Merge branch 'master' into tvpleextractor

.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 *2016.04.06*. 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 Issues with outdated version will be rejected.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 *2016.05.10*. 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 Issues with outdated version will be rejected.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 **2016.04.06**
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.05.10**
 
 ### 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 2016.04.06
+[debug] youtube-dl version 2016.05.10
 [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: {}

.gitignore

@@ -31,7 +31,9 @@ updates_key.pem
 *.part
 *.swp
 test/testdata
+test/local_parameters.json
 .tox
 youtube-dl.zsh
 .idea
 .idea/*
+tmp/

.travis.yml

@@ -7,6 +7,9 @@ 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

@@ -167,3 +167,8 @@ Kacper Michajłow
 José Joaquín Atria
 Viťas Strádal
 Kagami Hiiragi
+Philip Huppert
+blahgeek
+Kevin Deldycke
+inondle
+Tomáš Čech

Makefile

@@ -1,7 +1,7 @@
 all: youtube-dl README.md CONTRIBUTING.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish supportedsites
 
 clean:
-	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish youtube_dl/extractor/lazy_extractors.py *.dump *.part *.info.json *.mp4 *.flv *.mp3 *.avi CONTRIBUTING.md.tmp ISSUE_TEMPLATE.md.tmp youtube-dl youtube-dl.exe
+	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish youtube_dl/extractor/lazy_extractors.py *.dump *.part *.info.json *.mp4 *.flv *.mp3 *.avi *.mkv *.webm CONTRIBUTING.md.tmp ISSUE_TEMPLATE.md.tmp youtube-dl youtube-dl.exe
 	find . -name "*.pyc" -delete
 	find . -name "*.class" -delete
 
@@ -37,7 +37,7 @@ test:
 ot: offlinetest
 
 offlinetest: codetest
-	$(PYTHON) -m nose --verbose test --exclude test_download.py --exclude test_age_restriction.py --exclude test_subtitles.py --exclude test_write_annotations.py --exclude test_youtube_lists.py --exclude test_iqiyi_sdk_interpreter.py
+	$(PYTHON) -m nose --verbose test --exclude test_download.py --exclude test_age_restriction.py --exclude test_subtitles.py --exclude test_write_annotations.py --exclude test_youtube_lists.py --exclude test_iqiyi_sdk_interpreter.py --exclude test_socks.py
 
 tar: youtube-dl.tar.gz
 

README.md

@@ -85,9 +85,11 @@ which means you can modify it, redistribute it or use it however you like.
     --no-color                       Do not emit color codes in output
 
 ## Network Options:
-    --proxy URL                      Use the specified HTTP/HTTPS proxy. Pass in
+    --proxy URL                      Use the specified HTTP/HTTPS/SOCKS proxy.
-                                     an empty string (--proxy "") for direct
+                                     To enable experimental SOCKS proxy, specify
-                                     connection
+                                     a proper scheme. For example
+                                     socks5://127.0.0.1:1080/. Pass in an empty
+                                     string (--proxy "") for direct connection
     --socket-timeout SECONDS         Time to wait before giving up, in seconds
     --source-address IP              Client-side IP address to bind to
                                      (experimental)
@@ -176,7 +178,9 @@ which means you can modify it, redistribute it or use it however you like.
     --xattr-set-filesize             Set file xattribute ytdl.filesize with
                                      expected filesize (experimental)
     --hls-prefer-native              Use the native HLS downloader instead of
-                                     ffmpeg (experimental)
+                                     ffmpeg
+    --hls-prefer-ffmpeg              Use ffmpeg instead of the native HLS
+                                     downloader
     --hls-use-mpegts                 Use the mpegts container for HLS videos,
                                      allowing to play the video while
                                      downloading (some players may not be able
@@ -463,7 +467,7 @@ The basic usage is not to set any template arguments when downloading a single f
  - `display_id`: An alternative identifier for the video
  - `uploader`: Full name of the video uploader
  - `license`: License name the video is licensed under
- - `creator`: The main artist who created the video
+ - `creator`: The creator of the video
  - `release_date`: The date (YYYYMMDD) when the video was released
  - `timestamp`: UNIX timestamp of the moment the video became available
  - `upload_date`: Video upload date (YYYYMMDD)
@@ -515,6 +519,18 @@ Available for the video that is an episode of some series or programme:
  - `episode_number`: Number of the video episode within a season
  - `episode_id`: Id of the video episode
 
+Available for the media that is a track or a part of a music album:
+ - `track`: Title of the track
+ - `track_number`: Number of the track within an album or a disc
+ - `track_id`: Id of the track
+ - `artist`: Artist(s) of the track
+ - `genre`: Genre(s) of the track
+ - `album`: Title of the album the track belongs to
+ - `album_type`: Type of the album
+ - `album_artist`: List of all artists appeared on the album
+ - `disc_number`: Number of the disc or other physical medium the track belongs to
+ - `release_year`: Year (YYYY) when the album was released
+
 Each aforementioned sequence when referenced in output template will be replaced by the actual value corresponding to the sequence name. Note that some of the sequences are not guaranteed to be present since they depend on the metadata obtained by particular extractor, such sequences will be replaced with `NA`.
 
 For example for `-o %(title)s-%(id)s.%(ext)s` and mp4 video with title `youtube-dl test video` and id `BaW_jenozKcj` this will result in a `youtube-dl test video-BaW_jenozKcj.mp4` file created in the current directory.
@@ -683,6 +699,10 @@ YouTube changed their playlist format in March 2014 and later on, so you'll need
 
 If you have installed youtube-dl with a package manager, pip, setup.py or a tarball, please use that to update. Note that Ubuntu packages do not seem to get updated anymore. Since we are not affiliated with Ubuntu, there is little we can do. Feel free to [report bugs](https://bugs.launchpad.net/ubuntu/+source/youtube-dl/+filebug) to the [Ubuntu packaging guys](mailto:ubuntu-motu@lists.ubuntu.com?subject=outdated%20version%20of%20youtube-dl) - all they have to do is update the package to a somewhat recent version. See above for a way to update.
 
+### I'm getting an error when trying to use output template: `error: using output template conflicts with using title, video ID or auto number`
+
+Make sure you are not using `-o` with any of these options `-t`, `--title`, `--id`, `-A` or `--auto-number` set in command line or in a configuration file. Remove the latter if any.
+
 ### Do I always have to pass `-citw`?
 
 By default, youtube-dl intends to have the best options (incidentally, if you have a convincing case that these should be different, [please file an issue where you explain that](https://yt-dl.org/bug)). Therefore, it is unnecessary and sometimes harmful to copy long option strings from webpages. In particular, the only option out of `-citw` that is regularly useful is `-i`.
@@ -703,7 +723,7 @@ Videos or video formats streamed via RTMP protocol can only be downloaded when [
 
 ### I have downloaded a video but how can I play it?
 
-Once the video is fully downloaded, use any video player, such as [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/).
+Once the video is fully downloaded, use any video player, such as [mpv](https://mpv.io/), [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/).
 
 ### I extracted a video URL with `-g`, but it does not play on another machine / in my webbrowser.
 

devscripts/install_srelay.sh

@@ -0,0 +1,8 @@
+#!/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

docs/supportedsites.md

@@ -50,6 +50,7 @@
  - **arte.tv:ddc**
  - **arte.tv:embed**
  - **arte.tv:future**
+ - **arte.tv:info**
  - **arte.tv:magazine**
  - **AtresPlayer**
  - **ATTTechChannel**
@@ -76,6 +77,7 @@
  - **Bild**: Bild.de
  - **BiliBili**
  - **BioBioChileTV**
+ - **BIQLE**
  - **BleacherReport**
  - **BleacherReportCMS**
  - **blinkx**
@@ -115,6 +117,7 @@
  - **Cinemassacre**
  - **Clipfish**
  - **cliphunter**
+ - **ClipRs**
  - **Clipsyndicate**
  - **cloudtime**: CloudTime
  - **Cloudy**
@@ -143,6 +146,7 @@
  - **culturebox.francetvinfo.fr**
  - **CultureUnplugged**
  - **CWTV**
+ - **DailyMail**
  - **dailymotion**
  - **dailymotion:playlist**
  - **dailymotion:user**
@@ -161,6 +165,7 @@
  - **defense.gouv.fr**
  - **democracynow**
  - **DHM**: Filmarchiv - Deutsches Historisches Museum
+ - **DigitallySpeaking**
  - **Digiteka**
  - **Discovery**
  - **Dotsub**
@@ -172,7 +177,6 @@
  - **Dropbox**
  - **DrTuber**
  - **DRTV**
- - **Dump**
  - **Dumpert**
  - **dvtv**: http://video.aktualne.cz/
  - **dw**
@@ -286,7 +290,6 @@
  - **ivi:compilation**: ivi.ru compilations
  - **ivideon**: Ivideon TV
  - **Izlesene**
- - **JadoreCettePub**
  - **JeuxVideo**
  - **Jove**
  - **jpopsuki.tv**
@@ -324,6 +327,7 @@
  - **limelight**
  - **limelight:channel**
  - **limelight:channel_list**
+ - **LiTV**
  - **LiveLeak**
  - **livestream**
  - **livestream:original**
@@ -337,26 +341,28 @@
  - **mailru**: Видео@Mail.Ru
  - **MakersChannel**
  - **MakerTV**
- - **Malemotion**
  - **MatchTV**
  - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
  - **metacafe**
  - **Metacritic**
  - **Mgoon**
+ - **MGTV**: 芒果TV
  - **Minhateca**
  - **MinistryGrid**
  - **Minoto**
  - **miomio.tv**
  - **MiTele**: mitele.es
  - **mixcloud**
+ - **mixcloud:playlist**
+ - **mixcloud:stream**
+ - **mixcloud:user**
  - **MLB**
  - **Mnet**
  - **MoeVideo**: LetitBit video services: moevideo.net, playreplay.net and videochart.net
  - **Mofosex**
  - **Mojvideo**
  - **Moniker**: allmyvideos.net and vidspot.net
- - **mooshare**: Mooshare.biz
  - **Morningstar**: morningstar.com
  - **Motherless**
  - **Motorsport**: motorsport.com
@@ -371,8 +377,10 @@
  - **mtvservices:embedded**
  - **MuenchenTV**: münchen.tv
  - **MusicPlayOn**
- - **muzu.tv**
+ - **mva**: Microsoft Virtual Academy videos
+ - **mva:course**: Microsoft Virtual Academy courses
  - **Mwave**
+ - **MwaveMeetGreet**
  - **MySpace**
  - **MySpace:album**
  - **MySpass**
@@ -393,7 +401,6 @@
  - **ndr:embed:base**
  - **NDTV**
  - **NerdCubedFeed**
- - **Nerdist**
  - **netease:album**: 网易云音乐 - 专辑
  - **netease:djradio**: 网易云音乐 - 电台
  - **netease:mv**: 网易云音乐 - MV
@@ -411,7 +418,8 @@
  - **nfl.com**
  - **nhl.com**
  - **nhl.com:news**: NHL news
- - **nhl.com:videocenter**: NHL videocenter category
+ - **nhl.com:videocenter**
+ - **nhl.com:videocenter:category**: NHL videocenter category
  - **nick.com**
  - **niconico**: ニコニコ動画
  - **NiconicoPlaylist**
@@ -459,13 +467,14 @@
  - **Patreon**
  - **pbs**: Public Broadcasting Service (PBS) and member stations: PBS: Public Broadcasting Service, APT - Alabama Public Television (WBIQ), GPB/Georgia Public Broadcasting (WGTV), Mississippi Public Broadcasting (WMPN), Nashville Public Television (WNPT), WFSU-TV (WFSU), WSRE (WSRE), WTCI (WTCI), WPBA/Channel 30 (WPBA), Alaska Public Media (KAKM), Arizona PBS (KAET), KNME-TV/Channel 5 (KNME), Vegas PBS (KLVX), AETN/ARKANSAS ETV NETWORK (KETS), KET (WKLE), WKNO/Channel 10 (WKNO), LPB/LOUISIANA PUBLIC BROADCASTING (WLPB), OETA (KETA), Ozarks Public Television (KOZK), WSIU Public Broadcasting (WSIU), KEET TV (KEET), KIXE/Channel 9 (KIXE), KPBS San Diego (KPBS), KQED (KQED), KVIE Public Television (KVIE), PBS SoCal/KOCE (KOCE), ValleyPBS (KVPT), CONNECTICUT PUBLIC TELEVISION (WEDH), KNPB Channel 5 (KNPB), SOPTV (KSYS), Rocky Mountain PBS (KRMA), KENW-TV3 (KENW), KUED Channel 7 (KUED), Wyoming PBS (KCWC), Colorado Public Television / KBDI 12 (KBDI), KBYU-TV (KBYU), Thirteen/WNET New York (WNET), WGBH/Channel 2 (WGBH), WGBY (WGBY), NJTV Public Media NJ (WNJT), WLIW21 (WLIW), mpt/Maryland Public Television (WMPB), WETA Television and Radio (WETA), WHYY (WHYY), PBS 39 (WLVT), WVPT - Your Source for PBS and More! (WVPT), Howard University Television (WHUT), WEDU PBS (WEDU), WGCU Public Media (WGCU), WPBT2 (WPBT), WUCF TV (WUCF), WUFT/Channel 5 (WUFT), WXEL/Channel 42 (WXEL), WLRN/Channel 17 (WLRN), WUSF Public Broadcasting (WUSF), ETV (WRLK), UNC-TV (WUNC), PBS Hawaii - Oceanic Cable Channel 10 (KHET), Idaho Public Television (KAID), KSPS (KSPS), OPB (KOPB), KWSU/Channel 10 & KTNW/Channel 31 (KWSU), WILL-TV (WILL), Network Knowledge - WSEC/Springfield (WSEC), WTTW11 (WTTW), Iowa Public Television/IPTV (KDIN), Nine Network (KETC), PBS39 Fort Wayne (WFWA), WFYI Indianapolis (WFYI), Milwaukee Public Television (WMVS), WNIN (WNIN), WNIT Public Television (WNIT), WPT (WPNE), WVUT/Channel 22 (WVUT), WEIU/Channel 51 (WEIU), WQPT-TV (WQPT), WYCC PBS Chicago (WYCC), WIPB-TV (WIPB), WTIU (WTIU), CET  (WCET), ThinkTVNetwork (WPTD), WBGU-TV (WBGU), WGVU TV (WGVU), NET1 (KUON), Pioneer Public Television (KWCM), SDPB Television (KUSD), TPT (KTCA), KSMQ (KSMQ), KPTS/Channel 8 (KPTS), KTWU/Channel 11 (KTWU), East Tennessee PBS (WSJK), WCTE-TV (WCTE), WLJT, Channel 11 (WLJT), WOSU TV (WOSU), WOUB/WOUC (WOUB), WVPB (WVPB), WKYU-PBS (WKYU), KERA 13 (KERA), MPBN (WCBB), Mountain Lake PBS (WCFE), NHPTV (WENH), Vermont PBS (WETK), witf (WITF), WQED Multimedia (WQED), WMHT Educational Telecommunications (WMHT), Q-TV (WDCQ), WTVS Detroit Public TV (WTVS), CMU Public Television (WCMU), WKAR-TV (WKAR), WNMU-TV Public TV 13 (WNMU), WDSE - WRPT (WDSE), WGTE TV (WGTE), Lakeland Public Television (KAWE), KMOS-TV - Channels 6.1, 6.2 and 6.3 (KMOS), MontanaPBS (KUSM), KRWG/Channel 22 (KRWG), KACV (KACV), KCOS/Channel 13 (KCOS), WCNY/Channel 24 (WCNY), WNED (WNED), WPBS (WPBS), WSKG Public TV (WSKG), WXXI (WXXI), WPSU (WPSU), WVIA Public Media Studios (WVIA), WTVI (WTVI), Western Reserve PBS (WNEO), WVIZ/PBS ideastream (WVIZ), KCTS 9 (KCTS), Basin PBS (KPBT), KUHT / Channel 8 (KUHT), KLRN (KLRN), KLRU (KLRU), WTJX Channel 12 (WTJX), WCVE PBS (WCVE), KBTC Public Television (KBTC)
  - **pcmag**
- - **Periscope**: Periscope
+ - **People**
+ - **periscope**: Periscope
+ - **periscope:user**: Periscope user videos
  - **PhilharmonieDeParis**: Philharmonie de Paris
  - **phoenix.de**
  - **Photobucket**
  - **Pinkbike**
  - **Pladform**
- - **PlanetaPlay**
  - **play.fm**
  - **played.to**
  - **PlaysTV**
@@ -484,6 +493,7 @@
  - **Pornotube**
  - **PornoVoisines**
  - **PornoXO**
+ - **PressTV**
  - **PrimeShareTV**
  - **PromptFile**
  - **prosiebensat1**: ProSiebenSat.1 Digital
@@ -494,7 +504,6 @@
  - **qqmusic:playlist**: QQ音乐 - 歌单
  - **qqmusic:singer**: QQ音乐 - 歌手
  - **qqmusic:toplist**: QQ音乐 - 排行榜
- - **QuickVid**
  - **R7**
  - **radio.de**
  - **radiobremen**
@@ -550,7 +559,6 @@
  - **SenateISVP**
  - **ServingSys**
  - **Sexu**
- - **SexyKarma**: Sexy Karma and Watch Indian Porn
  - **Shahid**
  - **Shared**: shared.sx and vivo.sx
  - **ShareSix**
@@ -563,8 +571,6 @@
  - **smotri:broadcast**: Smotri.com broadcasts
  - **smotri:community**: Smotri.com community videos
  - **smotri:user**: Smotri.com user videos
- - **SnagFilms**
- - **SnagFilmsEmbed**
  - **Snotr**
  - **Sohu**
  - **soundcloud**
@@ -606,8 +612,10 @@
  - **Syfy**
  - **SztvHu**
  - **Tagesschau**
+ - **tagesschau:player**
  - **Tapely**
  - **Tass**
+ - **TDSLifeway**
  - **teachertube**: teachertube.com videos
  - **teachertube:user:collection**: teachertube.com user and collection videos
  - **TeachingChannel**
@@ -624,7 +632,6 @@
  - **TeleTask**
  - **TF1**
  - **TheIntercept**
- - **TheOnion**
  - **ThePlatform**
  - **ThePlatformFeed**
  - **TheScene**
@@ -684,7 +691,6 @@
  - **twitter**
  - **twitter:amplify**
  - **twitter:card**
- - **Ubu**
  - **udemy**
  - **udemy:course**
  - **UDNEmbed**: 聯合影音
@@ -701,6 +707,7 @@
  - **Vessel**
  - **Vesti**: Вести.Ru
  - **Vevo**
+ - **VevoPlaylist**
  - **VGTV**: VGTV, BTTV, FTV, Aftenposten and Aftonbladet
  - **vh1.com**
  - **Vice**
@@ -723,6 +730,8 @@
  - **Vidzi**
  - **vier**
  - **vier:videos**
+ - **ViewLift**
+ - **ViewLiftEmbed**
  - **Viewster**
  - **Viidea**
  - **viki**
@@ -754,7 +763,7 @@
  - **Walla**
  - **WashingtonPost**
  - **wat.tv**
- - **WayOfTheMaster**
+ - **WatchIndianPorn**: Watch Indian Porn
  - **WDR**
  - **wdr:mobile**
  - **WDRMaus**: Sendung mit der Maus
@@ -771,9 +780,13 @@
  - **WSJ**: Wall Street Journal
  - **XBef**
  - **XboxClips**
- - **XFileShare**: XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me
+ - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To
  - **XHamster**
  - **XHamsterEmbed**
+ - **xiami:album**: 虾米音乐 - 专辑
+ - **xiami:artist**: 虾米音乐 - 歌手
+ - **xiami:collection**: 虾米音乐 - 精选集
+ - **xiami:song**: 虾米音乐
  - **XMinus**
  - **XNXX**
  - **Xstream**

