = video_data.get('tracks')
-        if tracks and isinstance(tracks, list):
-            for track in tracks:
-                if track.get('file') and track.get('kind') == 'captions':
-                    subtitles.setdefault(track.get('label') or 'en', []).append({
-                        'url': self._proto_relative_url(track['file'])
                     })
+                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',
 
-        return {
+                        # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as
-            'id': video_id,
+                        # of jwplayer.flash.swf
-            'title': video_data['title'] if require_title else video_data.get('title'),
+                        rtmp_url_parts = re.split(
-            'description': video_data.get('description'),
+                            r'((?:mp4|mp3|flv):)', source_url, 1)
-            'thumbnail': self._proto_relative_url(video_data.get('image')),
+                        if len(rtmp_url_parts) == 3:
-            'timestamp': int_or_none(video_data.get('pubdate')),
+                            rtmp_url, prefix, play_path = rtmp_url_parts
-            'duration': float_or_none(jwplayer_data.get('duration')),
+                            a_format.update({
-            'subtitles': subtitles,
+                                'url': rtmp_url,
-            'formats': formats,
+                                'play_path': prefix + play_path,
-        }
+                            })
+                        if rtmp_params:
+                            a_format.update(rtmp_params)
+                    formats.append(a_format)
+            self._sort_formats(formats)
+
+            subtitles = {}
+            tracks = video_data.get('tracks')
+            if tracks and isinstance(tracks, list):
+                for track in tracks:
+                    if track.get('file') and track.get('kind') == 'captions':
+                        subtitles.setdefault(track.get('label') or 'en', []).append({
+                            'url': self._proto_relative_url(track['file'])
+                        })
+
+            entries.append({
+                'id': video_id,
+                'title': video_data['title'] if require_title else video_data.get('title'),
+                'description': video_data.get('description'),
+                'thumbnail': self._proto_relative_url(video_data.get('image')),
+                'timestamp': int_or_none(video_data.get('pubdate')),
+                '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

@@ -6,7 +6,6 @@ import base64
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse_urlencode,
     compat_urlparse,
     compat_parse_qs,
 )
@@ -15,6 +14,7 @@ from ..utils import (
     ExtractorError,
     int_or_none,
     unsmuggle_url,
+    smuggle_url,
 )
 
 
@@ -34,7 +34,8 @@ class KalturaIE(InfoExtractor):
                         )(?:/(?P<path>[^?]+))?(?:\?(?P<query>.*))?
                 )
                 '''
-    _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?'
+    _SERVICE_URL = 'http://cdnapi.kaltura.com'
+    _SERVICE_BASE = '/api_v3/index.php'
     _TESTS = [
         {
             'url': 'kaltura:269692:1_1jc2y3e4',
@@ -61,6 +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,
         }
     ]
 
@@ -88,18 +94,26 @@ class KalturaIE(InfoExtractor):
                     (?P<q3>["\'])(?P<id>.+?)(?P=q3)
                 ''', webpage))
         if mobj:
-            return 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict()
+            embed_info = mobj.groupdict()
+            url = 'kaltura:%(partner_id)s:%(id)s' % embed_info
+            escaped_pid = re.escape(embed_info['partner_id'])
+            service_url = re.search(
+                r'<script[^>]+src=["\']((?:https?:)?//.+?)/p/%s/sp/%s00/embedIframeJs' % (escaped_pid, escaped_pid),
+                webpage)
+            if service_url:
+                url = smuggle_url(url, {'service_url': service_url.group(1)})
+            return url
 
-    def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
+    def _kaltura_api_call(self, video_id, actions, service_url=None, *args, **kwargs):
         params = actions[0]
         if len(actions) > 1:
             for i, a in enumerate(actions[1:], start=1):
                 for k, v in a.items():
                     params['%d:%s' % (i, k)] = v
 
-        query = compat_urllib_parse_urlencode(params)
+        data = self._download_json(
-        url = self._API_BASE + query
+            (service_url or self._SERVICE_URL) + self._SERVICE_BASE,
-        data = self._download_json(url, video_id, *args, **kwargs)
+            video_id, query=params, *args, **kwargs)
 
         status = data if len(actions) == 1 else data[0]
         if status.get('objectType') == 'KalturaAPIException':
@@ -108,7 +122,7 @@ class KalturaIE(InfoExtractor):
 
         return data
 
-    def _get_kaltura_signature(self, video_id, partner_id):
+    def _get_kaltura_signature(self, video_id, partner_id, service_url=None):
         actions = [{
             'apiVersion': '3.1',
             'expiry': 86400,
@@ -118,10 +132,9 @@ class KalturaIE(InfoExtractor):
             'widgetId': '_%s' % partner_id,
         }]
         return self._kaltura_api_call(
-            video_id, actions, note='Downloading Kaltura signature')['ks']
+            video_id, actions, service_url, note='Downloading Kaltura signature')['ks']
 
-    def _get_video_info(self, video_id, partner_id):
+    def _get_video_info(self, video_id, partner_id, service_url=None):
-        signature = self._get_kaltura_signature(video_id, partner_id)
         actions = [
             {
                 'action': 'null',
@@ -129,22 +142,34 @@ 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(
-            video_id, actions, note='Downloading video info JSON')
+            video_id, actions, service_url, note='Downloading video info JSON')
 
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
@@ -152,8 +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)
+            _, 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:
@@ -172,7 +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)
@@ -201,12 +227,17 @@ class KalturaIE(InfoExtractor):
                 unsigned_url += '?referrer=%s' % referrer
             return unsigned_url
 
+        data_url = info['dataUrl']
+        if '/flvclipper/' in data_url:
+            data_url = re.sub(r'/flvclipper/.*', '/serveFlavor', data_url)
+
         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' % (info['dataUrl'], f['id']))
+            video_url = sign_url(
+                '%s/flavorId/%s' % (data_url, f['id']))
             formats.append({
                 'format_id': '%(fileExt)s-%(bitrate)s' % f,
                 'ext': f.get('fileExt'),
@@ -219,17 +250,31 @@ class KalturaIE(InfoExtractor):
                 'width': int_or_none(f.get('width')),
                 'url': video_url,
             })
-        m3u8_url = sign_url(info['dataUrl'].replace('format/url', 'format/applehttp'))
+        if '/playManifest/' in data_url:
-        formats.extend(self._extract_m3u8_formats(
+            m3u8_url = sign_url(data_url.replace(
-            m3u8_url, entry_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                'format/url', 'format/applehttp'))
+            formats.extend(self._extract_m3u8_formats(
+                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/kamcord.py