test/helper.py

@@ -24,8 +24,13 @@ from youtube_dl.utils import (
 def get_params(override=None):
     PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
                                    "parameters.json")
+    LOCAL_PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
+                                         "local_parameters.json")
     with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
         parameters = json.load(pf)
+    if os.path.exists(LOCAL_PARAMETERS_FILE):
+        with io.open(LOCAL_PARAMETERS_FILE, encoding='utf-8') as pf:
+            parameters.update(json.load(pf))
     if override:
         parameters.update(override)
     return parameters
@@ -143,6 +148,9 @@ def expect_value(self, got, expected, field):
             expect_value(self, item_got, item_expected, field)
     else:
         if isinstance(expected, compat_str) and expected.startswith('md5:'):
+            self.assertTrue(
+                isinstance(got, compat_str),
+                'Expected field %s to be a unicode object, but got value %r of type %r' % (field, got, type(got)))
             got = 'md5:' + md5(got)
         elif isinstance(expected, compat_str) and expected.startswith('mincount:'):
             self.assertTrue(

test/test_InfoExtractor.py

@@ -11,6 +11,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from test.helper import FakeYDL
 from youtube_dl.extractor.common import InfoExtractor
 from youtube_dl.extractor import YoutubeIE, get_info_extractor
+from youtube_dl.utils import encode_data_uri, strip_jsonp, ExtractorError
 
 
 class TestIE(InfoExtractor):
@@ -66,5 +67,14 @@ class TestInfoExtractor(unittest.TestCase):
         self.assertEqual(ie._html_search_meta('e', html), '5')
         self.assertEqual(ie._html_search_meta('f', html), '6')
 
+    def test_download_json(self):
+        uri = encode_data_uri(b'{"foo": "blah"}', 'application/json')
+        self.assertEqual(self.ie._download_json(uri, None), {'foo': 'blah'})
+        uri = encode_data_uri(b'callback({"foo": "blah"})', 'application/javascript')
+        self.assertEqual(self.ie._download_json(uri, None, transform_source=strip_jsonp), {'foo': 'blah'})
+        uri = encode_data_uri(b'{"foo": invalid}', 'application/json')
+        self.assertRaises(ExtractorError, self.ie._download_json, uri, None)
+        self.assertEqual(self.ie._download_json(uri, None, fatal=False), None)
+
 if __name__ == '__main__':
     unittest.main()

test/test_compat.py

@@ -10,13 +10,14 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
-from youtube_dl.utils import get_filesystem_encoding
 from youtube_dl.compat import (
     compat_getenv,
+    compat_setenv,
     compat_etree_fromstring,
     compat_expanduser,
     compat_shlex_split,
     compat_str,
+    compat_struct_unpack,
     compat_urllib_parse_unquote,
     compat_urllib_parse_unquote_plus,
     compat_urllib_parse_urlencode,
@@ -26,19 +27,22 @@ from youtube_dl.compat import (
 class TestCompat(unittest.TestCase):
     def test_compat_getenv(self):
         test_str = 'тест'
-        os.environ['YOUTUBE-DL-TEST'] = (
+        compat_setenv('YOUTUBE-DL-TEST', test_str)
-            test_str if sys.version_info >= (3, 0)
-            else test_str.encode(get_filesystem_encoding()))
         self.assertEqual(compat_getenv('YOUTUBE-DL-TEST'), test_str)
 
+    def test_compat_setenv(self):
+        test_var = 'YOUTUBE-DL-TEST'
+        test_str = 'тест'
+        compat_setenv(test_var, test_str)
+        compat_getenv(test_var)
+        self.assertEqual(compat_getenv(test_var), test_str)
+
     def test_compat_expanduser(self):
         old_home = os.environ.get('HOME')
         test_str = 'C:\Documents and Settings\тест\Application Data'
-        os.environ['HOME'] = (
+        compat_setenv('HOME', test_str)
-            test_str if sys.version_info >= (3, 0)
-            else test_str.encode(get_filesystem_encoding()))
         self.assertEqual(compat_expanduser('~'), test_str)
-        os.environ['HOME'] = old_home
+        compat_setenv('HOME', old_home or '')
 
     def test_all_present(self):
         import youtube_dl.compat
@@ -99,5 +103,9 @@ class TestCompat(unittest.TestCase):
         self.assertTrue(isinstance(doc.find('chinese').text, compat_str))
         self.assertTrue(isinstance(doc.find('foo/bar').text, compat_str))
 
+    def test_struct_unpack(self):
+        self.assertEqual(compat_struct_unpack('!B', b'\x00'), (0,))
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_socks.py

@@ -0,0 +1,107 @@
+#!/usr/bin/env python
+# coding: utf-8
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+import random
+import subprocess
+
+from test.helper import (
+    FakeYDL,
+    get_params,
+)
+from youtube_dl.compat import (
+    compat_str,
+    compat_urllib_request,
+)
+
+
+class TestMultipleSocks(unittest.TestCase):
+    @staticmethod
+    def _check_params(attrs):
+        params = get_params()
+        for attr in attrs:
+            if attr not in params:
+                print('Missing %s. Skipping.' % attr)
+                return
+        return params
+
+    def test_proxy_http(self):
+        params = self._check_params(['primary_proxy', 'primary_server_ip'])
+        if params is None:
+            return
+        ydl = FakeYDL({
+            'proxy': params['primary_proxy']
+        })
+        self.assertEqual(
+            ydl.urlopen('http://yt-dl.org/ip').read().decode('utf-8'),
+            params['primary_server_ip'])
+
+    def test_proxy_https(self):
+        params = self._check_params(['primary_proxy', 'primary_server_ip'])
+        if params is None:
+            return
+        ydl = FakeYDL({
+            'proxy': params['primary_proxy']
+        })
+        self.assertEqual(
+            ydl.urlopen('https://yt-dl.org/ip').read().decode('utf-8'),
+            params['primary_server_ip'])
+
+    def test_secondary_proxy_http(self):
+        params = self._check_params(['secondary_proxy', 'secondary_server_ip'])
+        if params is None:
+            return
+        ydl = FakeYDL()
+        req = compat_urllib_request.Request('http://yt-dl.org/ip')
+        req.add_header('Ytdl-request-proxy', params['secondary_proxy'])
+        self.assertEqual(
+            ydl.urlopen(req).read().decode('utf-8'),
+            params['secondary_server_ip'])
+
+    def test_secondary_proxy_https(self):
+        params = self._check_params(['secondary_proxy', 'secondary_server_ip'])
+        if params is None:
+            return
+        ydl = FakeYDL()
+        req = compat_urllib_request.Request('https://yt-dl.org/ip')
+        req.add_header('Ytdl-request-proxy', params['secondary_proxy'])
+        self.assertEqual(
+            ydl.urlopen(req).read().decode('utf-8'),
+            params['secondary_server_ip'])
+
+
+class TestSocks(unittest.TestCase):
+    def setUp(self):
+        self.port = random.randint(20000, 30000)
+        self.server_process = subprocess.Popen([
+            'srelay', '-f', '-i', '127.0.0.1:%d' % self.port],
+            stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+
+    def tearDown(self):
+        self.server_process.terminate()
+        self.server_process.communicate()
+
+    def _get_ip(self, protocol):
+        ydl = FakeYDL({
+            'proxy': '%s://127.0.0.1:%d' % (protocol, self.port),
+        })
+        return ydl.urlopen('http://yt-dl.org/ip').read().decode('utf-8')
+
+    def test_socks4(self):
+        self.assertTrue(isinstance(self._get_ip('socks4'), compat_str))
+
+    def test_socks4a(self):
+        self.assertTrue(isinstance(self._get_ip('socks4a'), compat_str))
+
+    def test_socks5(self):
+        self.assertTrue(isinstance(self._get_ip('socks5'), compat_str))
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -20,6 +20,7 @@ from youtube_dl.utils import (
     args_to_str,
     encode_base_n,
     clean_html,
+    date_from_str,
     DateRange,
     detect_exe_version,
     determine_ext,
@@ -54,7 +55,6 @@ from youtube_dl.utils import (
     smuggle_url,
     str_to_int,
     strip_jsonp,
-    struct_unpack,
     timeconvert,
     unescapeHTML,
     unified_strdate,
@@ -138,8 +138,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual('yes_no', sanitize_filename('yes? no', restricted=True))
         self.assertEqual('this_-_that', sanitize_filename('this: that', restricted=True))
 
-        tests = 'a\xe4b\u4e2d\u56fd\u7684c'
+        tests = 'aäb\u4e2d\u56fd\u7684c'
-        self.assertEqual(sanitize_filename(tests, restricted=True), 'a_b_c')
+        self.assertEqual(sanitize_filename(tests, restricted=True), 'aab_c')
         self.assertTrue(sanitize_filename('\xf6', restricted=True) != '')  # No empty filename
 
         forbidden = '"\0\\/&!: \'\t\n()[]{}$;`^,#'
@@ -154,6 +154,10 @@ class TestUtil(unittest.TestCase):
         self.assertTrue(sanitize_filename('-', restricted=True) != '')
         self.assertTrue(sanitize_filename(':', restricted=True) != '')
 
+        self.assertEqual(sanitize_filename(
+            'ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØŒÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøœùúûüýþÿ', restricted=True),
+            'AAAAAAAECEEEEIIIIDNOOOOOOOEUUUUYPssaaaaaaaeceeeeiiiionoooooooeuuuuypy')
+
     def test_sanitize_ids(self):
         self.assertEqual(sanitize_filename('_n_cd26wFpw', is_id=True), '_n_cd26wFpw')
         self.assertEqual(sanitize_filename('_BD_eEpuzXw', is_id=True), '_BD_eEpuzXw')
@@ -234,6 +238,13 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unescapeHTML('é'), 'é')
         self.assertEqual(unescapeHTML('�'), '�')
 
+    def test_date_from_str(self):
+        self.assertEqual(date_from_str('yesterday'), date_from_str('now-1day'))
+        self.assertEqual(date_from_str('now+7day'), date_from_str('now+1week'))
+        self.assertEqual(date_from_str('now+14day'), date_from_str('now+2week'))
+        self.assertEqual(date_from_str('now+365day'), date_from_str('now+1year'))
+        self.assertEqual(date_from_str('now+30day'), date_from_str('now+1month'))
+
     def test_daterange(self):
         _20century = DateRange("19000101", "20000101")
         self.assertFalse("17890714" in _20century)
@@ -405,6 +416,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('01:02:03:04'), 93784)
         self.assertEqual(parse_duration('1 hour 3 minutes'), 3780)
         self.assertEqual(parse_duration('87 Min.'), 5220)
+        self.assertEqual(parse_duration('PT1H0.040S'), 3600.04)
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(
@@ -444,9 +456,6 @@ class TestUtil(unittest.TestCase):
         testPL(5, 2, (2, 99), [2, 3, 4])
         testPL(5, 2, (20, 99), [])
 
-    def test_struct_unpack(self):
-        self.assertEqual(struct_unpack('!B', b'\x00'), (0,))
-
     def test_read_batch_urls(self):
         f = io.StringIO('''\xef\xbb\xbf foo
             bar\r

test/test_youtube_lists.py

@@ -44,7 +44,7 @@ class TestYoutubeLists(unittest.TestCase):
         ie = YoutubePlaylistIE(dl)
         result = ie.extract('https://www.youtube.com/watch?v=W01L70IGBgE&index=2&list=RDOQpdSVF_k_w')
         entries = result['entries']
-        self.assertTrue(len(entries) >= 20)
+        self.assertTrue(len(entries) >= 50)
         original_video = entries[0]
         self.assertEqual(original_video['id'], 'OQpdSVF_k_w')
 

tox.ini

@@ -9,5 +9,6 @@ passenv = HOME
 defaultargs = test --exclude test_download.py --exclude test_age_restriction.py
     --exclude test_subtitles.py --exclude test_write_annotations.py
     --exclude test_youtube_lists.py --exclude test_iqiyi_sdk_interpreter.py
+    --exclude test_socks.py
 commands = nosetests --verbose {posargs:{[testenv]defaultargs}}  # --with-coverage --cover-package=youtube_dl --cover-html
                                                # test.test_download:TestDownload.test_NowVideo

youtube_dl/YoutubeDL.py

@@ -64,6 +64,7 @@ from .utils import (
     PostProcessingError,
     preferredencoding,
     prepend_extension,
+    register_socks_protocols,
     render_table,
     replace_extension,
     SameFileError,
@@ -260,7 +261,9 @@ class YoutubeDL(object):
     The following options determine which downloader is picked:
     external_downloader: Executable of the external downloader to call.
                        None or unset for standard (built-in) downloader.
-    hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv.
+    hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv
+                       if True, otherwise use ffmpeg/avconv if False, otherwise
+                       use downloader suggested by extractor if None.
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the downloader (see youtube_dl/downloader/common.py):
@@ -359,6 +362,8 @@ class YoutubeDL(object):
         for ph in self.params.get('progress_hooks', []):
             self.add_progress_hook(ph)
 
+        register_socks_protocols()
+
     def warn_if_short_id(self, argv):
         # short YouTube ID starting with dash?
         idxs = [
@@ -578,7 +583,7 @@ class YoutubeDL(object):
                 is_id=(k == 'id'))
             template_dict = dict((k, sanitize(k, v))
                                  for k, v in template_dict.items()
-                                 if v is not None)
+                                 if v is not None and not isinstance(v, (list, tuple, dict)))
             template_dict = collections.defaultdict(lambda: 'NA', template_dict)
 
             outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
@@ -715,6 +720,7 @@ class YoutubeDL(object):
         result_type = ie_result.get('_type', 'video')
 
         if result_type in ('url', 'url_transparent'):
+            ie_result['url'] = sanitize_url(ie_result['url'])
             extract_flat = self.params.get('extract_flat', False)
             if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
                     extract_flat is True):
@@ -1637,7 +1643,7 @@ class YoutubeDL(object):
                     # Just a single file
                     success = dl(filename, info_dict)
             except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                self.report_error('unable to download video data: %s' % str(err))
+                self.report_error('unable to download video data: %s' % error_to_compat_str(err))
                 return
             except (OSError, IOError) as err:
                 raise UnavailableVideoError(err)
@@ -2016,6 +2022,7 @@ class YoutubeDL(object):
         if opts_cookiefile is None:
             self.cookiejar = compat_cookiejar.CookieJar()
         else:
+            opts_cookiefile = compat_expanduser(opts_cookiefile)
             self.cookiejar = compat_cookiejar.MozillaCookieJar(
                 opts_cookiefile)
             if os.access(opts_cookiefile, os.R_OK):

youtube_dl/__init__.py

@@ -67,9 +67,9 @@ def _real_main(argv=None):
     # Custom HTTP headers
     if opts.headers is not None:
         for h in opts.headers:
-            if h.find(':', 1) < 0:
+            if ':' not in h:
                 parser.error('wrong header formatting, it should be key:value, not "%s"' % h)
-            key, value = h.split(':', 2)
+            key, value = h.split(':', 1)
             if opts.verbose:
                 write_string('[debug] Adding header from command line option %s:%s\n' % (key, value))
             std_headers[key] = value
@@ -86,7 +86,9 @@ def _real_main(argv=None):
             if opts.batchfile == '-':
                 batchfd = sys.stdin
             else:
-                batchfd = io.open(opts.batchfile, 'r', encoding='utf-8', errors='ignore')
+                batchfd = io.open(
+                    compat_expanduser(opts.batchfile),
+                    'r', encoding='utf-8', errors='ignore')
             batch_urls = read_batch_urls(batchfd)
             if opts.verbose:
                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
@@ -404,7 +406,7 @@ def _real_main(argv=None):
 
         try:
             if opts.load_info_filename is not None:
-                retcode = ydl.download_with_info_file(opts.load_info_filename)
+                retcode = ydl.download_with_info_file(compat_expanduser(opts.load_info_filename))
             else:
                 retcode = ydl.download(all_urls)
         except MaxDownloadsReached:

youtube_dl/compat.py

@@ -11,6 +11,7 @@ import re
 import shlex
 import shutil
 import socket
+import struct
 import subprocess
 import sys
 import itertools
@@ -340,9 +341,9 @@ except ImportError:  # Python 2
         return parsed_result
 
 try:
-    from shlex import quote as shlex_quote
+    from shlex import quote as compat_shlex_quote
 except ImportError:  # Python < 3.3
-    def shlex_quote(s):
+    def compat_shlex_quote(s):
         if re.match(r'^[-_\w./]+$', s):
             return s
         else:
@@ -373,6 +374,9 @@ compat_os_name = os._name if os.name == 'java' else os.name
 if sys.version_info >= (3, 0):
     compat_getenv = os.getenv
     compat_expanduser = os.path.expanduser
+
+    def compat_setenv(key, value, env=os.environ):
+        env[key] = value
 else:
     # Environment variables should be decoded with filesystem encoding.
     # Otherwise it will fail if any non-ASCII characters present (see #3854 #3217 #2918)
@@ -384,6 +388,12 @@ else:
             env = env.decode(get_filesystem_encoding())
         return env
 
+    def compat_setenv(key, value, env=os.environ):
+        def encode(v):
+            from .utils import get_filesystem_encoding
+            return v.encode(get_filesystem_encoding()) if isinstance(v, compat_str) else v
+        env[encode(key)] = encode(value)
+
     # HACK: The default implementations of os.path.expanduser from cpython do not decode
     # environment variables with filesystem encoding. We will work around this by
     # providing adjusted implementations.
@@ -456,18 +466,6 @@ else:
         print(s)
 
 
-try:
-    subprocess_check_output = subprocess.check_output
-except AttributeError:
-    def subprocess_check_output(*args, **kwargs):
-        assert 'input' not in kwargs
-        p = subprocess.Popen(*args, stdout=subprocess.PIPE, **kwargs)
-        output, _ = p.communicate()
-        ret = p.poll()
-        if ret:
-            raise subprocess.CalledProcessError(ret, p.args, output=output)
-        return output
-
 if sys.version_info < (3, 0) and sys.platform == 'win32':
     def compat_getpass(prompt, *args, **kwargs):
         if isinstance(prompt, compat_str):
@@ -583,6 +581,26 @@ if sys.version_info >= (3, 0):
 else:
     from tokenize import generate_tokens as compat_tokenize_tokenize
 
+
+try:
+    struct.pack('!I', 0)
+except TypeError:
+    # In Python 2.6 and 2.7.x < 2.7.7, struct requires a bytes argument
+    # See https://bugs.python.org/issue19099
+    def compat_struct_pack(spec, *args):
+        if isinstance(spec, compat_str):
+            spec = spec.encode('ascii')
+        return struct.pack(spec, *args)
+
+    def compat_struct_unpack(spec, *args):
+        if isinstance(spec, compat_str):
+            spec = spec.encode('ascii')
+        return struct.unpack(spec, *args)
+else:
+    compat_struct_pack = struct.pack
+    compat_struct_unpack = struct.unpack
+
+
 __all__ = [
     'compat_HTMLParser',
     'compat_HTTPError',
@@ -604,9 +622,13 @@ __all__ = [
     'compat_os_name',
     'compat_parse_qs',
     'compat_print',
+    'compat_setenv',
+    'compat_shlex_quote',
     'compat_shlex_split',
     'compat_socket_create_connection',
     'compat_str',
+    'compat_struct_pack',
+    'compat_struct_unpack',
     'compat_subprocess_get_DEVNULL',
     'compat_tokenize_tokenize',
     'compat_urllib_error',
@@ -623,7 +645,5 @@ __all__ = [
     'compat_urlretrieve',
     'compat_xml_parse_error',
     'compat_xpath',
-    'shlex_quote',
-    'subprocess_check_output',
     'workaround_optparse_bug9161',
 ]

youtube_dl/downloader/__init__.py

@@ -41,9 +41,12 @@ def get_suitable_downloader(info_dict, params={}):
         if ed.can_download(info_dict):
             return ed
 
-    if protocol == 'm3u8' and params.get('hls_prefer_native'):
+    if protocol == 'm3u8' and params.get('hls_prefer_native') is True:
         return HlsFD
 
+    if protocol == 'm3u8_native' and params.get('hls_prefer_native') is False:
+        return FFmpegFD
+
     return PROTOCOL_MAP.get(protocol, HttpFD)
 
 

youtube_dl/downloader/external.py

@@ -6,6 +6,7 @@ import sys
 import re
 
 from .common import FileDownloader
+from ..compat import compat_setenv
 from ..postprocessor.ffmpeg import FFmpegPostProcessor, EXT_TO_OUT_FORMATS
 from ..utils import (
     cli_option,
@@ -198,6 +199,18 @@ class FFmpegFD(ExternalFD):
                 '-headers',
                 ''.join('%s: %s\r\n' % (key, val) for key, val in headers.items())]
 
+        env = None
+        proxy = self.params.get('proxy')
+        if proxy:
+            if not re.match(r'^[\da-zA-Z]+://', proxy):
+                proxy = 'http://%s' % proxy
+            # Since December 2015 ffmpeg supports -http_proxy option (see
+            # http://git.videolan.org/?p=ffmpeg.git;a=commit;h=b4eb1f29ebddd60c41a2eb39f5af701e38e0d3fd)
+            # We could switch to the following code if we are able to detect version properly
+            # args += ['-http_proxy', proxy]
+            env = os.environ.copy()
+            compat_setenv('HTTP_PROXY', proxy, env=env)
+
         protocol = info_dict.get('protocol')
 
         if protocol == 'rtmp':
@@ -224,8 +237,8 @@ class FFmpegFD(ExternalFD):
                 args += ['-rtmp_live', 'live']
 
         args += ['-i', url, '-c', 'copy']
-        if protocol == 'm3u8':
+        if protocol in ('m3u8', 'm3u8_native'):
-            if self.params.get('hls_use_mpegts', False):
+            if self.params.get('hls_use_mpegts', False) or tmpfilename == '-':
                 args += ['-f', 'mpegts']
             else:
                 args += ['-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
@@ -239,7 +252,7 @@ class FFmpegFD(ExternalFD):
 
         self._debug_cmd(args)
 
-        proc = subprocess.Popen(args, stdin=subprocess.PIPE)
+        proc = subprocess.Popen(args, stdin=subprocess.PIPE, env=env)
         try:
             retval = proc.wait()
         except KeyboardInterrupt:

youtube_dl/downloader/f4m.py

@@ -12,37 +12,49 @@ from ..compat import (
     compat_urlparse,
     compat_urllib_error,
     compat_urllib_parse_urlparse,
+    compat_struct_pack,
+    compat_struct_unpack,
 )
 from ..utils import (
     encodeFilename,
     fix_xml_ampersands,
     sanitize_open,
-    struct_pack,
-    struct_unpack,
     xpath_text,
 )
 
 
+class DataTruncatedError(Exception):
+    pass
+
+
 class FlvReader(io.BytesIO):
     """
     Reader for Flv files
     The file format is documented in https://www.adobe.com/devnet/f4v.html
     """
 
+    def read_bytes(self, n):
+        data = self.read(n)
+        if len(data) < n:
+            raise DataTruncatedError(
+                'FlvReader error: need %d bytes while only %d bytes got' % (
+                    n, len(data)))
+        return data
+
     # Utility functions for reading numbers and strings
     def read_unsigned_long_long(self):
-        return struct_unpack('!Q', self.read(8))[0]
+        return compat_struct_unpack('!Q', self.read_bytes(8))[0]
 
     def read_unsigned_int(self):
-        return struct_unpack('!I', self.read(4))[0]
+        return compat_struct_unpack('!I', self.read_bytes(4))[0]
 
     def read_unsigned_char(self):
-        return struct_unpack('!B', self.read(1))[0]
+        return compat_struct_unpack('!B', self.read_bytes(1))[0]
 
     def read_string(self):
         res = b''
         while True:
-            char = self.read(1)
+            char = self.read_bytes(1)
             if char == b'\x00':
                 break
             res += char
@@ -53,18 +65,18 @@ class FlvReader(io.BytesIO):
         Read a box and return the info as a tuple: (box_size, box_type, box_data)
         """
         real_size = size = self.read_unsigned_int()
-        box_type = self.read(4)
+        box_type = self.read_bytes(4)
         header_end = 8
         if size == 1:
             real_size = self.read_unsigned_long_long()
             header_end = 16
-        return real_size, box_type, self.read(real_size - header_end)
+        return real_size, box_type, self.read_bytes(real_size - header_end)
 
     def read_asrt(self):
         # version
         self.read_unsigned_char()
         # flags
-        self.read(3)
+        self.read_bytes(3)
         quality_entry_count = self.read_unsigned_char()
         # QualityEntryCount
         for i in range(quality_entry_count):
@@ -85,7 +97,7 @@ class FlvReader(io.BytesIO):
         # version
         self.read_unsigned_char()
         # flags
-        self.read(3)
+        self.read_bytes(3)
         # time scale
         self.read_unsigned_int()
 