@@ -0,0 +1,71 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    qualities,
+)
+
+
+class KamcordIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?kamcord\.com/v/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://www.kamcord.com/v/hNYRduDgWb4',
+        'md5': 'c3180e8a9cfac2e86e1b88cb8751b54c',
+        'info_dict': {
+            'id': 'hNYRduDgWb4',
+            'ext': 'mp4',
+            'title': 'Drinking Madness',
+            'uploader': 'jacksfilms',
+            'uploader_id': '3044562',
+            'view_count': int,
+            'like_count': int,
+            'comment_count': int,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video = self._parse_json(
+            self._search_regex(
+                r'window\.__props\s*=\s*({.+?});?(?:\n|\s*</script)',
+                webpage, 'video'),
+            video_id)['video']
+
+        title = video['title']
+
+        formats = self._extract_m3u8_formats(
+            video['play']['hls'], video_id, 'mp4', entry_protocol='m3u8_native')
+        self._sort_formats(formats)
+
+        uploader = video.get('user', {}).get('username')
+        uploader_id = video.get('user', {}).get('id')
+
+        view_count = int_or_none(video.get('viewCount'))
+        like_count = int_or_none(video.get('heartCount'))
+        comment_count = int_or_none(video.get('messageCount'))
+
+        preference_key = qualities(('small', 'medium', 'large'))
+
+        thumbnails = [{
+            'url': thumbnail_url,
+            'id': thumbnail_id,
+            'preference': preference_key(thumbnail_id),
+        } for thumbnail_id, thumbnail_url in (video.get('thumbnail') or {}).items()
+            if isinstance(thumbnail_id, compat_str) and isinstance(thumbnail_url, compat_str)]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'view_count': view_count,
+            'like_count': like_count,
+            'comment_count': comment_count,
+            'thumbnails': thumbnails,
+            'formats': formats,
+        }

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,
@@ -26,11 +27,6 @@ class KuwoBaseIE(InfoExtractor):
     def _get_formats(self, song_id, tolerate_ip_deny=False):
         formats = []
         for file_format in self._FORMATS:
-            headers = {}
-            cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-            if cn_verification_proxy:
-                headers['Ytdl-request-proxy'] = cn_verification_proxy
-
             query = {
                 'format': file_format['ext'],
                 'br': file_format.get('br', ''),
@@ -42,7 +38,7 @@ class KuwoBaseIE(InfoExtractor):
             song_url = self._download_webpage(
                 'http://antiserver.kuwo.cn/anti.s',
                 song_id, note='Download %s url info' % file_format['format'],
-                query=query, headers=headers,
+                query=query, headers=self.geo_verification_headers(),
             )
 
             if song_url == 'IPDeny' and not tolerate_ip_deny:
@@ -247,8 +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(
+                self.url_result(compat_urlparse.urljoin(url, song_url), 'Kuwo')
-                    r'<div[^>]+class="name"><a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+)',
+                for song_url in re.findall(
+                    r'<div[^>]+class="name"><a[^>]+href="(/yinyue/\d+)',
                     webpage)
             ]
 

youtube_dl/extractor/la7.py

@@ -3,8 +3,8 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
     js_to_json,
+    smuggle_url,
 )
 
 
@@ -18,13 +18,16 @@ class LA7IE(InfoExtractor):
     _TESTS = [{
         # 'src' is a plain URL
         'url': 'http://www.la7.it/crozza/video/inccool8-02-10-2015-163722',
-        'md5': '6054674766e7988d3e02f2148ff92180',
+        'md5': '8b613ffc0c4bf9b9e377169fc19c214c',
         'info_dict': {
             'id': 'inccool8-02-10-2015-163722',
             'ext': 'mp4',
             'title': 'Inc.Cool8',
             'description': 'Benvenuti nell\'incredibile mondo della INC. COOL. 8. dove “INC.” sta per “Incorporated” “COOL” sta per “fashion” ed Eight sta per il gesto  atletico',
             'thumbnail': 're:^https?://.*',
+            'uploader_id': 'kdla7pillole@iltrovatore.it',
+            'timestamp': 1443814869,
+            'upload_date': '20151002',
         },
     }, {
         # 'src' is a dictionary
@@ -49,26 +52,14 @@ class LA7IE(InfoExtractor):
             self._search_regex(r'videoLa7\(({[^;]+})\);', webpage, 'player data'),
             video_id, transform_source=js_to_json)
 
-        source = player_data['src']
-        source_urls = source.values() if isinstance(source, dict) else [source]
-
-        formats = []
-        for source_url in source_urls:
-            ext = determine_ext(source_url)
-            if ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    source_url, video_id, ext='mp4',
-                    entry_protocol='m3u8_native', m3u8_id='hls'))
-            else:
-                formats.append({
-                    'url': source_url,
-                })
-        self._sort_formats(formats)
-
         return {
+            '_type': 'url_transparent',
+            'url': smuggle_url('kaltura:103:%s' % player_data['vid'], {
+                'service_url': 'http://kdam.iltrovatore.it',
+            }),
             'id': video_id,
             'title': player_data['title'],
             'description': self._og_search_description(webpage, default=None),
             'thumbnail': player_data.get('poster'),
-            'formats': formats,
+            'ie_key': 'Kaltura',
         }

youtube_dl/extractor/lcp.py

@@ -0,0 +1,90 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from .arkena import ArkenaIE
+
+
+class LcpPlayIE(ArkenaIE):
+    _VALID_URL = r'https?://play\.lcp\.fr/embed/(?P<id>[^/]+)/(?P<account_id>[^/]+)/[^/]+/[^/]+'
+    _TESTS = [{
+        'url': 'http://play.lcp.fr/embed/327336/131064/darkmatter/0',
+        'md5': 'b8bd9298542929c06c1c15788b1f277a',
+        'info_dict': {
+            'id': '327336',
+            'ext': 'mp4',
+            'title': '327336',
+            'timestamp': 1456391602,
+            'upload_date': '20160225',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+
+class LcpIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?lcp\.fr/(?:[^/]+/)*(?P<id>[^/]+)'
+
+    _TESTS = [{
+        # arkena embed
+        'url': 'http://www.lcp.fr/la-politique-en-video/schwartzenberg-prg-preconise-francois-hollande-de-participer-une-primaire',
+        'md5': 'b8bd9298542929c06c1c15788b1f277a',
+        'info_dict': {
+            'id': 'd56d03e9',
+            'ext': 'mp4',
+            'title': 'Schwartzenberg (PRG) préconise à François Hollande de participer à une primaire à gauche',
+            'description': 'md5:96ad55009548da9dea19f4120c6c16a8',
+            'timestamp': 1456488895,
+            'upload_date': '20160226',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # dailymotion live stream
+        'url': 'http://www.lcp.fr/le-direct',
+        'info_dict': {
+            'id': 'xji3qy',
+            'ext': 'mp4',
+            'title': 'La Chaine Parlementaire (LCP), Live TNT',
+            'description': 'md5:5c69593f2de0f38bd9a949f2c95e870b',
+            'uploader': 'LCP',
+            'uploader_id': 'xbz33d',
+            'timestamp': 1308923058,
+            'upload_date': '20110624',
+        },
+        'params': {
+            # m3u8 live stream
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.lcp.fr/emissions/277792-les-volontaires',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        play_url = self._search_regex(
+            r'<iframe[^>]+src=(["\'])(?P<url>%s?(?:(?!\1).)*)\1' % LcpPlayIE._VALID_URL,
+            webpage, 'play iframe', default=None, group='url')
+
+        if not play_url:
+            return self.url_result(url, 'Generic')
+
+        title = self._og_search_title(webpage, default=None) or self._html_search_meta(
+            'twitter:title', webpage, fatal=True)
+        description = self._html_search_meta(
+            ('description', 'twitter:description'), webpage)
+
+        return {
+            '_type': 'url_transparent',
+            'ie_key': LcpPlayIE.ie_key(),
+            'url': play_url,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+        }

youtube_dl/extractor/leeco.py

@@ -20,10 +20,10 @@ from ..utils import (
     int_or_none,
     orderedSet,
     parse_iso8601,
-    sanitized_Request,
     str_or_none,
     url_basename,
     urshift,
+    update_url_query,
 )
 
 
@@ -90,6 +90,10 @@ class LeIE(InfoExtractor):
         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
         return _loc3_
 
+    # reversed from http://jstatic.letvcdn.com/sdk/player.js
+    def get_mms_key(self, time):
+        return self.ror(time, 8) ^ 185025305
+
     # see M3U8Encryption class in KLetvPlayer.swf
     @staticmethod
     def decrypt_m3u8(encrypted_data):
@@ -110,28 +114,7 @@ class LeIE(InfoExtractor):
 
         return bytes(_loc7_)
 
-    def _real_extract(self, url):
+    def _check_errors(self, play_json):
-        media_id = self._match_id(url)
-        page = self._download_webpage(url, media_id)
-        params = {
-            'id': media_id,
-            'platid': 1,
-            'splatid': 101,
-            'format': 1,
-            'tkey': self.calc_time_key(int(time.time())),
-            'domain': 'www.le.com'
-        }
-        play_json_req = sanitized_Request(
-            'http://api.le.com/mms/out/video/playJson?' + compat_urllib_parse_urlencode(params)
-        )
-        cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-        if cn_verification_proxy:
-            play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy)
-
-        play_json = self._download_json(
-            play_json_req,
-            media_id, 'Downloading playJson data')
-
         # Check for errors
         playstatus = play_json['playstatus']
         if playstatus['status'] == 0:
@@ -142,43 +125,99 @@ class LeIE(InfoExtractor):
                 msg = 'Generic error. flag = %d' % flag
             raise ExtractorError(msg, expected=True)
 
-        playurl = play_json['playurl']
+    def _real_extract(self, url):
+        media_id = self._match_id(url)
+        page = self._download_webpage(url, media_id)
 
-        formats = ['350', '1000', '1300', '720p', '1080p']
+        play_json_h5 = self._download_json(
-        dispatch = playurl['dispatch']
+            'http://api.le.com/mms/out/video/playJsonH5',
+            media_id, 'Downloading html5 playJson data', query={
+                'id': media_id,
+                'platid': 3,
+                'splatid': 304,
+                'format': 1,
+                'tkey': self.get_mms_key(int(time.time())),
+                'domain': 'www.le.com',
+                'tss': 'no',
+            },
+            headers=self.geo_verification_headers())
+        self._check_errors(play_json_h5)
 
-        urls = []
+        play_json_flash = self._download_json(
-        for format_id in formats:
+            'http://api.le.com/mms/out/video/playJson',
-            if format_id in dispatch:
+            media_id, 'Downloading flash playJson data', query={
-                media_url = playurl['domain'][0] + dispatch[format_id][0]
+                'id': media_id,
-                media_url += '&' + compat_urllib_parse_urlencode({
+                'platid': 1,
-                    'm3v': 1,
+                'splatid': 101,
+                'format': 1,
+                'tkey': self.calc_time_key(int(time.time())),
+                'domain': 'www.le.com',
+            },
+            headers=self.geo_verification_headers())
+        self._check_errors(play_json_flash)
+
+        def get_h5_urls(media_url, format_id):
+            location = self._download_json(
+                media_url, media_id,
+                'Download JSON metadata for format %s' % format_id, query={
                     'format': 1,
                     'expect': 3,
-                    'rateid': format_id,
+                    'tss': 'no',
-                })
+                })['location']
 
-                nodes_data = self._download_json(
+            return {
-                    media_url, media_id,
+                'http': update_url_query(location, {'tss': 'no'}),
-                    'Download JSON metadata for format %s' % format_id)
+                'hls': update_url_query(location, {'tss': 'ios'}),
+            }
 
-                req = self._request_webpage(
+        def get_flash_urls(media_url, format_id):
-                    nodes_data['nodelist'][0]['location'], media_id,
+            media_url += '&' + compat_urllib_parse_urlencode({
-                    note='Downloading m3u8 information for format %s' % format_id)
+                'm3v': 1,
+                'format': 1,
+                'expect': 3,
+                'rateid': format_id,
+            })
 
-                m3u8_data = self.decrypt_m3u8(req.read())
+            nodes_data = self._download_json(
+                media_url, media_id,
+                'Download JSON metadata for format %s' % format_id)
 
-                url_info_dict = {
+            req = self._request_webpage(
-                    'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
+                nodes_data['nodelist'][0]['location'], media_id,
-                    'ext': determine_ext(dispatch[format_id][1]),
+                note='Downloading m3u8 information for format %s' % format_id)
-                    'format_id': format_id,
-                    'protocol': 'm3u8',
-                }
 
-                if format_id[-1:] == 'p':
+            m3u8_data = self.decrypt_m3u8(req.read())
-                    url_info_dict['height'] = int_or_none(format_id[:-1])
 
-                urls.append(url_info_dict)
+            return {
+                'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
+            }
+
+        extracted_formats = []
+        formats = []
+        for play_json, get_urls in ((play_json_h5, get_h5_urls), (play_json_flash, get_flash_urls)):
+            playurl = play_json['playurl']
+            play_domain = playurl['domain'][0]
+
+            for format_id, format_data in playurl.get('dispatch', []).items():
+                if format_id in extracted_formats:
+                    continue
+                extracted_formats.append(format_id)
+
+                media_url = play_domain + format_data[0]
+                for protocol, format_url in get_urls(media_url, format_id).items():
+                    f = {
+                        'url': format_url,
+                        'ext': determine_ext(format_data[1]),
+                        'format_id': '%s-%s' % (protocol, format_id),
+                        'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
+                        'quality': int_or_none(format_id),
+                    }
+
+                    if format_id[-1:] == 'p':
+                        f['height'] = int_or_none(format_id[:-1])
+
+                    formats.append(f)
+        self._sort_formats(formats, ('height', 'quality', 'format_id'))
 
         publish_time = parse_iso8601(self._html_search_regex(
             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
@@ -187,7 +226,7 @@ class LeIE(InfoExtractor):
 
         return {
             'id': media_id,
-            'formats': urls,
+            'formats': formats,
             'title': playurl['title'],
             'thumbnail': playurl['pic'],
             'description': description,

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/lynda.py

@@ -100,7 +100,7 @@ class LyndaIE(LyndaBaseIE):
 