@@ -119,7 +131,7 @@ class FlvReader(io.BytesIO):
         # version
         self.read_unsigned_char()
         # flags
-        self.read(3)
+        self.read_bytes(3)
 
         self.read_unsigned_int()  # BootstrapinfoVersion
         # Profile,Live,Update,Reserved
@@ -194,11 +206,11 @@ def build_fragments_list(boot_info):
 
 
 def write_unsigned_int(stream, val):
-    stream.write(struct_pack('!I', val))
+    stream.write(compat_struct_pack('!I', val))
 
 
 def write_unsigned_int_24(stream, val):
-    stream.write(struct_pack('!I', val)[1:])
+    stream.write(compat_struct_pack('!I', val)[1:])
 
 
 def write_flv_header(stream):
@@ -374,7 +386,17 @@ class F4mFD(FragmentFD):
                 down.close()
                 reader = FlvReader(down_data)
                 while True:
+                    try:
                         _, box_type, box_data = reader.read_box_info()
+                    except DataTruncatedError:
+                        if test:
+                            # In tests, segments may be truncated, and thus
+                            # FlvReader may not be able to parse the whole
+                            # chunk. If so, write the segment as is
+                            # See https://github.com/rg3/youtube-dl/issues/9214
+                            dest_stream.write(down_data)
+                            break
+                        raise
                     if box_type == b'mdat':
                         dest_stream.write(box_data)
                         break

youtube_dl/downloader/hls.py

@@ -4,6 +4,7 @@ import os.path
 import re
 
 from .fragment import FragmentFD
+from .external import FFmpegFD
 
 from ..compat import compat_urlparse
 from ..utils import (
@@ -17,12 +18,39 @@ class HlsFD(FragmentFD):
 
     FD_NAME = 'hlsnative'
 
+    @staticmethod
+    def can_download(manifest):
+        UNSUPPORTED_FEATURES = (
+            r'#EXT-X-KEY:METHOD=(?!NONE)',  # encrypted streams [1]
+            r'#EXT-X-BYTERANGE',  # playlists composed of byte ranges of media files [2]
+            # Live streams heuristic does not always work (e.g. geo restricted to Germany
+            # http://hls-geo.daserste.de/i/videoportal/Film/c_620000/622873/format,716451,716457,716450,716458,716459,.mp4.csmil/index_4_av.m3u8?null=0)
+            # r'#EXT-X-MEDIA-SEQUENCE:(?!0$)',  # live streams [3]
+            r'#EXT-X-PLAYLIST-TYPE:EVENT',  # media segments may be appended to the end of
+                                            # event media playlists [4]
+            # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.4
+            # 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.2
+            # 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.2
+            # 4. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.5
+        )
+        return all(not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES)
+
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         self.to_screen('[%s] Downloading m3u8 manifest' % self.FD_NAME)
         manifest = self.ydl.urlopen(man_url).read()
 
         s = manifest.decode('utf-8', 'ignore')
+
+        if not self.can_download(s):
+            self.report_warning(
+                'hlsnative has detected features it does not support, '
+                'extraction will be delegated to ffmpeg')
+            fd = FFmpegFD(self.ydl, self.params)
+            for ph in self._progress_hooks:
+                fd.add_progress_hook(ph)
+            return fd.real_download(filename, info_dict)
+
         fragment_urls = []
         for line in s.splitlines():
             line = line.strip()

youtube_dl/downloader/rtsp.py

@@ -27,6 +27,8 @@ class RtspFD(FileDownloader):
             self.report_error('MMS or RTSP download detected but neither "mplayer" nor "mpv" could be run. Please install any.')
             return False
 
+        self._debug_cmd(args)
+
         retval = subprocess.call(args)
         if retval == 0:
             fsize = os.path.getsize(encodeFilename(tmpfilename))

youtube_dl/extractor/aol.py

@@ -1,26 +1,113 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+)
 
 
 class AolIE(InfoExtractor):
     IE_NAME = 'on.aol.com'
-    _VALID_URL = r'(?:aol-video:|https?://on\.aol\.com/video/.*-)(?P<id>[0-9]+)(?:$|\?)'
+    _VALID_URL = r'(?:aol-video:|https?://on\.aol\.com/(?:[^/]+/)*(?:[^/?#&]+-)?)(?P<id>[^/?#&]+)'
 
     _TESTS = [{
+        # video with 5min ID
         'url': 'http://on.aol.com/video/u-s--official-warns-of-largest-ever-irs-phone-scam-518167793?icid=OnHomepageC2Wide_MustSee_Img',
         'md5': '18ef68f48740e86ae94b98da815eec42',
         'info_dict': {
             'id': '518167793',
             'ext': 'mp4',
             'title': 'U.S. Official Warns Of \'Largest Ever\' IRS Phone Scam',
+            'description': 'A major phone scam has cost thousands of taxpayers more than $1 million, with less than a month until income tax returns are due to the IRS.',
+            'timestamp': 1395405060,
+            'upload_date': '20140321',
+            'uploader': 'Newsy Studio',
         },
-        'add_ie': ['FiveMin'],
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
+    }, {
+        # video with vidible ID
+        'url': 'http://on.aol.com/video/netflix-is-raising-rates-5707d6b8e4b090497b04f706?context=PC:homepage:PL1944:1460189336183',
+        'info_dict': {
+            'id': '5707d6b8e4b090497b04f706',
+            'ext': 'mp4',
+            'title': 'Netflix is Raising Rates',
+            'description': 'Netflix is rewarding millions of it’s long-standing members with an increase in cost. Veuer’s Carly Figueroa has more.',
+            'upload_date': '20160408',
+            'timestamp': 1460123280,
+            'uploader': 'Veuer',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
+    }, {
+        'url': 'http://on.aol.com/partners/abc-551438d309eab105804dbfe8/sneak-peek-was-haley-really-framed-570eaebee4b0448640a5c944',
+        'only_matching': True,
+    }, {
+        'url': 'http://on.aol.com/shows/park-bench-shw518173474-559a1b9be4b0c3bfad3357a7?context=SH:SHW518173474:PL4327:1460619712763',
+        'only_matching': True,
+    }, {
+        'url': 'http://on.aol.com/video/519442220',
+        'only_matching': True,
+    }, {
+        'url': 'aol-video:5707d6b8e4b090497b04f706',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        return self.url_result('5min:%s' % video_id)
+
+        response = self._download_json(
+            'https://feedapi.b2c.on.aol.com/v1.0/app/videos/aolon/%s/details' % video_id,
+            video_id)['response']
+        if response['statusText'] != 'Ok':
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, response['statusText']), expected=True)
+
+        video_data = response['data']
+        formats = []
+        m3u8_url = video_data.get('videoMasterPlaylist')
+        if m3u8_url:
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
+        for rendition in video_data.get('renditions', []):
+            video_url = rendition.get('url')
+            if not video_url:
+                continue
+            ext = rendition.get('format')
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
+            else:
+                f = {
+                    'url': video_url,
+                    'format_id': rendition.get('quality'),
+                }
+                mobj = re.search(r'(\d+)x(\d+)', video_url)
+                if mobj:
+                    f.update({
+                        'width': int(mobj.group(1)),
+                        'height': int(mobj.group(2)),
+                    })
+                formats.append(f)
+        self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
+
+        return {
+            'id': video_id,
+            'title': video_data['title'],
+            'duration': int_or_none(video_data.get('duration')),
+            'timestamp': int_or_none(video_data.get('publishDate')),
+            'view_count': int_or_none(video_data.get('views')),
+            'description': video_data.get('description'),
+            'uploader': video_data.get('videoOwner'),
+            'formats': formats,
+        }
 
 
 class AolFeaturesIE(InfoExtractor):

youtube_dl/extractor/ard.py

@@ -83,7 +83,7 @@ class ARDMediathekIE(InfoExtractor):
         subtitle_url = media_info.get('_subtitleUrl')
         if subtitle_url:
             subtitles['de'] = [{
-                'ext': 'srt',
+                'ext': 'ttml',
                 'url': subtitle_url,
             }]
 

youtube_dl/extractor/arte.py

@@ -63,7 +63,7 @@ class ArteTvIE(InfoExtractor):
 
 class ArteTVPlus7IE(InfoExtractor):
     IE_NAME = 'arte.tv:+7'
-    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&+])'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&]+)'
 
     @classmethod
     def _extract_url_info(cls, url):
@@ -161,24 +161,53 @@ class ArteTVPlus7IE(InfoExtractor):
             'es': 'E[ESP]',
         }
 
+        langcode = LANGS.get(lang, lang)
+
         formats = []
         for format_id, format_dict in player_info['VSR'].items():
             f = dict(format_dict)
             versionCode = f.get('versionCode')
-            langcode = LANGS.get(lang, lang)
+            l = re.escape(langcode)
-            lang_rexs = [r'VO?%s-' % re.escape(langcode), r'VO?.-ST%s$' % re.escape(langcode)]
+
-            lang_pref = None
+            # Language preference from most to least priority
-            if versionCode:
+            # Reference: section 5.6.3 of
-                matched_lang_rexs = [r for r in lang_rexs if re.match(r, versionCode)]
+            # http://www.arte.tv/sites/en/corporate/files/complete-technical-guidelines-arte-geie-v1-05.pdf
-                lang_pref = -10 if not matched_lang_rexs else 10 * len(matched_lang_rexs)
+            PREFERENCES = (
-            source_pref = 0
+                # original version in requested language, without subtitles
-            if versionCode is not None:
+                r'VO{0}$'.format(l),
-                # The original version with subtitles has lower relevance
+                # original version in requested language, with partial subtitles in requested language
-                if re.match(r'VO-ST(F|A|E)', versionCode):
+                r'VO{0}-ST{0}$'.format(l),
-                    source_pref -= 10
+                # original version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
-                # The version with sourds/mal subtitles has also lower relevance
+                r'VO{0}-STM{0}$'.format(l),
-                elif re.match(r'VO?(F|A|E)-STM\1', versionCode):
+                # non-original (dubbed) version in requested language, without subtitles
-                    source_pref -= 9
+                r'V{0}$'.format(l),
+                # non-original (dubbed) version in requested language, with subtitles partial subtitles in requested language
+                r'V{0}-ST{0}$'.format(l),
+                # non-original (dubbed) version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
+                r'V{0}-STM{0}$'.format(l),
+                # original version in requested language, with partial subtitles in different language
+                r'VO{0}-ST(?!{0}).+?$'.format(l),
+                # original version in requested language, with subtitles for the deaf and hard-of-hearing in different language
+                r'VO{0}-STM(?!{0}).+?$'.format(l),
+                # original version in different language, with partial subtitles in requested language
+                r'VO(?:(?!{0}).+?)?-ST{0}$'.format(l),
+                # original version in different language, with subtitles for the deaf and hard-of-hearing in requested language
+                r'VO(?:(?!{0}).+?)?-STM{0}$'.format(l),
+                # original version in different language, without subtitles
+                r'VO(?:(?!{0}))?$'.format(l),
+                # original version in different language, with partial subtitles in different language
+                r'VO(?:(?!{0}).+?)?-ST(?!{0}).+?$'.format(l),
+                # original version in different language, with subtitles for the deaf and hard-of-hearing in different language
+                r'VO(?:(?!{0}).+?)?-STM(?!{0}).+?$'.format(l),
+            )
+
+            for pref, p in enumerate(PREFERENCES):
+                if re.match(p, versionCode):
+                    lang_pref = len(PREFERENCES) - pref
+                    break
+            else:
+                lang_pref = -1
+
             format = {
                 'format_id': format_id,
                 'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
@@ -188,7 +217,6 @@ class ArteTVPlus7IE(InfoExtractor):
                 'height': int_or_none(f.get('height')),
                 'tbr': int_or_none(f.get('bitrate')),
                 'quality': qfunc(f.get('quality')),
-                'source_preference': source_pref,
             }
 
             if f.get('mediaType') == 'rtmp':
@@ -210,7 +238,7 @@ class ArteTVPlus7IE(InfoExtractor):
 # It also uses the arte_vp_url url from the webpage to extract the information
 class ArteTVCreativeIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:creative'
-    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:magazine?/)?(?P<id>[^/?#&]+)'
+    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
     _TESTS = [{
         'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
@@ -229,9 +257,27 @@ class ArteTVCreativeIE(ArteTVPlus7IE):
             'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
             'upload_date': '20140805',
         }
+    }, {
+        'url': 'http://creative.arte.tv/de/episode/agentur-amateur-4-der-erste-kunde',
+        'only_matching': True,
     }]
 
 
+class ArteTVInfoIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:info'
+    _VALID_URL = r'https?://info\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+
+    _TEST = {
+        'url': 'http://info.arte.tv/fr/service-civique-un-cache-misere',
+        'info_dict': {
+            'id': '067528-000-A',
+            'ext': 'mp4',
+            'title': 'Service civique, un cache misère ?',
+            'upload_date': '20160403',
+        },
+    }
+
+
 class ArteTVFutureIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:future'
     _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
@@ -337,7 +383,7 @@ class ArteTVEmbedIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:embed'
     _VALID_URL = r'''(?x)
         http://www\.arte\.tv
-        /playerv2/embed\.php\?json_url=
+        /(?:playerv2/embed|arte_vp/index)\.php\?json_url=
         (?P<json_url>
             http://arte\.tv/papi/tvguide/videos/stream/player/
             (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*

youtube_dl/extractor/audiomack.py

@@ -30,14 +30,14 @@ class AudiomackIE(InfoExtractor):
         # audiomack wrapper around soundcloud song
         {
             'add_ie': ['Soundcloud'],
-            'url': 'http://www.audiomack.com/song/xclusiveszone/take-kare',
+            'url': 'http://www.audiomack.com/song/hip-hop-daily/black-mamba-freestyle',
             'info_dict': {
-                'id': '172419696',
+                'id': '258901379',
                 'ext': 'mp3',
-                'description': 'md5:1fc3272ed7a635cce5be1568c2822997',
+                'description': 'mamba day freestyle for the legend Kobe Bryant ',
-                'title': 'Young Thug ft Lil Wayne - Take Kare',
+                'title': 'Black Mamba Freestyle [Prod. By Danny Wolf]',
-                'uploader': 'Young Thug World',
+                'uploader': 'ILOVEMAKONNEN',
-                'upload_date': '20141016',
+                'upload_date': '20160414',
             }
         },
     ]

youtube_dl/extractor/bbc.py

@@ -671,6 +671,7 @@ class BBCIE(BBCCoUkIE):
         'info_dict': {
             'id': '34475836',
             'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
+            'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
         },
         'playlist_count': 3,
     }, {

youtube_dl/extractor/biqle.py

@@ -0,0 +1,39 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class BIQLEIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?biqle\.(?:com|org|ru)/watch/(?P<id>-?\d+_\d+)'
+    _TESTS = [{
+        'url': 'http://www.biqle.ru/watch/847655_160197695',
+        'md5': 'ad5f746a874ccded7b8f211aeea96637',
+        'info_dict': {
+            'id': '160197695',
+            'ext': 'mp4',
+            'title': 'Foo Fighters - The Pretender (Live at Wembley Stadium)',
+            'uploader': 'Andrey Rogozin',
+            'upload_date': '20110605',
+        }
+    }, {
+        'url': 'https://biqle.org/watch/-44781847_168547604',
+        'md5': '7f24e72af1db0edf7c1aaba513174f97',
+        'info_dict': {
+            'id': '168547604',
+            'ext': 'mp4',
+            'title': 'Ребенок в шоке от автоматической мойки',
+            'uploader': 'Dmitry Kotov',
+        }
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        embed_url = self._proto_relative_url(self._search_regex(
+            r'<iframe.+?src="((?:http:)?//daxab\.com/[^"]+)".*?></iframe>', webpage, 'embed url'))
+
+        return {
+            '_type': 'url_transparent',
+            'url': embed_url,
+        }

youtube_dl/extractor/bloomberg.py

@@ -17,6 +17,9 @@ class BloombergIE(InfoExtractor):
             'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',
             'description': 'md5:a8ba0302912d03d246979735c17d2761',
         },
+        '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,

youtube_dl/extractor/brightcove.py

@@ -307,9 +307,10 @@ class BrightcoveLegacyIE(InfoExtractor):
                                     playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
 
     def _extract_video_info(self, video_info):
+        video_id = compat_str(video_info['id'])
         publisher_id = video_info.get('publisherId')
         info = {
-            'id': compat_str(video_info['id']),
+            'id': video_id,
             'title': video_info['displayName'].strip(),
             'description': video_info.get('shortDescription'),
             'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
@@ -331,7 +332,8 @@ class BrightcoveLegacyIE(InfoExtractor):
                     url_comp = compat_urllib_parse_urlparse(url)
                     if url_comp.path.endswith('.m3u8'):
                         formats.extend(
-                            self._extract_m3u8_formats(url, info['id'], 'mp4'))
+                            self._extract_m3u8_formats(
+                                url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
                         continue
                     elif 'akamaihd.net' in url_comp.netloc:
                         # This type of renditions are served through
@@ -340,7 +342,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                         ext = 'flv'
                 if ext is None:
                     ext = determine_ext(url)
-                tbr = int_or_none(rend.get('encodingRate'), 1000),
+                tbr = int_or_none(rend.get('encodingRate'), 1000)
                 a_format = {
                     'format_id': 'http%s' % ('-%s' % tbr if tbr else ''),
                     'url': url,
@@ -365,7 +367,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                     a_format.update({
                         'format_id': 'hls%s' % ('-%s' % tbr if tbr else ''),
                         'ext': 'mp4',
-                        'protocol': 'm3u8',
+                        'protocol': 'm3u8_native',
                     })
 
                 formats.append(a_format)
@@ -395,7 +397,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                     return ad_info
 
         if 'url' not in info and not info.get('formats'):
-            raise ExtractorError('Unable to extract video url for %s' % info['id'])
+            raise ExtractorError('Unable to extract video url for %s' % video_id)
         return info
 
 
@@ -527,7 +529,7 @@ class BrightcoveNewIE(InfoExtractor):
                 if not src:
                     continue
                 formats.extend(self._extract_m3u8_formats(
-                    src, video_id, 'mp4', m3u8_id='hls', fatal=False))
+                    src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
             elif source_type == 'application/dash+xml':
                 if not src:
                     continue

youtube_dl/extractor/cbc.py

@@ -33,6 +33,7 @@ class CBCIE(InfoExtractor):
             'title': 'Robin Williams freestyles on 90 Minutes Live',
             'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.',
             'upload_date': '19700101',
+            'uploader': 'CBCC-NEW',
         },
         'params': {
             # rtmp download

youtube_dl/extractor/cbs.py

@@ -5,7 +5,6 @@ from ..utils import (
     xpath_text,
     xpath_element,
     int_or_none,
-    ExtractorError,
     find_xpath_attr,
 )
 
@@ -64,7 +63,7 @@ class CBSIE(CBSBaseIE):
         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
         'only_matching': True,
     }]
-    TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?manifest=m3u&mbr=true'
+    TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
@@ -84,11 +83,11 @@ class CBSIE(CBSBaseIE):
             pid = xpath_text(item, 'pid')
             if not pid:
                 continue
-            try:
+            tp_release_url = self.TP_RELEASE_URL_TEMPLATE % pid
+            if '.m3u8' in xpath_text(item, 'contentUrl', default=''):
+                tp_release_url += '&manifest=m3u'
             tp_formats, tp_subtitles = self._extract_theplatform_smil(
-                    self.TP_RELEASE_URL_TEMPLATE % pid, content_id, 'Downloading %s SMIL data' % pid)
+                tp_release_url, content_id, 'Downloading %s SMIL data' % pid)
-            except ExtractorError:
-                continue
             formats.extend(tp_formats)
             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
         self._sort_formats(formats)

youtube_dl/extractor/ccc.py

@@ -1,13 +1,9 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
-    parse_duration,
+    parse_iso8601,
-    qualities,
-    unified_strdate,
 )
 
 
@@ -19,14 +15,14 @@ class CCCIE(InfoExtractor):
         'url': 'https://media.ccc.de/v/30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor#video',
         'md5': '3a1eda8f3a29515d27f5adb967d7e740',
         'info_dict': {
-            'id': '30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor',
+            'id': '1839',
             'ext': 'mp4',
             'title': 'Introduction to Processor Design',
-            'description': 'md5:80be298773966f66d56cb11260b879af',
+            'description': 'md5:df55f6d073d4ceae55aae6f2fd98a0ac',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'view_count': int,
             'upload_date': '20131228',
-            'duration': 3660,
+            'timestamp': 1388188800,
+            'duration': 3710,
         }
     }, {
         'url': 'https://media.ccc.de/v/32c3-7368-shopshifting#download',
@@ -34,79 +30,48 @@ class CCCIE(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)
+        event_id = self._search_regex("data-id='(\d+)'", webpage, 'event id')
+        event_data = self._download_json('https://media.ccc.de/public/events/%s' % event_id, event_id)
 
-        if self._downloader.params.get('prefer_free_formats'):
-            preference = qualities(['mp3', 'opus', 'mp4-lq', 'webm-lq', 'h264-sd', 'mp4-sd', 'webm-sd', 'mp4', 'webm', 'mp4-hd', 'h264-hd', 'webm-hd'])
-        else:
-            preference = qualities(['opus', 'mp3', 'webm-lq', 'mp4-lq', 'webm-sd', 'h264-sd', 'mp4-sd', 'webm', 'mp4', 'webm-hd', 'mp4-hd', 'h264-hd'])
-
-        title = self._html_search_regex(
-            r'(?s)<h1>(.*?)</h1>', webpage, 'title')
-        description = self._html_search_regex(
-            r'(?s)<h3>About</h3>(.+?)<h3>',
-            webpage, 'description', fatal=False)
-        upload_date = unified_strdate(self._html_search_regex(
-            r"(?s)<span[^>]+class='[^']*fa-calendar-o'[^>]*>(.+?)</span>",
-            webpage, 'upload date', fatal=False))
-        view_count = int_or_none(self._html_search_regex(
-            r"(?s)<span class='[^']*fa-eye'></span>(.*?)</li>",
-            webpage, 'view count', fatal=False))
-        duration = parse_duration(self._html_search_regex(
-            r'(?s)<span[^>]+class=(["\']).*?fa-clock-o.*?\1[^>]*></span>(?P<duration>.+?)</li',
-            webpage, 'duration', fatal=False, group='duration'))
-
-        matches = re.finditer(r'''(?xs)
-            <(?:span|div)\s+class='label\s+filetype'>(?P<format>[^<]*)</(?:span|div)>\s*
-            <(?:span|div)\s+class='label\s+filetype'>(?P<lang>[^<]*)</(?:span|div)>\s*
-            <a\s+download\s+href='(?P<http_url>[^']+)'>\s*
-            (?:
-                .*?
-                <a\s+(?:download\s+)?href='(?P<torrent_url>[^']+\.torrent)'
-            )?''', webpage)
         formats = []
-        for m in matches:
+        for recording in event_data.get('recordings', []):
-            format = m.group('format')
+            recording_url = recording.get('recording_url')
-            format_id = self._search_regex(
+            if not recording_url:
-                r'.*/([a-z0-9_-]+)/[^/]*$',
+                continue
-                m.group('http_url'), 'format id', default=None)
+            language = recording.get('language')
-            if format_id:
+            folder = recording.get('folder')
-                format_id = m.group('lang') + '-' + format_id
+            format_id = None
-            vcodec = 'h264' if 'h264' in format_id else (
+            if language:
-                'none' if format_id in ('mp3', 'opus') else None
+                format_id = language
+            if folder:
+                if language:
+                    format_id += '-' + folder
+                else:
+                    format_id = folder
+            vcodec = 'h264' if 'h264' in folder else (
+                'none' if folder in ('mp3', 'opus') else None
             )
             formats.append({
                 'format_id': format_id,
-                'format': format,
+                'url': recording_url,
-                'language': m.group('lang'),
+                'width': int_or_none(recording.get('width')),
-                'url': m.group('http_url'),
+                'height': int_or_none(recording.get('height')),
+                'filesize': int_or_none(recording.get('size'), invscale=1024 * 1024),
+                'language': language,
                 'vcodec': vcodec,
-                'preference': preference(format_id),
-            })
-
-            if m.group('torrent_url'):
-                formats.append({
-                    'format_id': 'torrent-%s' % (format if format_id is None else format_id),
-                    'format': '%s (torrent)' % format,
-                    'proto': 'torrent',
-                    'format_note': '(unsupported; will just download the .torrent file)',
-                    'vcodec': vcodec,
-                    'preference': -100 + preference(format_id),
-                    'url': m.group('torrent_url'),
             })
         self._sort_formats(formats)
 
-        thumbnail = self._html_search_regex(
-            r"<video.*?poster='([^']+)'", webpage, 'thumbnail', fatal=False)
-
         return {
-            'id': video_id,
+            'id': event_id,
-            'title': title,
+            'display_id': display_id,
-            'description': description,
+            'title': event_data['title'],
-            'thumbnail': thumbnail,
+            'description': event_data.get('description'),
-            'view_count': view_count,
+            'thumbnail': event_data.get('thumb_url'),
-            'upload_date': upload_date,
+            'timestamp': parse_iso8601(event_data.get('date')),
-            'duration': duration,
+            'duration': int_or_none(event_data.get('length')),
+            'tags': event_data.get('tags'),
             'formats': formats,
         }

youtube_dl/extractor/ceskatelevize.py

@@ -33,19 +33,33 @@ class CeskaTelevizeIE(InfoExtractor):
             'skip_download': True,
         },
     }, {
-        'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
+        'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
         'info_dict': {
-            'id': '61924494876844374',
+            'id': '61924494877028507',
             'ext': 'mp4',
-            'title': 'První republika: Zpěvačka z Dupárny Bobina',
+            'title': 'Hyde Park Civilizace: Bonus 01 - En',
-            'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.',
+            'description': 'English Subtittles',
             'thumbnail': 're:^https?://.*\.jpg',
-            'duration': 88.4,
+            'duration': 81.3,
         },
         'params': {
             # m3u8 download
             'skip_download': True,
         },
+    }, {
+        # live stream
+        'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
+        'info_dict': {
+            'id': 402,
+            'ext': 'mp4',
+            'title': 're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
+            'is_live': True,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'skip': 'Georestricted to Czech Republic',
     }, {
         # video with 18+ caution trailer
         'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
@@ -118,19 +132,21 @@ class CeskaTelevizeIE(InfoExtractor):
         req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
         req.add_header('Referer', url)
 
-        playlist_title = self._og_search_title(webpage)
+        playlist_title = self._og_search_title(webpage, default=None)
-        playlist_description = self._og_search_description(webpage)
+        playlist_description = self._og_search_description(webpage, default=None)
 
         playlist = self._download_json(req, playlist_id)['playlist']
         playlist_len = len(playlist)
 
         entries = []
         for item in playlist:
+            is_live = item.get('type') == 'LIVE'
             formats = []
             for format_id, stream_url in item['streamUrls'].items():
                 formats.extend(self._extract_m3u8_formats(
                     stream_url, playlist_id, 'mp4',
-                    entry_protocol='m3u8_native', fatal=False))
+                    entry_protocol='m3u8' if is_live else 'm3u8_native',
+                    fatal=False))
             self._sort_formats(formats)
 
             item_id = item.get('id') or item['assetId']
@@ -145,14 +161,22 @@ class CeskaTelevizeIE(InfoExtractor):
                 if subs:
                     subtitles = self.extract_subtitles(episode_id, subs)
 
+            if playlist_len == 1:
+                final_title = playlist_title or title
+                if is_live:
+                    final_title = self._live_title(final_title)
+            else:
+                final_title = '%s (%s)' % (playlist_title, title)
+
             entries.append({
                 'id': item_id,
-                'title': playlist_title if playlist_len == 1 else '%s (%s)' % (playlist_title, title),
+                'title': final_title,
                 'description': playlist_description if playlist_len == 1 else None,
                 'thumbnail': thumbnail,
                 'duration': duration,
                 'formats': formats,
                 'subtitles': subtitles,
+                'is_live': is_live,
             })
 
         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)

youtube_dl/extractor/cinemassacre.py

@@ -1,119 +0,0 @@
-# encoding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import ExtractorError
-from .screenwavemedia import ScreenwaveMediaIE
-
-
-class CinemassacreIE(InfoExtractor):
-    _VALID_URL = 'https?://(?:www\.)?cinemassacre\.com/(?P<date_y>[0-9]{4})/(?P<date_m>[0-9]{2})/(?P<date_d>[0-9]{2})/(?P<display_id>[^?#/]+)'
-    _TESTS = [
-        {
-            'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
-            'md5': 'fde81fbafaee331785f58cd6c0d46190',
-            'info_dict': {
-                'id': 'Cinemassacre-19911',
-                'ext': 'mp4',
-                'upload_date': '20121110',
-                'title': '“Angry Video Game Nerd: The Movie” – Trailer',
-                'description': 'md5:fb87405fcb42a331742a0dce2708560b',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        },
-        {
-            'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
-            'md5': 'd72f10cd39eac4215048f62ab477a511',
-            'info_dict': {
-                'id': 'Cinemassacre-521be8ef82b16',
-                'ext': 'mp4',
-                'upload_date': '20131002',
-                'title': 'The Mummy’s Hand (1940)',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        },
-        {
-            # Youtube embedded video
-            'url': 'http://cinemassacre.com/2006/12/07/chronologically-confused-about-bad-movie-and-video-game-sequel-titles/',
-            'md5': 'ec9838a5520ef5409b3e4e42fcb0a3b9',
-            'info_dict': {
-                'id': 'OEVzPCY2T-g',
-                'ext': 'webm',
-                'title': 'AVGN: Chronologically Confused about Bad Movie and Video Game Sequel Titles',
-                'upload_date': '20061207',
-                'uploader': 'Cinemassacre',
-                'uploader_id': 'JamesNintendoNerd',
-                'description': 'md5:784734696c2b8b7f4b8625cc799e07f6',
-            }
-        },
-        {
-            # Youtube embedded video
-            'url': 'http://cinemassacre.com/2006/09/01/mckids/',
-            'md5': '7393c4e0f54602ad110c793eb7a6513a',
-            'info_dict': {
-                'id': 'FnxsNhuikpo',
-                'ext': 'webm',
-                'upload_date': '20060901',
-                'uploader': 'Cinemassacre Extra',
-                'description': 'md5:de9b751efa9e45fbaafd9c8a1123ed53',
-                'uploader_id': 'Cinemassacre',
-                'title': 'AVGN: McKids',
-            }
-        },
-        {
-            'url': 'http://cinemassacre.com/2015/05/25/mario-kart-64-nintendo-64-james-mike-mondays/',
-            'md5': '1376908e49572389e7b06251a53cdd08',
-            'info_dict': {
-                'id': 'Cinemassacre-555779690c440',
-                'ext': 'mp4',
-                'description': 'Let’s Play Mario Kart 64 !! Mario Kart 64 is a classic go-kart racing game released for the Nintendo 64 (N64). Today James & Mike do 4 player Battle Mode with Kyle and Bootsy!',
-                'title': 'Mario Kart 64 (Nintendo 64) James & Mike Mondays',
-                'upload_date': '20150525',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        }
-    ]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('display_id')
-        video_date = mobj.group('date_y') + mobj.group('date_m') + mobj.group('date_d')
-
-        webpage = self._download_webpage(url, display_id)
-
-        playerdata_url = self._search_regex(
-            [
-                ScreenwaveMediaIE.EMBED_PATTERN,
-                r'<iframe[^>]+src="(?P<url>(?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"',
-            ],
-            webpage, 'player data URL', default=None, group='url')
-        if not playerdata_url:
-            raise ExtractorError('Unable to find player data')
-
-        video_title = self._html_search_regex(
-            r'<title>(?P<title>.+?)\|', webpage, 'title')
-        video_description = self._html_search_regex(
-            r'<div class="entry-content">(?P<description>.+?)</div>',
-            webpage, 'description', flags=re.DOTALL, fatal=False)
-        video_thumbnail = self._og_search_thumbnail(webpage)
-
-        return {
-            '_type': 'url_transparent',
-            'display_id': display_id,
-            'title': video_title,
-            'description': video_description,
-            'upload_date': video_date,
-            'thumbnail': video_thumbnail,
-            'url': playerdata_url,
-        }

youtube_dl/extractor/cliprs.py

@@ -0,0 +1,90 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    int_or_none,
+    parse_iso8601,
+)
+
+
+class ClipRsIE(InfoExtractor):
+    _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',
+        'md5': 'c412d57815ba07b56f9edc7b5d6a14e5',
+        'info_dict': {
+            'id': '1488842.1399140381',
+            'ext': 'mp4',
+            'title': 'PREMIJERA Frajle predstavljaju novi spot za pesmu Moli me, moli',
+            'description': 'md5:56ce2c3b4ab31c5a2e0b17cb9a453026',
+            'duration': 229,
+            'timestamp': 1459850243,
+            'upload_date': '20160405',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_id = self._search_regex(
+            r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
+
+        response = self._download_json(
+            'http://qi.ckm.onetapi.pl/', video_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')
+        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