     _TESTS = [{
         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
-        'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
+        # md5 is unstable
         'info_dict': {
             'id': '114408',
             'ext': 'mp4',

youtube_dl/extractor/metacafe.py

@@ -11,13 +11,14 @@ from ..utils import (
     determine_ext,
     ExtractorError,
     int_or_none,
-    sanitized_Request,
     urlencode_postdata,
+    get_element_by_attribute,
+    mimetype2ext,
 )
 
 
 class MetacafeIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
+    _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/(?P<video_id>[^/]+)/(?P<display_id>[^/?#]+)'
     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
     IE_NAME = 'metacafe'
@@ -47,6 +48,7 @@ class MetacafeIE(InfoExtractor):
                 'uploader': 'ign',
                 'description': 'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.',
             },
+            'skip': 'Page is temporarily unavailable.',
         },
         # AnyClip video
         {
@@ -55,8 +57,8 @@ class MetacafeIE(InfoExtractor):
                 'id': 'an-dVVXnuY7Jh77J',
                 'ext': 'mp4',
                 'title': 'The Andromeda Strain (1971): Stop the Bomb Part 3',
-                'uploader': 'anyclip',
+                'uploader': 'AnyClip',
-                'description': 'md5:38c711dd98f5bb87acf973d573442e67',
+                'description': 'md5:cbef0460d31e3807f6feb4e7a5952e5b',
             },
         },
         # age-restricted video
@@ -110,28 +112,25 @@ class MetacafeIE(InfoExtractor):
     def report_disclaimer(self):
         self.to_screen('Retrieving disclaimer')
 
-    def _real_initialize(self):
+    def _confirm_age(self):
         # Retrieve disclaimer
         self.report_disclaimer()
         self._download_webpage(self._DISCLAIMER, None, False, 'Unable to retrieve disclaimer')
 
         # Confirm age
-        disclaimer_form = {
-            'filters': '0',
-            'submit': "Continue - I'm over 18",
-        }
-        request = sanitized_Request(self._FILTER_POST, urlencode_postdata(disclaimer_form))
-        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         self.report_age_confirmation()
-        self._download_webpage(request, None, False, 'Unable to confirm age')
+        self._download_webpage(
+            self._FILTER_POST, None, False, 'Unable to confirm age',
+            data=urlencode_postdata({
+                'filters': '0',
+                'submit': "Continue - I'm over 18",
+            }), headers={
+                'Content-Type': 'application/x-www-form-urlencoded',
+            })
 
     def _real_extract(self, url):
         # Extract id and simplified title from URL
-        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
-        if mobj is None:
-            raise ExtractorError('Invalid URL: %s' % url)
-
-        video_id = mobj.group(1)
 
         # the video may come from an external site
         m_external = re.match('^(\w{2})-(.*)$', video_id)
@@ -144,15 +143,24 @@ class MetacafeIE(InfoExtractor):
             if prefix == 'cb':
                 return self.url_result('theplatform:%s' % ext_id, 'ThePlatform')
 
-        # Retrieve video webpage to extract further information
+        # self._confirm_age()
-        req = sanitized_Request('http://www.metacafe.com/watch/%s/' % video_id)
 
         # AnyClip videos require the flashversion cookie so that we get the link
         # to the mp4 file
-        mobj_an = re.match(r'^an-(.*?)$', video_id)
+        headers = {}
-        if mobj_an:
+        if video_id.startswith('an-'):
-            req.headers['Cookie'] = 'flashVersion=0;'
+            headers['Cookie'] = 'flashVersion=0;'
-        webpage = self._download_webpage(req, video_id)
+
+        # Retrieve video webpage to extract further information
+        webpage = self._download_webpage(url, video_id, headers=headers)
+
+        error = get_element_by_attribute(
+            'class', 'notfound-page-title', webpage)
+        if error:
+            raise ExtractorError(error, expected=True)
+
+        video_title = self._html_search_meta(
+            ['og:title', 'twitter:title'], webpage, 'title', default=None) or self._search_regex(r'<h1>(.*?)</h1>', webpage, 'title')
 
         # Extract URL, uploader and title from webpage
         self.report_extraction(video_id)
@@ -216,20 +224,40 @@ class MetacafeIE(InfoExtractor):
                         'player_url': player_url,
                         'ext': play_path.partition(':')[0],
                     })
+        if video_url is None:
+            flashvars = self._parse_json(self._search_regex(
+                r'flashvars\s*=\s*({.*});', webpage, 'flashvars',
+                default=None), video_id, fatal=False)
+            if flashvars:
+                video_url = []
+                for source in flashvars.get('sources'):
+                    source_url = source.get('src')
+                    if not source_url:
+                        continue
+                    ext = mimetype2ext(source.get('type')) or determine_ext(source_url)
+                    if ext == 'm3u8':
+                        video_url.extend(self._extract_m3u8_formats(
+                            source_url, video_id, 'mp4',
+                            'm3u8_native', m3u8_id='hls', fatal=False))
+                    else:
+                        video_url.append({
+                            'url': source_url,
+                            'ext': ext,
+                        })
 
         if video_url is None:
             raise ExtractorError('Unsupported video type')
 
-        video_title = self._html_search_regex(
+        description = self._html_search_meta(
-            r'(?im)<title>(.*) - Video</title>', webpage, 'title')
+            ['og:description', 'twitter:description', 'description'],
-        description = self._og_search_description(webpage)
+            webpage, 'title', fatal=False)
-        thumbnail = self._og_search_thumbnail(webpage)
+        thumbnail = self._html_search_meta(
+            ['og:image', 'twitter:image'], webpage, 'title', fatal=False)
         video_uploader = self._html_search_regex(
             r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
             webpage, 'uploader nickname', fatal=False)
         duration = int_or_none(
-            self._html_search_meta('video:duration', webpage))
+            self._html_search_meta('video:duration', webpage, default=None))
-
         age_limit = (
             18
             if re.search(r'(?:"contentRating":|"rating",)"restricted"', webpage)
@@ -242,10 +270,11 @@ class MetacafeIE(InfoExtractor):
                 'url': video_url,
                 'ext': video_ext,
             }]
-
         self._sort_formats(formats)
+
         return {
             'id': video_id,
+            'display_id': display_id,
             'description': description,
             'uploader': video_uploader,
             'title': video_title,

youtube_dl/extractor/mgtv.py

@@ -9,7 +9,7 @@ class MGTVIE(InfoExtractor):
     _VALID_URL = r'https?://www\.mgtv\.com/v/(?:[^/]+/)*(?P<id>\d+)\.html'
     IE_DESC = '芒果TV'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
         'md5': '1bdadcf760a0b90946ca68ee9a2db41a',
         'info_dict': {
@@ -20,13 +20,18 @@ class MGTVIE(InfoExtractor):
             'duration': 7461,
             'thumbnail': 're:^https?://.*\.jpg$',
         },
-    }
+    }, {
+        # no tbr extracted from stream_url
+        'url': 'http://www.mgtv.com/v/1/1/f/3324755.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         api_data = self._download_json(
             'http://v.api.mgtv.com/player/video', video_id,
-            query={'video_id': video_id})['data']
+            query={'video_id': video_id},
+            headers=self.geo_verification_headers())['data']
         info = api_data['info']
 
         formats = []
@@ -40,7 +45,8 @@ class MGTVIE(InfoExtractor):
             def extract_format(stream_url, format_id, idx, query={}):
                 format_info = self._download_json(
                     stream_url, video_id,
-                    note='Download video info for format %s' % format_id or '#%d' % idx, query=query)
+                    note='Download video info for format %s' % (format_id or '#%d' % idx),
+                    query=query)
                 return {
                     'format_id': format_id,
                     'url': format_info['info'],