@@ -19,7 +19,7 @@ from ..utils import (
 class CloudyIE(InfoExtractor):
     _IE_DESC = 'cloudy.ec and videoraj.ch'
     _VALID_URL = r'''(?x)
-        https?://(?:www\.)?(?P<host>cloudy\.ec|videoraj\.ch)/
+        https?://(?:www\.)?(?P<host>cloudy\.ec|videoraj\.(?:ch|to))/
         (?:v/|embed\.php\?id=)
         (?P<id>[A-Za-z0-9]+)
         '''
@@ -37,7 +37,7 @@ class CloudyIE(InfoExtractor):
             }
         },
         {
-            'url': 'http://www.videoraj.ch/v/47f399fd8bb60',
+            'url': 'http://www.videoraj.to/v/47f399fd8bb60',
             'md5': '7d0f8799d91efd4eda26587421c3c3b0',
             'info_dict': {
                 'id': '47f399fd8bb60',

youtube_dl/extractor/common.py

@@ -163,7 +163,7 @@ class InfoExtractor(object):
     description:    Full video description.
     uploader:       Full name of the video uploader.
     license:        License name the video is licensed under.
-    creator:        The main artist who created the video.
+    creator:        The creator of the video.
     release_date:   The date (YYYYMMDD) when the video was released.
     timestamp:      UNIX timestamp of the moment the video became available.
     upload_date:    Video upload date (YYYYMMDD).
@@ -376,14 +376,13 @@ class InfoExtractor(object):
                 self.to_screen('%s' % (note,))
             else:
                 self.to_screen('%s: %s' % (video_id, note))
-        # data, headers and query params will be ignored for `Request` objects
         if isinstance(url_or_request, compat_urllib_request.Request):
             url_or_request = update_Request(
                 url_or_request, data=data, headers=headers, query=query)
         else:
             if query:
                 url_or_request = update_url_query(url_or_request, query)
-            if data or headers:
+            if data is not None or headers:
                 url_or_request = sanitized_Request(url_or_request, data, headers)
         try:
             return self._downloader.urlopen(url_or_request)
@@ -1007,6 +1006,13 @@ class InfoExtractor(object):
     def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, f4m_id=None,
                            transform_source=lambda s: fix_xml_ampersands(s).strip(),
                            fatal=True):
+        # currently youtube-dl cannot decode the playerVerificationChallenge as Akamai uses Adobe Alchemy
+        akamai_pv = manifest.find('{http://ns.adobe.com/f4m/1.0}pv-2.0')
+        if akamai_pv is not None and ';' in akamai_pv.text:
+            playerVerificationChallenge = akamai_pv.text.split(';')[0]
+            if playerVerificationChallenge.strip() != '':
+                return []
+
         formats = []
         manifest_version = '1.0'
         media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
@@ -1055,7 +1061,7 @@ class InfoExtractor(object):
     def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
                               entry_protocol='m3u8', preference=None,
                               m3u8_id=None, note=None, errnote=None,
-                              fatal=True):
+                              fatal=True, live=False):
 
         formats = [{
             'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
@@ -1133,6 +1139,10 @@ class InfoExtractor(object):
                 if m3u8_id:
                     format_id.append(m3u8_id)
                 last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None
+                # Bandwidth of live streams may differ over time thus making
+                # format_id unpredictable. So it's better to keep provided
+                # format_id intact.
+                if not live:
                     format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats)))
                 f = {
                     'format_id': '-'.join(format_id),

youtube_dl/extractor/crunchyroll.py

@@ -11,7 +11,6 @@ from math import pow, sqrt, floor
 from .common import InfoExtractor
 from ..compat import (
     compat_etree_fromstring,
-    compat_urllib_parse_unquote,
     compat_urllib_parse_urlencode,
     compat_urllib_request,
     compat_urlparse,
@@ -27,6 +26,7 @@ from ..utils import (
     unified_strdate,
     urlencode_postdata,
     xpath_text,
+    extract_attributes,
 )
 from ..aes import (
     aes_cbc_decrypt,
@@ -306,28 +306,36 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage,
             'video_uploader', fatal=False)
 
-        playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
+        available_fmts = []
-        playerdata_req = sanitized_Request(playerdata_url)
+        for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
-        playerdata_req.data = urlencode_postdata({'current_page': webpage_url})
+            attrs = extract_attributes(a)
-        playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+            href = attrs.get('href')
-        playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
+            if href and '/freetrial' in href:
-
+                continue
-        stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
+            available_fmts.append(fmt)
-        video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
+        if not available_fmts:
-
+            for p in (r'token=["\']showmedia\.([0-9]{3,4})p"', r'showmedia\.([0-9]{3,4})p'):
+                available_fmts = re.findall(p, webpage)
+                if available_fmts:
+                    break
+        video_encode_ids = []
         formats = []
-        for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
+        for fmt in available_fmts:
             stream_quality, stream_format = self._FORMAT_IDS[fmt]
             video_format = fmt + 'p'
             streamdata_req = sanitized_Request(
                 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
-                % (stream_id, stream_format, stream_quality),
+                % (video_id, stream_format, stream_quality),
                 compat_urllib_parse_urlencode({'current_page': url}).encode('utf-8'))
             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
             streamdata = self._download_xml(
                 streamdata_req, video_id,
                 note='Downloading media info for %s' % video_format)
             stream_info = streamdata.find('./{default}preload/stream_info')
+            video_encode_id = xpath_text(stream_info, './video_encode_id')
+            if video_encode_id in video_encode_ids:
+                continue
+            video_encode_ids.append(video_encode_id)
             video_url = xpath_text(stream_info, './host')
             video_play_path = xpath_text(stream_info, './file')
             if not video_url or not video_play_path:
@@ -359,6 +367,14 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                 'ext': 'flv',
             })
             formats.append(format_info)
+        self._sort_formats(formats)
+
+        metadata = self._download_xml(
+            'http://www.crunchyroll.com/xml', video_id,
+            note='Downloading media info', query={
+                'req': 'RpcApiVideoPlayer_GetMediaMetadata',
+                'media_id': video_id,
+            })
 
         subtitles = self.extract_subtitles(video_id, webpage)
 
@@ -366,9 +382,12 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             'id': video_id,
             'title': video_title,
             'description': video_description,
-            'thumbnail': video_thumbnail,
+            'thumbnail': xpath_text(metadata, 'episode_image_url'),
             'uploader': video_uploader,
             'upload_date': video_upload_date,
+            'series': xpath_text(metadata, 'series_title'),
+            'episode': xpath_text(metadata, 'episode_title'),
+            'episode_number': int_or_none(xpath_text(metadata, 'episode_number')),
             'subtitles': subtitles,
             'formats': formats,
         }

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}\?play=(?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|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})'
     _TESTS = [{
         'url': 'http://cwtv.com/shows/arrow/legends-of-yesterday/?play=6b15e985-9345-4f60-baf8-56e96be57c63',
         'info_dict': {
@@ -48,6 +48,9 @@ class CWTVIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        'url': 'http://cwtv.com/thecw/chroniclesofcisco/?play=8adebe35-f447-465f-ab52-e863506ff6d6',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/dailymail.py

@@ -0,0 +1,61 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    determine_protocol,
+)
+
+
+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',
+        'md5': '2f639d446394f53f3a33658b518b6615',
+        'info_dict': {
+            'id': '1288527',
+            'ext': 'mp4',
+            'title': 'Turn any video into an impressionist masterpiece',
+            'description': 'md5:88ddbcb504367987b2708bb38677c9d2',
+        }
+    }
+
+    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"data-opts='({.+?})'", webpage, 'video data'), video_id)
+        title = 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)
+
+        formats = []
+        for rendition in video_sources['renditions']:
+            rendition_url = rendition.get('url')
+            if not rendition_url:
+                continue
+            tbr = int_or_none(rendition.get('encodingRate'), 1000)
+            container = rendition.get('videoContainer')
+            is_hls = container == 'M2TS'
+            protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url})
+            formats.append({
+                'format_id': ('hls' if is_hls else protocol) + ('-%d' % tbr if tbr else ''),
+                'url': rendition_url,
+                'width': int_or_none(rendition.get('frameWidth')),
+                'height': int_or_none(rendition.get('frameHeight')),
+                'tbr': tbr,
+                'vcodec': rendition.get('videoCodec'),
+                'container': container,
+                'protocol': protocol,
+                'ext': 'mp4' if is_hls else None,
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': video_data.get('descr'),
+            'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
+            'formats': formats,
+        }

youtube_dl/extractor/dfb.py

@@ -12,39 +12,46 @@ class DFBIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://tv.dfb.de/video/u-19-em-stimmen-zum-spiel-gegen-russland/11633/',
-        # The md5 is different each time
+        'md5': 'ac0f98a52a330f700b4b3034ad240649',
         'info_dict': {
             'id': '11633',
             'display_id': 'u-19-em-stimmen-zum-spiel-gegen-russland',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'U 19-EM: Stimmen zum Spiel gegen Russland',
             'upload_date': '20150714',
         },
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        display_id, video_id = re.match(self._VALID_URL, url).groups()
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id')
 
-        webpage = self._download_webpage(url, display_id)
         player_info = self._download_xml(
             'http://tv.dfb.de/server/hd_video.php?play=%s' % video_id,
             display_id)
         video_info = player_info.find('video')
+        stream_access_url = self._proto_relative_url(video_info.find('url').text.strip())
 
-        f4m_info = self._download_xml(
+        formats = []
-            self._proto_relative_url(video_info.find('url').text.strip()), display_id)
+        # see http://tv.dfb.de/player/js/ajax.js for the method to extract m3u8 formats
-        token_el = f4m_info.find('token')
+        for sa_url in (stream_access_url, stream_access_url + '&area=&format=iphone'):
-        manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0'
+            stream_access_info = self._download_xml(sa_url, display_id)
-        formats = self._extract_f4m_formats(manifest_url, display_id)
+            token_el = stream_access_info.find('token')
+            manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth']
+            if '.f4m' in manifest_url:
+                formats.extend(self._extract_f4m_formats(
+                    manifest_url + '&hdcore=3.2.0',
+                    display_id, f4m_id='hds', fatal=False))
+            else:
+                formats.extend(self._extract_m3u8_formats(
+                    manifest_url, display_id, 'mp4',
+                    'm3u8_native', m3u8_id='hls', fatal=False))
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': video_info.find('title').text,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'thumbnail': 'http://tv.dfb.de/images/%s_640x360.jpg' % video_id,
             'upload_date': unified_strdate(video_info.find('time_date').text),
             'formats': formats,
         }

youtube_dl/extractor/discovery.py

@@ -33,6 +33,7 @@ class DiscoveryIE(InfoExtractor):
             'duration': 156,
             'timestamp': 1302032462,
             'upload_date': '20110405',
+            'uploader_id': '103207',
         },
         'params': {
             'skip_download': True,  # requires ffmpeg
@@ -54,7 +55,11 @@ class DiscoveryIE(InfoExtractor):
             'upload_date': '20140725',
             'timestamp': 1406246400,
             'duration': 116,
+            'uploader_id': '103207',
         },
+        'params': {
+            'skip_download': True,  # requires ffmpeg
+        }
     }]
 
     def _real_extract(self, url):
@@ -66,13 +71,19 @@ class DiscoveryIE(InfoExtractor):
         entries = []
 
         for idx, video_info in enumerate(info['playlist']):
-            formats = self._extract_m3u8_formats(
+            subtitles = {}
-                video_info['src'], display_id, 'mp4', 'm3u8_native', m3u8_id='hls',
+            caption_url = video_info.get('captionsUrl')
-                note='Download m3u8 information for video %d' % (idx + 1))
+            if caption_url:
-            self._sort_formats(formats)
+                subtitles = {
+                    'en': [{
+                        'url': caption_url,
+                    }]
+                }
+
             entries.append({
+                '_type': 'url_transparent',
+                'url': 'http://players.brightcove.net/103207/default_default/index.html?videoId=ref:%s' % video_info['referenceId'],
                 'id': compat_str(video_info['id']),
-                'formats': formats,
                 'title': video_info['title'],
                 'description': video_info.get('description'),
                 'duration': parse_duration(video_info.get('video_length')),
@@ -80,6 +91,7 @@ class DiscoveryIE(InfoExtractor):
                 'thumbnail': video_info.get('thumbnailURL'),
                 'alt_title': video_info.get('secondary_title'),
                 'timestamp': parse_iso8601(video_info.get('publishedDate')),
+                'subtitles': subtitles,
             })
 
         return self.playlist_result(entries, display_id, video_title)

youtube_dl/extractor/dispeak.py

@@ -0,0 +1,114 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    remove_end,
+    xpath_element,
+    xpath_text,
+)
+
+
+class DigitallySpeakingIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:evt\.dispeak|events\.digitallyspeaking)\.com/(?:[^/]+/)+xml/(?P<id>[^.]+)\.xml'
+
+    _TESTS = [{
+        # From http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface
+        'url': 'http://evt.dispeak.com/ubm/gdc/sf16/xml/840376_BQRC.xml',
+        'md5': 'a8efb6c31ed06ca8739294960b2dbabd',
+        'info_dict': {
+            'id': '840376_BQRC',
+            'ext': 'mp4',
+            'title': 'Tenacious Design and The Interface of \'Destiny\'',
+        },
+    }, {
+        # From http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC
+        'url': 'http://events.digitallyspeaking.com/gdc/sf11/xml/12396_1299111843500GMPX.xml',
+        'only_matching': True,
+    }]
+
+    def _parse_mp4(self, metadata):
+        video_formats = []
+        video_root = None
+
+        mp4_video = xpath_text(metadata, './mp4video', default=None)
+        if mp4_video is not None:
+            mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video)
+            video_root = mobj.group('root')
+        if video_root is None:
+            http_host = xpath_text(metadata, 'httpHost', default=None)
+            if http_host:
+                video_root = 'http://%s/' % http_host
+        if video_root is None:
+            # Hard-coded in http://evt.dispeak.com/ubm/gdc/sf16/custom/player2.js
+            # Works for GPUTechConf, too
+            video_root = 'http://s3-2u.digitallyspeaking.com/'
+
+        formats = metadata.findall('./MBRVideos/MBRVideo')
+        if not formats:
+            return None
+        for a_format in formats:
+            stream_name = xpath_text(a_format, 'streamName', fatal=True)
+            video_path = re.match(r'mp4\:(?P<path>.*)', stream_name).group('path')
+            url = video_root + video_path
+            vbr = xpath_text(a_format, 'bitrate')
+            video_formats.append({
+                'url': url,
+                'vbr': int_or_none(vbr),
+            })
+        return video_formats
+
+    def _parse_flv(self, metadata):
+        formats = []
+        akamai_url = xpath_text(metadata, './akamaiHost', fatal=True)
+        audios = metadata.findall('./audios/audio')
+        for audio in audios:
+            formats.append({
+                'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
+                'play_path': remove_end(audio.get('url'), '.flv'),
+                'ext': 'flv',
+                'vcodec': 'none',
+                'format_id': audio.get('code'),
+            })
+        slide_video_path = xpath_text(metadata, './slideVideo', fatal=True)
+        formats.append({
+            'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
+            'play_path': remove_end(slide_video_path, '.flv'),
+            'ext': 'flv',
+            'format_note': 'slide deck video',
+            'quality': -2,
+            'preference': -2,
+            'format_id': 'slides',
+        })
+        speaker_video_path = xpath_text(metadata, './speakerVideo', fatal=True)
+        formats.append({
+            'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
+            'play_path': remove_end(speaker_video_path, '.flv'),
+            'ext': 'flv',
+            'format_note': 'speaker video',
+            'quality': -1,
+            'preference': -1,
+            'format_id': 'speaker',
+        })
+        return formats
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        xml_description = self._download_xml(url, video_id)
+        metadata = xpath_element(xml_description, 'metadata')
+
+        video_formats = self._parse_mp4(metadata)
+        if video_formats is None:
+            video_formats = self._parse_flv(metadata)
+
+        return {
+            'id': video_id,
+            'formats': video_formats,
+            'title': xpath_text(metadata, 'title', fatal=True),
+            'duration': parse_duration(xpath_text(metadata, 'endTime')),
+            'creator': xpath_text(metadata, 'speaker'),
+        }

youtube_dl/extractor/douyutv.py

@@ -18,7 +18,7 @@ class DouyuTVIE(InfoExtractor):
             'display_id': 'iseven',
             'ext': 'flv',
             'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
-            'description': 'md5:f34981259a03e980a3c6404190a3ed61',
+            'description': 're:.*m7show@163\.com.*',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': '7师傅',
             'uploader_id': '431925',
@@ -43,7 +43,7 @@ class DouyuTVIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
-        'skip': 'Romm not found',
+        'skip': 'Room not found',
     }, {
         'url': 'http://www.douyutv.com/17732',
         'info_dict': {
@@ -51,7 +51,7 @@ class DouyuTVIE(InfoExtractor):
             'display_id': '17732',
             'ext': 'flv',
             'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
-            'description': 'md5:f34981259a03e980a3c6404190a3ed61',
+            'description': 're:.*m7show@163\.com.*',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': '7师傅',
             'uploader_id': '431925',
@@ -75,13 +75,28 @@ class DouyuTVIE(InfoExtractor):
             room_id = self._html_search_regex(
                 r'"room_id"\s*:\s*(\d+),', page, 'room id')
 
+        config = None
+        # Douyu API sometimes returns error "Unable to load the requested class: eticket_redis_cache"
+        # Retry with different parameters - same parameters cause same errors
+        for i in range(5):
             prefix = 'room/%s?aid=android&client_sys=android&time=%d' % (
                 room_id, int(time.time()))
-
             auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest()
-        config = self._download_json(
+
+            config_page = self._download_webpage(
                 'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth),
                 video_id)
+            try:
+                config = self._parse_json(config_page, video_id, fatal=False)
+            except ExtractorError:
+                # Wait some time before retrying to get a different time() value
+                self._sleep(1, video_id, msg_template='%(video_id)s: Error occurs. '
+                                                      'Waiting for %(timeout)s seconds before retrying')
+                continue
+            else:
+                break
+        if config is None:
+            raise ExtractorError('Unable to fetch API result')
 
         data = config['data']
 

youtube_dl/extractor/dplay.py

@@ -6,13 +6,18 @@ import re
 import time
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..compat import compat_urlparse
+from ..utils import (
+    int_or_none,
+    update_url_query,
+)
 
 
 class DPlayIE(InfoExtractor):
     _VALID_URL = r'https?://(?P<domain>it\.dplay\.com|www\.dplay\.(?:dk|se|no))/[^/]+/(?P<id>[^/?#]+)'
 
     _TESTS = [{
+        # geo restricted, via direct unsigned hls URL
         'url': 'http://it.dplay.com/take-me-out/stagione-1-episodio-25/',
         'info_dict': {
             'id': '1255600',
@@ -31,11 +36,12 @@ class DPlayIE(InfoExtractor):
         },
         'expected_warnings': ['Unable to download f4m manifest'],
     }, {
+        # non geo restricted, via secure api, unsigned download hls URL
         'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/',
         'info_dict': {
             'id': '3172',
             'display_id': 'season-1-svensken-lar-sig-njuta-av-livet',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Svensken lär sig njuta av livet',
             'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
             'duration': 2650,
@@ -48,23 +54,25 @@ class DPlayIE(InfoExtractor):
             'age_limit': 0,
         },
     }, {
+        # geo restricted, via secure api, unsigned download hls URL
         'url': 'http://www.dplay.dk/mig-og-min-mor/season-6-episode-12/',
         'info_dict': {
             'id': '70816',
             'display_id': 'season-6-episode-12',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Episode 12',
             'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90',
             'duration': 2563,
             'timestamp': 1429696800,
             'upload_date': '20150422',
-            'creator': 'Kanal 4',
+            'creator': 'Kanal 4 (Home)',
             'series': 'Mig og min mor',
             'season_number': 6,
             'episode_number': 12,
             'age_limit': 0,
         },
     }, {
+        # geo restricted, via direct unsigned hls URL
         'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
         'only_matching': True,
     }]
@@ -90,17 +98,24 @@ class DPlayIE(InfoExtractor):
 
         def extract_formats(protocol, manifest_url):
             if protocol == 'hls':
-                formats.extend(self._extract_m3u8_formats(
+                m3u8_formats = self._extract_m3u8_formats(
                     manifest_url, video_id, ext='mp4',
-                    entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False))
+                    entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False)
+                # Sometimes final URLs inside m3u8 are unsigned, let's fix this
+                # ourselves
+                query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
+                for m3u8_format in m3u8_formats:
+                    m3u8_format['url'] = update_url_query(m3u8_format['url'], query)
+                formats.extend(m3u8_formats)
             elif protocol == 'hds':
                 formats.extend(self._extract_f4m_formats(
                     manifest_url + '&hdcore=3.8.0&plugin=flowplayer-3.8.0.0',
                     video_id, f4m_id=protocol, fatal=False))
 
         domain_tld = domain.split('.')[-1]
-        if domain_tld in ('se', 'dk'):
+        if domain_tld in ('se', 'dk', 'no'):
             for protocol in PROTOCOLS:
+                # Providing dsc-geo allows to bypass geo restriction in some cases
                 self._set_cookie(
                     'secure.dplay.%s' % domain_tld, 'dsc-geo',
                     json.dumps({
@@ -113,13 +128,24 @@ class DPlayIE(InfoExtractor):
                     'Downloading %s stream JSON' % protocol, fatal=False)
                 if stream and stream.get(protocol):
                     extract_formats(protocol, stream[protocol])
-        else:
+
+        # The last resort is to try direct unsigned hls/hds URLs from info dictionary.
+        # Sometimes this does work even when secure API with dsc-geo has failed (e.g.
+        # http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/).
+        if not formats:
             for protocol in PROTOCOLS:
                 if info.get(protocol):
                     extract_formats(protocol, info[protocol])
 
         self._sort_formats(formats)
 
+        subtitles = {}
+        for lang in ('se', 'sv', 'da', 'nl', 'no'):
+            for format_id in ('web_vtt', 'vtt', 'srt'):
+                subtitle_url = info.get('subtitles_%s_%s' % (lang, format_id))
+                if subtitle_url:
+                    subtitles.setdefault(lang, []).append({'url': subtitle_url})
+
         return {
             'id': video_id,
             'display_id': display_id,
@@ -133,4 +159,5 @@ class DPlayIE(InfoExtractor):
             'episode_number': int_or_none(info.get('episode')),
             'age_limit': int_or_none(info.get('minimum_age')),
             'formats': formats,
+            'subtitles': subtitles,
         }

youtube_dl/extractor/dump.py

@@ -1,39 +0,0 @@
-# encoding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-
-
-class DumpIE(InfoExtractor):
-    _VALID_URL = r'^https?://(?:www\.)?dump\.com/(?P<id>[a-zA-Z0-9]+)/'
-
-    _TEST = {
-        'url': 'http://www.dump.com/oneus/',
-        'md5': 'ad71704d1e67dfd9e81e3e8b42d69d99',
-        'info_dict': {
-            'id': 'oneus',
-            'ext': 'flv',
-            'title': "He's one of us.",
-            'thumbnail': 're:^https?://.*\.jpg$',
-        },
-    }
-
-    def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('id')
-
-        webpage = self._download_webpage(url, video_id)
-        video_url = self._search_regex(
-            r's1.addVariable\("file",\s*"([^"]+)"', webpage, 'video URL')
-
-        title = self._og_search_title(webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'url': video_url,
-            'thumbnail': thumbnail,
-        }

youtube_dl/extractor/eagleplatform.py

@@ -4,9 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_HTTPError
 from ..utils import (
     ExtractorError,
     int_or_none,
+    url_basename,
 )
 
 
@@ -21,7 +23,7 @@ class EaglePlatformIE(InfoExtractor):
     _TESTS = [{
         # http://lenta.ru/news/2015/03/06/navalny/
         'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
-        'md5': '70f5187fb620f2c1d503b3b22fd4efe3',
+        # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
         'info_dict': {
             'id': '227304',
             'ext': 'mp4',
@@ -36,7 +38,7 @@ class EaglePlatformIE(InfoExtractor):
         # http://muz-tv.ru/play/7129/
         # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
         'url': 'eagleplatform:media.clipyou.ru:12820',
-        'md5': '90b26344ba442c8e44aa4cf8f301164a',
+        'md5': '358597369cf8ba56675c1df15e7af624',
         'info_dict': {
             'id': '12820',
             'ext': 'mp4',
@@ -55,8 +57,13 @@ class EaglePlatformIE(InfoExtractor):
             raise ExtractorError(' '.join(response['errors']), expected=True)
 
     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
+        try:
             response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
+        except ExtractorError as ee:
+            if isinstance(ee.cause, compat_HTTPError):
+                response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
                 self._handle_error(response)
+            raise
         return response
 
     def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
@@ -84,17 +91,33 @@ class EaglePlatformIE(InfoExtractor):
 
         secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
 
+        formats = []
+
         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
-        formats = self._extract_m3u8_formats(
+        m3u8_formats = self._extract_m3u8_formats(
             m3u8_url, video_id,
             'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
+        formats.extend(m3u8_formats)
 
         mp4_url = self._get_video_url(
             # Secure mp4 URL is constructed according to Player.prototype.mp4 from
             # http://lentaru.media.eagleplatform.com/player/player.js
             re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4', secure_m3u8),
             video_id, 'Downloading mp4 JSON')
-        formats.append({'url': mp4_url, 'format_id': 'mp4'})
+        mp4_url_basename = url_basename(mp4_url)
+        for m3u8_format in m3u8_formats:
+            mobj = re.search('/([^/]+)/index\.m3u8', m3u8_format['url'])
+            if mobj:
+                http_format = m3u8_format.copy()
+                video_url = mp4_url.replace(mp4_url_basename, mobj.group(1))
+                if not self._is_valid_url(video_url, video_id):
+                    continue
+                http_format.update({
+                    'url': video_url,
+                    'format_id': m3u8_format['format_id'].replace('hls', 'http'),
+                    'protocol': 'http',
+                })
+                formats.append(http_format)
 
         self._sort_formats(formats)
 

youtube_dl/extractor/ebaumsworld.py

@@ -4,10 +4,10 @@ from .common import InfoExtractor
 
 
 class EbaumsWorldIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.ebaumsworld\.com/video/watch/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?ebaumsworld\.com/videos/[^/]+/(?P<id>\d+)'
 