youtube_dl/extractor/miomio.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import random
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     xpath_text,
     int_or_none,
@@ -18,13 +19,16 @@ class MioMioIE(InfoExtractor):
     _TESTS = [{
         # "type=video" in flashvars
         'url': 'http://www.miomio.tv/watch/cc88912/',
-        'md5': '317a5f7f6b544ce8419b784ca8edae65',
         'info_dict': {
             'id': '88912',
             'ext': 'flv',
             'title': '【SKY】字幕 铠武昭和VS平成 假面骑士大战FEAT战队 魔星字幕组 字幕',
             'duration': 5923,
         },
+        'params': {
+            # The server provides broken file
+            'skip_download': True,
+        }
     }, {
         'url': 'http://www.miomio.tv/watch/cc184024/',
         'info_dict': {
@@ -32,7 +36,7 @@ class MioMioIE(InfoExtractor):
             'title': '《动漫同人插画绘制》',
         },
         'playlist_mincount': 86,
-        'skip': 'This video takes time too long for retrieving the URL',
+        'skip': 'Unable to load videos',
     }, {
         'url': 'http://www.miomio.tv/watch/cc173113/',
         'info_dict': {
@@ -40,20 +44,23 @@ class MioMioIE(InfoExtractor):
             'title': 'The New Macbook 2015 上手试玩与简评'
         },
         'playlist_mincount': 2,
+        'skip': 'Unable to load videos',
+    }, {
+        # new 'h5' player
+        'url': 'http://www.miomio.tv/watch/cc273295/',
+        'md5': '',
+        'info_dict': {
+            'id': '273295',
+            'ext': 'mp4',
+            'title': 'アウト×デラックス 20160526',
+        },
+        'params': {
+            # intermittent HTTP 500
+            'skip_download': True,
+        },
     }]
 
-    def _real_extract(self, url):
+    def _extract_mioplayer(self, webpage, video_id, title, http_headers):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        title = self._html_search_meta(
-            'description', webpage, 'title', fatal=True)
-
-        mioplayer_path = self._search_regex(
-            r'src="(/mioplayer/[^"]+)"', webpage, 'ref_path')
-
-        http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}
-
         xml_config = self._search_regex(
             r'flashvars="type=(?:sina|video)&(.+?)&',
             webpage, 'xml config')
@@ -92,10 +99,34 @@ class MioMioIE(InfoExtractor):
                 'http_headers': http_headers,
             })
 
+        return entries
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_meta(
+            'description', webpage, 'title', fatal=True)
+
+        mioplayer_path = self._search_regex(
+            r'src="(/mioplayer(?:_h5)?/[^"]+)"', webpage, 'ref_path')
+
+        if '_h5' in mioplayer_path:
+            player_url = compat_urlparse.urljoin(url, mioplayer_path)
+            player_webpage = self._download_webpage(
+                player_url, video_id,
+                note='Downloading player webpage', headers={'Referer': url})
+            entries = self._parse_html5_media_entries(player_url, player_webpage)
+            http_headers = {'Referer': player_url}
+        else:
+            http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}
+            entries = self._extract_mioplayer(webpage, video_id, title, http_headers)
+
         if len(entries) == 1:
             segment = entries[0]
             segment['id'] = video_id
             segment['title'] = title
+            segment['http_headers'] = http_headers
             return segment
 
         return {

youtube_dl/extractor/mitele.py

@@ -12,12 +12,69 @@ from ..utils import (
     get_element_by_attribute,
     int_or_none,
     remove_start,
+    extract_attributes,
+    determine_ext,
 )
 
 
-class MiTeleIE(InfoExtractor):
+class MiTeleBaseIE(InfoExtractor):
+    def _get_player_info(self, url, webpage):
+        player_data = extract_attributes(self._search_regex(
+            r'(?s)(<ms-video-player.+?</ms-video-player>)',
+            webpage, 'ms video player'))
+        video_id = player_data['data-media-id']
+        config_url = compat_urlparse.urljoin(url, player_data['data-config'])
+        config = self._download_json(
+            config_url, video_id, 'Downloading config JSON')
+        mmc_url = config['services']['mmc']
+
+        duration = None
+        formats = []
+        for m_url in (mmc_url, mmc_url.replace('/flash.json', '/html5.json')):
+            mmc = self._download_json(
+                m_url, video_id, 'Downloading mmc JSON')
+            if not duration:
+                duration = int_or_none(mmc.get('duration'))
+            for location in mmc['locations']:
+                gat = self._proto_relative_url(location.get('gat'), 'http:')
+                bas = location.get('bas')
+                loc = location.get('loc')
+                ogn = location.get('ogn')
+                if None in (gat, bas, loc, ogn):
+                    continue
+                token_data = {
+                    'bas': bas,
+                    'icd': loc,
+                    'ogn': ogn,
+                    'sta': '0',
+                }
+                media = self._download_json(
+                    '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)),
+                    video_id, 'Downloading %s JSON' % location['loc'])
+                file_ = media.get('file')
+                if not file_:
+                    continue
+                ext = determine_ext(file_)
+                if ext == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
+                        video_id, f4m_id='hds', fatal=False))
+                elif ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        file_, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'thumbnail': player_data.get('data-poster') or config.get('poster', {}).get('imageUrl'),
+            'duration': duration,
+        }
+
+
+class MiTeleIE(MiTeleBaseIE):
     IE_DESC = 'mitele.es'
-    _VALID_URL = r'https?://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
+    _VALID_URL = r'https?://www\.mitele\.es/(?:[^/]+/){3}(?P<id>[^/]+)/'
 
     _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
@@ -25,7 +82,7 @@ class MiTeleIE(InfoExtractor):
         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Tor, la web invisible',
             'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
             'series': 'Diario de',
@@ -40,7 +97,7 @@ class MiTeleIE(InfoExtractor):
         'info_dict': {
             'id': 'eLZSwoEd1S3pVyUm8lc6F',
             'display_id': 'programa-226',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Cuarto Milenio - Temporada 6 - Programa 226',
             'description': 'md5:50daf9fadefa4e62d9fc866d0c015701',
             'series': 'Cuarto Milenio',
@@ -59,40 +116,7 @@ class MiTeleIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        config_url = self._search_regex(
+        info = self._get_player_info(url, webpage)
-            r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url')
-        config_url = compat_urlparse.urljoin(url, config_url)
-
-        config = self._download_json(
-            config_url, display_id, 'Downloading config JSON')
-
-        mmc = self._download_json(
-            config['services']['mmc'], display_id, 'Downloading mmc JSON')
-
-        formats = []
-        for location in mmc['locations']:
-            gat = self._proto_relative_url(location.get('gat'), 'http:')
-            bas = location.get('bas')
-            loc = location.get('loc')
-            ogn = location.get('ogn')
-            if None in (gat, bas, loc, ogn):
-                continue
-            token_data = {
-                'bas': bas,
-                'icd': loc,
-                'ogn': ogn,
-                'sta': '0',
-            }
-            media = self._download_json(
-                '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)),
-                display_id, 'Downloading %s JSON' % location['loc'])
-            file_ = media.get('file')
-            if not file_:
-                continue
-            formats.extend(self._extract_f4m_formats(
-                file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
-                display_id, f4m_id=loc))
-        self._sort_formats(formats)
 
         title = self._search_regex(
             r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>',
@@ -112,21 +136,12 @@ class MiTeleIE(InfoExtractor):
                 title = remove_start(self._search_regex(
                     r'<title>([^<]+)</title>', webpage, 'title'), 'Ver online ')
 
-        video_id = self._search_regex(
+        info.update({
-            r'data-media-id\s*=\s*"([^"]+)"', webpage,
-            'data media id', default=None) or display_id
-        thumbnail = config.get('poster', {}).get('imageUrl')
-        duration = int_or_none(mmc.get('duration'))
-
-        return {
-            'id': video_id,
             'display_id': display_id,
             'title': title,
             'description': get_element_by_attribute('class', 'text', webpage),
             'series': series,
             'season': season,
             'episode': episode,
-            'thumbnail': thumbnail,
+        })
-            'duration': duration,
+        return info
-            'formats': formats,
-        }

youtube_dl/extractor/mtv.py

@@ -15,6 +15,8 @@ from ..utils import (
     float_or_none,
     HEADRequest,
     sanitized_Request,
+    strip_or_none,
+    timeconvert,
     unescapeHTML,
     url_basename,
     RegexNotFoundError,
@@ -35,13 +37,13 @@ class MTVServicesInfoExtractor(InfoExtractor):
         return uri.split(':')[-1]
 
     # This was originally implemented for ComedyCentral, but it also works here
-    @staticmethod
+    @classmethod
-    def _transform_rtmp_url(rtmp_video_url):
+    def _transform_rtmp_url(cls, rtmp_video_url):
         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
         if not m:
-            return rtmp_video_url
+            return {'rtmp': rtmp_video_url}
         base = 'http://viacommtvstrmfs.fplive.net/'
-        return base + m.group('finalid')
+        return {'http': base + m.group('finalid')}
 
     def _get_feed_url(self, uri):
         return self._FEED_URL
@@ -85,14 +87,14 @@ class MTVServicesInfoExtractor(InfoExtractor):
                 rtmp_video_url = rendition.find('./src').text
                 if rtmp_video_url.endswith('siteunavail.png'):
                     continue
-                new_url = self._transform_rtmp_url(rtmp_video_url)
+                new_urls = self._transform_rtmp_url(rtmp_video_url)
-                formats.append({
+                formats.extend([{
                     'ext': 'flv' if new_url.startswith('rtmp') else ext,
                     'url': new_url,
-                    'format_id': rendition.get('bitrate'),
+                    'format_id': '-'.join(filter(None, [kind, rendition.get('bitrate')])),
                     'width': int(rendition.get('width')),
                     'height': int(rendition.get('height')),
-                })
+                } for kind, new_url in new_urls.items()])
             except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')
         self._sort_formats(formats)
@@ -133,7 +135,9 @@ class MTVServicesInfoExtractor(InfoExtractor):
             message += item.text
             raise ExtractorError(message, expected=True)
 
-        description = xpath_text(itemdoc, 'description')
+        description = strip_or_none(xpath_text(itemdoc, 'description'))
+
+        timestamp = timeconvert(xpath_text(itemdoc, 'pubDate'))
 
         title_el = None
         if title_el is None:
@@ -167,6 +171,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,
             'duration': float_or_none(content_el.attrib.get('duration')),
+            'timestamp': timestamp,
         }
 
     def _get_feed_query(self, uri):
@@ -185,8 +190,13 @@ class MTVServicesInfoExtractor(InfoExtractor):
         idoc = self._download_xml(
             url, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
+
+        title = xpath_text(idoc, './channel/title')
+        description = xpath_text(idoc, './channel/description')
+
         return self.playlist_result(
-            [self._get_video_info(item) for item in idoc.findall('.//item')])
+            [self._get_video_info(item) for item in idoc.findall('.//item')],
+            playlist_title=title, playlist_description=description)
 
     def _extract_mgid(self, webpage):
         try:
@@ -232,6 +242,8 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
             'ext': 'mp4',
             'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds',
             'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.',
+            'timestamp': 1400126400,
+            'upload_date': '20140515',
         },
     }
 
@@ -274,6 +286,8 @@ class MTVIE(MTVServicesInfoExtractor):
                 'ext': 'mp4',
                 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
                 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
+                'timestamp': 1352610000,
+                'upload_date': '20121111',
             },
         },
     ]
@@ -300,20 +314,6 @@ class MTVIE(MTVServicesInfoExtractor):
         return self._get_videos_info(uri)
 
 
-class MTVIggyIE(MTVServicesInfoExtractor):
-    IE_NAME = 'mtviggy.com'
-    _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
-    _TEST = {
-        'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
-        'info_dict': {
-            'id': '984696',
-            'ext': 'mp4',
-            'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
-        }
-    }
-    _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
-
-
 class MTVDEIE(MTVServicesInfoExtractor):
     IE_NAME = 'mtv.de'
     _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
@@ -321,7 +321,7 @@ class MTVDEIE(MTVServicesInfoExtractor):
         'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
         'info_dict': {
             'id': 'music_video-a50bc5f0b3aa4b3190aa',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'MusicVideo_cro-traum',
             'description': 'Cro - Traum',
         },
@@ -329,20 +329,21 @@ class MTVDEIE(MTVServicesInfoExtractor):
             # rtmp download
             'skip_download': True,
         },
+        'skip': 'Blocked at Travis CI',
     }, {
         # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
         'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen',
         'info_dict': {
             'id': 'local_playlist-f5ae778b9832cc837189',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
         },
         'params': {
             # rtmp download
             'skip_download': True,
         },
+        'skip': 'Blocked at Travis CI',
     }, {
-        # single video in pagePlaylist with different id
         'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
         'info_dict': {
             'id': 'local_playlist-4e760566473c4c8c5344',
@@ -354,6 +355,7 @@ class MTVDEIE(MTVServicesInfoExtractor):
             # rtmp download
             'skip_download': True,
         },
+        'skip': 'Das Video kann zur Zeit nicht abgespielt werden.',
     }]
 
     def _real_extract(self, url):
@@ -366,11 +368,14 @@ class MTVDEIE(MTVServicesInfoExtractor):
                 r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'),
             video_id)
 
+        def _mrss_url(item):
+            return item['mrss'] + item.get('mrssvars', '')
+
         # news pages contain single video in playlist with different id
         if len(playlist) == 1:
-            return self._get_videos_info_from_url(playlist[0]['mrss'], video_id)
+            return self._get_videos_info_from_url(_mrss_url(playlist[0]), video_id)
 
         for item in playlist:
             item_id = item.get('id')
             if item_id and compat_str(item_id) == video_id:
-                return self._get_videos_info_from_url(item['mrss'], video_id)
+                return self._get_videos_info_from_url(_mrss_url(item), video_id)

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):
+class NationalGeographicVideoIE(InfoExtractor):
-    IE_NAME = 'natgeo'
+    IE_NAME = 'natgeo:video'
     _VALID_URL = r'https?://video\.nationalgeographic\.com/.*?'
 
     _TESTS = [
@@ -62,16 +65,16 @@ class NationalGeographicIE(InfoExtractor):
         }
 
 
-class NationalGeographicChannelIE(ThePlatformIE):
+class NationalGeographicIE(ThePlatformIE):
-    IE_NAME = 'natgeo:channel'
+    IE_NAME = 'natgeo'
-    _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:wild/)?[^/]+/videos/(?P<id>[^/?]+)'
+    _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)
+        video_data = self._download_json(
-        key = m_id.group(2)
+            'http://play.rmcnmv.naver.com/vod/play/v2.0/' + m_id.group(1),
-        query = compat_urllib_parse_urlencode({'vid': vid, 'inKey': key, })
+            video_id, query={
-        query_urls = compat_urllib_parse_urlencode({
+                'key': m_id.group(2),
-            'masterVid': vid,
+            })
-            'protocol': 'p2p',
+        meta = video_data['meta']
-            'inKey': key,
+        title = meta['subject']
-        })
-        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')
-
         formats = []
-        for format_el in urls.findall('EncodingOptions/EncodingOption'):
+
-            domain = format_el.find('Domain').text
+        def extract_formats(streams, stream_type, query={}):
-            uri = format_el.find('uri').text
+            for stream in streams:
-            f = {
+                stream_url = stream.get('source')
-                'url': compat_urlparse.urljoin(domain, uri),
+                if not stream_url:
-                'ext': 'mp4',
+                    continue
-                'width': int(format_el.find('width').text),
+                stream_url = update_url_query(stream_url, query)
-                'height': int(format_el.find('height').text),
+                encoding_option = stream.get('encodingOption', {})
-            }
+                bitrate = stream.get('bitrate', {})
-            if domain.startswith('rtmp'):
+                formats.append({
-                # urlparse does not support custom schemes
+                    'format_id': '%s_%s' % (stream.get('type') or stream_type, encoding_option.get('id') or encoding_option.get('name')),
-                # https://bugs.python.org/issue18828
+                    'url': stream_url,
-                f.update({
+                    'width': int_or_none(encoding_option.get('width')),
-                    'url': domain + uri,
+                    'height': int_or_none(encoding_option.get('height')),
-                    'ext': 'flv',
+                    'vbr': int_or_none(bitrate.get('video')),
-                    'rtmp_protocol': '1',  # rtmpt
+                    '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),
+            'thumbnail': meta.get('cover', {}).get('source') or self._og_search_thumbnail(webpage),
-            'upload_date': info.find('WriteDate').text.replace('.', ''),
+            'view_count': int_or_none(meta.get('count')),
-            'view_count': int(info.find('PlayCount').text),
+            'upload_date': upload_date,
         }

youtube_dl/extractor/nextmovie.py

@@ -1,30 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-from .mtv import MTVServicesInfoExtractor
-from ..compat import compat_urllib_parse_urlencode
-
-
-class NextMovieIE(MTVServicesInfoExtractor):
-    IE_NAME = 'nextmovie.com'
-    _VALID_URL = r'https?://(?:www\.)?nextmovie\.com/shows/[^/]+/\d{4}-\d{2}-\d{2}/(?P<id>[^/?#]+)'
-    _FEED_URL = 'http://lite.dextr.mtvi.com/service1/dispatch.htm'
-    _TESTS = [{
-        'url': 'http://www.nextmovie.com/shows/exclusives/2013-03-10/mgid:uma:videolist:nextmovie.com:1715019/',
-        'md5': '09a9199f2f11f10107d04fcb153218aa',
-        'info_dict': {
-            'id': '961726',
-            'ext': 'mp4',
-            'title': 'The Muppets\' Gravity',
-        },
-    }]
-
-    def _get_feed_query(self, uri):
-        return compat_urllib_parse_urlencode({
-            'feed': '1505',
-            'mgid': uri,
-        })
-
-    def _real_extract(self, url):
-        mgid = self._match_id(url)
-        return self._get_videos_info(mgid)

youtube_dl/extractor/nick.py

@@ -7,8 +7,9 @@ from ..utils import update_url_query
 
 
 class NickIE(MTVServicesInfoExtractor):
+    # None of videos on the website are still alive?
     IE_NAME = 'nick.com'
-    _VALID_URL = r'https?://(?:www\.)?nick\.com/videos/clip/(?P<id>[^/?#.]+)'
+    _VALID_URL = r'https?://(?:www\.)?nick(?:jr)?\.com/(?:videos/clip|[^/]+/videos)/(?P<id>[^/?#.]+)'
     _FEED_URL = 'http://udat.mtvnservices.com/service1/dispatch.htm'
     _TESTS = [{
         'url': 'http://www.nick.com/videos/clip/alvinnn-and-the-chipmunks-112-full-episode.html',
@@ -52,6 +53,9 @@ class NickIE(MTVServicesInfoExtractor):
                 }
             },
         ],
+    }, {
+        'url': 'http://www.nickjr.com/paw-patrol/videos/pups-save-a-goldrush-s3-ep302-full-episode/',
+        'only_matching': True,
     }]
 
     def _get_feed_query(self, uri):

youtube_dl/extractor/ninenow.py

@@ -0,0 +1,72 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    float_or_none,
+    ExtractorError,
+)
+
+
+class NineNowIE(InfoExtractor):
+    IE_NAME = '9now.com.au'
+    _VALID_URL = r'https?://(?:www\.)?9now\.com\.au/(?:[^/]+/){2}(?P<id>[^/?#]+)'
+    _TESTS = [{
+        # clip
+        'url': 'https://www.9now.com.au/afl-footy-show/2016/clip-ciql02091000g0hp5oktrnytc',
+        'md5': '17cf47d63ec9323e562c9957a968b565',
+        'info_dict': {
+            'id': '16801',
+            'ext': 'mp4',
+            'title': 'St. Kilda\'s Joey Montagna on the potential for a player\'s strike',
+            'description': 'Is a boycott of the NAB Cup "on the table"?',
+            'uploader_id': '4460760524001',
+            'upload_date': '20160713',
+            'timestamp': 1468421266,
+        },
+        'skip': 'Only available in Australia',
+    }, {
+        # episode
+        'url': 'https://www.9now.com.au/afl-footy-show/2016/episode-19',
+        'only_matching': True,
+    }, {
+        # DRM protected
+        'url': 'https://www.9now.com.au/andrew-marrs-history-of-the-world/season-1/episode-1',
+        'only_matching': True,
+    }]
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/4460760524001/default_default/index.html?videoId=%s'
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        page_data = self._parse_json(self._search_regex(
+            r'window\.__data\s*=\s*({.*?});', webpage,
+            'page data'), display_id)
+        common_data = page_data.get('episode', {}).get('episode') or page_data.get('clip', {}).get('clip')
+        video_data = common_data['video']
+
+        if video_data.get('drm'):
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        brightcove_id = video_data.get('brightcoveId') or 'ref:' + video_data['referenceId']
+        video_id = compat_str(video_data.get('id') or brightcove_id)
+        title = common_data['name']
+
+        thumbnails = [{
+            'id': thumbnail_id,
+            'url': thumbnail_url,
+            'width': int_or_none(thumbnail_id[1:])
+        } for thumbnail_id, thumbnail_url in common_data.get('image', {}).get('sizes', {}).items()]
+
+        return {
+            '_type': 'url_transparent',
+            'url': self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id,
+            'id': video_id,
+            'title': title,
+            'description': common_data.get('description'),
+            'duration': float_or_none(video_data.get('duration'), 1000),
+            'thumbnails': thumbnails,
+            'ie_key': 'BrightcoveNew',
+        }

youtube_dl/extractor/nintendo.py

@@ -0,0 +1,46 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from .ooyala import OoyalaIE
+from ..utils import unescapeHTML
+
+
+class NintendoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?nintendo\.com/games/detail/(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'http://www.nintendo.com/games/detail/yEiAzhU2eQI1KZ7wOHhngFoAHc1FpHwj',
+        'info_dict': {
+            'id': 'MzMmticjp0VPzO3CCj4rmFOuohEuEWoW',
+            'ext': 'flv',
+            'title': 'Duck Hunt Wii U VC NES - Trailer',
+            'duration': 60.326,
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': ['Ooyala'],
+    }, {
+        'url': 'http://www.nintendo.com/games/detail/tokyo-mirage-sessions-fe-wii-u',
+        'info_dict': {
+            'id': 'tokyo-mirage-sessions-fe-wii-u',
+            'title': 'Tokyo Mirage Sessions ♯FE',
+        },
+        'playlist_count': 3,
+    }]
+
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, page_id)
+
+        entries = [
+            OoyalaIE._build_url_result(m.group('code'))
+            for m in re.finditer(
+                r'class=(["\'])embed-video\1[^>]+data-video-code=(["\'])(?P<code>(?:(?!\2).)+)\2',
+                webpage)]
+
+        return self.playlist_result(
+            entries, page_id, unescapeHTML(self._og_search_title(webpage, fatal=False)))

youtube_dl/extractor/ntvru.py

@@ -11,70 +11,64 @@ 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/',
-            'url': 'http://www.ntv.ru/novosti/863142/',
+        'md5': 'ba7ea172a91cb83eb734cad18c10e723',
-            'md5': 'ba7ea172a91cb83eb734cad18c10e723',
+        'info_dict': {
-            'info_dict': {
+            'id': '746000',
-                'id': '746000',
+            'ext': 'mp4',
-                'ext': 'mp4',
+            'title': 'Командующий Черноморским флотом провел переговоры в штабе ВМС Украины',
-                'title': 'Командующий Черноморским флотом провел переговоры в штабе ВМС Украины',
+            'description': 'Командующий Черноморским флотом провел переговоры в штабе ВМС Украины',
-                'description': 'Командующий Черноморским флотом провел переговоры в штабе ВМС Украины',
+            'thumbnail': 're:^http://.*\.jpg',
-                'thumbnail': 're:^http://.*\.jpg',
+            'duration': 136,
-                'duration': 136,
-            },
         },
-        {
+    }, {
-            'url': 'http://www.ntv.ru/video/novosti/750370/',
+        'url': 'http://www.ntv.ru/video/novosti/750370/',
-            'md5': 'adecff79691b4d71e25220a191477124',
+        'md5': 'adecff79691b4d71e25220a191477124',
-            'info_dict': {
+        'info_dict': {
-                'id': '750370',
+            'id': '750370',
-                'ext': 'mp4',
+            'ext': 'mp4',
-                'title': 'Родные пассажиров пропавшего Boeing не верят в трагический исход',
+            'title': 'Родные пассажиров пропавшего Boeing не верят в трагический исход',
-                'description': 'Родные пассажиров пропавшего Boeing не верят в трагический исход',
+            'description': 'Родные пассажиров пропавшего Boeing не верят в трагический исход',
-                'thumbnail': 're:^http://.*\.jpg',
+            'thumbnail': 're:^http://.*\.jpg',
-                'duration': 172,
+            'duration': 172,
-            },
         },
-        {
+    }, {
-            'url': 'http://www.ntv.ru/peredacha/segodnya/m23700/o232416',
+        'url': 'http://www.ntv.ru/peredacha/segodnya/m23700/o232416',
-            'md5': '82dbd49b38e3af1d00df16acbeab260c',
+        'md5': '82dbd49b38e3af1d00df16acbeab260c',
-            'info_dict': {
+        'info_dict': {
-                'id': '747480',
+            'id': '747480',
-                'ext': 'mp4',
+            'ext': 'mp4',
-                'title': '«Сегодня». 21 марта 2014 года. 16:00',
+            'title': '«Сегодня». 21 марта 2014 года. 16:00',
-                'description': '«Сегодня». 21 марта 2014 года. 16:00',
+            'description': '«Сегодня». 21 марта 2014 года. 16:00',
-                'thumbnail': 're:^http://.*\.jpg',
+            'thumbnail': 're:^http://.*\.jpg',
-                'duration': 1496,
+            'duration': 1496,
-            },
         },
-        {
+    }, {
-            'url': 'http://www.ntv.ru/kino/Koma_film',
+        'url': 'http://www.ntv.ru/kino/Koma_film',
-            'md5': 'f825770930937aa7e5aca0dc0d29319a',
+        'md5': 'f825770930937aa7e5aca0dc0d29319a',
-            'info_dict': {
+        'info_dict': {
-                'id': '1007609',
+            'id': '1007609',
-                'ext': 'mp4',
+            'ext': 'mp4',
-                'title': 'Остросюжетный фильм «Кома»',
+            'title': 'Остросюжетный фильм «Кома»',
-                'description': 'Остросюжетный фильм «Кома»',
+            'description': 'Остросюжетный фильм «Кома»',
-                'thumbnail': 're:^http://.*\.jpg',
+            'thumbnail': 're:^http://.*\.jpg',
-                'duration': 5592,
+            'duration': 5592,
-            },
         },
-        {
+    }, {
-            'url': 'http://www.ntv.ru/serial/Delo_vrachey/m31760/o233916/',
+        'url': 'http://www.ntv.ru/serial/Delo_vrachey/m31760/o233916/',
-            'md5': '9320cd0e23f3ea59c330dc744e06ff3b',
+        'md5': '9320cd0e23f3ea59c330dc744e06ff3b',
-            'info_dict': {
+        'info_dict': {
-                'id': '751482',
+            'id': '751482',
-                'ext': 'mp4',
+            'ext': 'mp4',
-                'title': '«Дело врачей»: «Деревце жизни»',
+            'title': '«Дело врачей»: «Деревце жизни»',
-                'description': '«Дело врачей»: «Деревце жизни»',
+            'description': '«Дело врачей»: «Деревце жизни»',
-                'thumbnail': 're:^http://.*\.jpg',
+            'thumbnail': 're:^http://.*\.jpg',
-                'duration': 2590,
+            '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'))