     _TEST = {
-        'url': 'http://www.ebaumsworld.com/video/watch/83367677/',
+        'url': 'http://www.ebaumsworld.com/videos/a-giant-python-opens-the-door/83367677/',
         'info_dict': {
             'id': '83367677',
             'ext': 'mp4',

youtube_dl/extractor/extractors.py

@@ -46,6 +46,7 @@ from .arte import (
     ArteTVPlus7IE,
     ArteTVCreativeIE,
     ArteTVConcertIE,
+    ArteTVInfoIE,
     ArteTVFutureIE,
     ArteTVCinemaIE,
     ArteTVDDCIE,
@@ -74,6 +75,7 @@ from .bigflix import BigflixIE
 from .bild import BildIE
 from .bilibili import BiliBiliIE
 from .biobiochiletv import BioBioChileTVIE
+from .biqle import BIQLEIE
 from .bleacherreport import (
     BleacherReportIE,
     BleacherReportCMSIE,
@@ -122,7 +124,7 @@ from .chirbit import (
     ChirbitProfileIE,
 )
 from .cinchcast import CinchcastIE
-from .cinemassacre import CinemassacreIE
+from .cliprs import ClipRsIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
 from .clipsyndicate import ClipsyndicateIE
@@ -155,6 +157,7 @@ from .cspan import CSpanIE
 from .ctsnews import CtsNewsIE
 from .cultureunplugged import CultureUnpluggedIE
 from .cwtv import CWTVIE
+from .dailymail import DailyMailIE
 from .dailymotion import (
     DailymotionIE,
     DailymotionPlaylistIE,
@@ -191,10 +194,10 @@ from .drbonanza import DRBonanzaIE
 from .drtuber import DrTuberIE
 from .drtv import DRTVIE
 from .dvtv import DVTVIE
-from .dump import DumpIE
 from .dumpert import DumpertIE
 from .defense import DefenseGouvFrIE
 from .discovery import DiscoveryIE
+from .dispeak import DigitallySpeakingIE
 from .dropbox import DropboxIE
 from .dw import (
     DWIE,
@@ -335,7 +338,6 @@ from .ivi import (
 )
 from .ivideon import IvideonIE
 from .izlesene import IzleseneIE
-from .jadorecettepub import JadoreCettePubIE
 from .jeuxvideo import JeuxVideoIE
 from .jove import JoveIE
 from .jwplatform import JWPlatformIE
@@ -381,6 +383,7 @@ from .limelight import (
     LimelightChannelIE,
     LimelightChannelListIE,
 )
+from .litv import LiTVIE
 from .liveleak import LiveLeakIE
 from .livestream import (
     LivestreamIE,
@@ -399,19 +402,28 @@ from .macgamestore import MacGameStoreIE
 from .mailru import MailRuIE
 from .makerschannel import MakersChannelIE
 from .makertv import MakerTVIE
-from .malemotion import MalemotionIE
 from .matchtv import MatchTVIE
 from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mgoon import MgoonIE
+from .mgtv import MGTVIE
+from .microsoftvirtualacademy import (
+    MicrosoftVirtualAcademyIE,
+    MicrosoftVirtualAcademyCourseIE,
+)
 from .minhateca import MinhatecaIE
 from .ministrygrid import MinistryGridIE
 from .minoto import MinotoIE
 from .miomio import MioMioIE
 from .mit import TechTVMITIE, MITIE, OCWMITIE
 from .mitele import MiTeleIE
-from .mixcloud import MixcloudIE
+from .mixcloud import (
+    MixcloudIE,
+    MixcloudUserIE,
+    MixcloudPlaylistIE,
+    MixcloudStreamIE,
+)
 from .mlb import MLBIE
 from .mnet import MnetIE
 from .mpora import MporaIE
@@ -419,7 +431,6 @@ from .moevideo import MoeVideoIE
 from .mofosex import MofosexIE
 from .mojvideo import MojvideoIE
 from .moniker import MonikerIE
-from .mooshare import MooshareIE
 from .morningstar import MorningstarIE
 from .motherless import MotherlessIE
 from .motorsport import MotorsportIE
@@ -433,8 +444,7 @@ from .mtv import (
 )
 from .muenchentv import MuenchenTVIE
 from .musicplayon import MusicPlayOnIE
-from .muzu import MuzuTVIE
+from .mwave import MwaveIE, MwaveMeetGreetIE
-from .mwave import MwaveIE
 from .myspace import MySpaceIE, MySpaceAlbumIE
 from .myspass import MySpassIE
 from .myvi import MyviIE
@@ -464,7 +474,6 @@ from .ndr import (
 from .ndtv import NDTVIE
 from .netzkino import NetzkinoIE
 from .nerdcubed import NerdCubedFeedIE
-from .nerdist import NerdistIE
 from .neteasemusic import (
     NetEaseMusicIE,
     NetEaseMusicAlbumIE,
@@ -485,9 +494,10 @@ from .nextmovie import NextMovieIE
 from .nfb import NFBIE
 from .nfl import NFLIE
 from .nhl import (
-    NHLIE,
-    NHLNewsIE,
     NHLVideocenterIE,
+    NHLNewsIE,
+    NHLVideocenterCategoryIE,
+    NHLIE,
 )
 from .nick import NickIE
 from .niconico import NiconicoIE, NiconicoPlaylistIE
@@ -555,12 +565,15 @@ from .pandoratv import PandoraTVIE
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE
-from .periscope import PeriscopeIE
+from .people import PeopleIE
+from .periscope import (
+    PeriscopeIE,
+    PeriscopeUserIE,
+)
 from .philharmoniedeparis import PhilharmonieDeParisIE
 from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
 from .pinkbike import PinkbikeIE
-from .planetaplay import PlanetaPlayIE
 from .pladform import PladformIE
 from .played import PlayedIE
 from .playfm import PlayFMIE
@@ -583,6 +596,7 @@ from .pornhub import (
 from .pornotube import PornotubeIE
 from .pornovoisines import PornoVoisinesIE
 from .pornoxo import PornoXOIE
+from .presstv import PressTVIE
 from .primesharetv import PrimeShareTVIE
 from .promptfile import PromptFileIE
 from .prosiebensat1 import ProSiebenSat1IE
@@ -595,7 +609,6 @@ from .qqmusic import (
     QQMusicToplistIE,
     QQMusicPlaylistIE,
 )
-from .quickvid import QuickVidIE
 from .r7 import R7IE
 from .radiode import RadioDeIE
 from .radiojavan import RadioJavanIE
@@ -653,7 +666,6 @@ from .screenwavemedia import ScreenwaveMediaIE, TeamFourIE
 from .senateisvp import SenateISVPIE
 from .servingsys import ServingSysIE
 from .sexu import SexuIE
-from .sexykarma import SexyKarmaIE
 from .shahid import ShahidIE
 from .shared import SharedIE
 from .sharesix import ShareSixIE
@@ -670,10 +682,6 @@ from .smotri import (
     SmotriUserIE,
     SmotriBroadcastIE,
 )
-from .snagfilms import (
-    SnagFilmsIE,
-    SnagFilmsEmbedIE,
-)
 from .snotr import SnotrIE
 from .sohu import SohuIE
 from .soundcloud import (
@@ -725,9 +733,13 @@ from .svt import (
 from .swrmediathek import SWRMediathekIE
 from .syfy import SyfyIE
 from .sztvhu import SztvHuIE
-from .tagesschau import TagesschauIE
+from .tagesschau import (
+    TagesschauPlayerIE,
+    TagesschauIE,
+)
 from .tapely import TapelyIE
 from .tass import TassIE
+from .tdslifeway import TDSLifewayIE
 from .teachertube import (
     TeacherTubeIE,
     TeacherTubeUserIE,
@@ -745,7 +757,6 @@ from .teletask import TeleTaskIE
 from .testurl import TestURLIE
 from .tf1 import TF1IE
 from .theintercept import TheInterceptIE
-from .theonion import TheOnionIE
 from .theplatform import (
     ThePlatformIE,
     ThePlatformFeedIE,
@@ -823,7 +834,6 @@ from .twitch import (
     TwitchVodIE,
     TwitchProfileIE,
     TwitchPastBroadcastsIE,
-    TwitchBookmarksIE,
     TwitchStreamIE,
 )
 from .twitter import (
@@ -831,7 +841,6 @@ from .twitter import (
     TwitterIE,
     TwitterAmplifyIE,
 )
-from .ubu import UbuIE
 from .udemy import (
     UdemyIE,
     UdemyCourseIE
@@ -842,14 +851,20 @@ from .unistra import UnistraIE
 from .urort import UrortIE
 from .usatoday import USATodayIE
 from .ustream import UstreamIE, UstreamChannelIE
-from .ustudio import UstudioIE
+from .ustudio import (
+    UstudioIE,
+    UstudioEmbedIE,
+)
 from .varzesh3 import Varzesh3IE
 from .vbox7 import Vbox7IE
 from .veehd import VeeHDIE
 from .veoh import VeohIE
 from .vessel import VesselIE
 from .vesti import VestiIE
-from .vevo import VevoIE
+from .vevo import (
+    VevoIE,
+    VevoPlaylistIE,
+)
 from .vgtv import (
     BTArticleIE,
     BTVestlendingenIE,
@@ -878,6 +893,10 @@ from .vidme import (
 )
 from .vidzi import VidziIE
 from .vier import VierIE, VierVideosIE
+from .viewlift import (
+    ViewLiftIE,
+    ViewLiftEmbedIE,
+)
 from .viewster import ViewsterIE
 from .viidea import ViideaIE
 from .vimeo import (
@@ -916,7 +935,7 @@ from .vulture import VultureIE
 from .walla import WallaIE
 from .washingtonpost import WashingtonPostIE
 from .wat import WatIE
-from .wayofthemaster import WayOfTheMasterIE
+from .watchindianporn import WatchIndianPornIE
 from .wdr import (
     WDRIE,
     WDRMobileIE,
@@ -940,6 +959,12 @@ from .xhamster import (
     XHamsterIE,
     XHamsterEmbedIE,
 )
+from .xiami import (
+    XiamiSongIE,
+    XiamiAlbumIE,
+    XiamiArtistIE,
+    XiamiCollectionIE
+)
 from .xminus import XMinusIE
 from .xnxx import XNXXIE
 from .xstream import XstreamIE

youtube_dl/extractor/fczenit.py

@@ -1,20 +1,19 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 
 
 class FczenitIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?fc-zenit\.ru/video/gl(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?fc-zenit\.ru/video/(?P<id>[0-9]+)'
     _TEST = {
-        'url': 'http://fc-zenit.ru/video/gl6785/',
+        'url': 'http://fc-zenit.ru/video/41044/',
-        'md5': '458bacc24549173fe5a5aa29174a5606',
+        'md5': '0e3fab421b455e970fa1aa3891e57df0',
         'info_dict': {
-            'id': '6785',
+            'id': '41044',
             'ext': 'mp4',
-            'title': '«Зенит-ТВ»: как Олег Шатов играл против «Урала»',
+            'title': 'Так пишется история: казанский разгром ЦСКА на «Зенит-ТВ»',
         },
     }
 
@@ -22,15 +21,23 @@ class FczenitIE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        video_title = self._html_search_regex(r'<div class=\"photoalbum__title\">([^<]+)', webpage, 'title')
+        video_title = self._html_search_regex(
+            r'<[^>]+class=\"photoalbum__title\">([^<]+)', webpage, 'title')
 
-        bitrates_raw = self._html_search_regex(r'bitrates:.*\n(.*)\]', webpage, 'video URL')
+        video_items = self._parse_json(self._search_regex(
-        bitrates = re.findall(r'url:.?\'(.+?)\'.*?bitrate:.?([0-9]{3}?)', bitrates_raw)
+            r'arrPath\s*=\s*JSON\.parse\(\'(.+)\'\)', webpage, 'video items'),
+            video_id)
+
+        def merge_dicts(*dicts):
+            ret = {}
+            for a_dict in dicts:
+                ret.update(a_dict)
+            return ret
 
         formats = [{
-            'url': furl,
+            'url': compat_urlparse.urljoin(url, video_url),
-            'tbr': tbr,
+            'tbr': int(tbr),
-        } for furl, tbr in bitrates]
+        } for tbr, video_url in merge_dicts(*video_items).items()]
 
         self._sort_formats(formats)
 

youtube_dl/extractor/firsttv.py

@@ -2,78 +2,133 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..compat import compat_xpath
+from ..utils import (
+    int_or_none,
+    qualities,
+    unified_strdate,
+    xpath_attr,
+    xpath_element,
+    xpath_text,
+    xpath_with_ns,
+)
 
 
 class FirstTVIE(InfoExtractor):
     IE_NAME = '1tv'
     IE_DESC = 'Первый канал'
-    _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+(?P<id>.+)'
+    _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+p?(?P<id>\d+)'
 
     _TESTS = [{
-        'url': 'http://www.1tv.ru/videoarchive/73390',
+        # single format via video_materials.json API
-        'md5': '777f525feeec4806130f4f764bc18a4f',
-        'info_dict': {
-            'id': '73390',
-            'ext': 'mp4',
-            'title': 'Олимпийские канатные дороги',
-            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
-            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
-            'duration': 149,
-            'like_count': int,
-            'dislike_count': int,
-        },
-        'skip': 'Only works from Russia',
-    }, {
         'url': 'http://www.1tv.ru/prj/inprivate/vypusk/35930',
-        'md5': 'a1b6b60d530ebcf8daacf4565762bbaf',
+        'md5': '82a2777648acae812d58b3f5bd42882b',
         'info_dict': {
             'id': '35930',
             'ext': 'mp4',
-            'title': 'Наедине со всеми. Людмила Сенчина',
+            'title': 'Гость Людмила Сенчина. Наедине со всеми. Выпуск от 12.02.2015',
-            'description': 'md5:89553aed1d641416001fe8d450f06cb9',
+            'description': 'md5:357933adeede13b202c7c21f91b871b2',
             'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
+            'upload_date': '20150212',
             'duration': 2694,
         },
-        'skip': 'Only works from Russia',
+    }, {
+        # multiple formats via video_materials.json API
+        'url': 'http://www.1tv.ru/video_archive/projects/dobroeutro/p113641',
+        'info_dict': {
+            'id': '113641',
+            'ext': 'mp4',
+            'title': 'Весенняя аллергия. Доброе утро. Фрагмент выпуска от 07.04.2016',
+            'description': 'md5:8dcebb3dded0ff20fade39087fd1fee2',
+            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
+            'upload_date': '20160407',
+            'duration': 179,
+            'formats': 'mincount:3',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # single format only available via ONE_ONLINE_VIDEOS.archive_single_xml API
+        'url': 'http://www.1tv.ru/video_archive/series/f7552/p47038',
+        'md5': '519d306c5b5669761fd8906c39dbee23',
+        'info_dict': {
+            'id': '47038',
+            'ext': 'mp4',
+            'title': '"Побег". Второй сезон. 3 серия',
+            'description': 'md5:3abf8f6b9bce88201c33e9a3d794a00b',
+            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
+            'upload_date': '20120516',
+            'duration': 3080,
+        },
+    }, {
+        'url': 'http://www.1tv.ru/videoarchive/9967',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id, 'Downloading page')
+        # Videos with multiple formats only available via this API
+        video = self._download_json(
+            'http://www.1tv.ru/video_materials.json?legacy_id=%s' % video_id,
+            video_id, fatal=False)
 
-        video_url = self._html_search_regex(
+        description, thumbnail, upload_date, duration = [None] * 4
-            r'''(?s)(?:jwplayer\('flashvideoportal_1'\)\.setup\({|var\s+playlistObj\s*=).*?'file'\s*:\s*'([^']+)'.*?}\);''',
-            webpage, 'video URL')
 
+        if video:
+            item = video[0]
+            title = item['title']
+            quality = qualities(('ld', 'sd', 'hd', ))
+            formats = [{
+                'url': f['src'],
+                'format_id': f.get('name'),
+                'quality': quality(f.get('name')),
+            } for f in item['mbr'] if f.get('src')]
+            thumbnail = item.get('poster')
+        else:
+            # Some videos are not available via video_materials.json
+            video = self._download_xml(
+                'http://www.1tv.ru/owa/win/ONE_ONLINE_VIDEOS.archive_single_xml?pid=%s' % video_id,
+                video_id)
+
+            NS_MAP = {
+                'media': 'http://search.yahoo.com/mrss/',
+            }
+
+            item = xpath_element(video, './channel/item', fatal=True)
+            title = xpath_text(item, './title', fatal=True)
+            formats = [{
+                'url': content.attrib['url'],
+            } for content in item.findall(
+                compat_xpath(xpath_with_ns('./media:content', NS_MAP))) if content.attrib.get('url')]
+            thumbnail = xpath_attr(
+                item, xpath_with_ns('./media:thumbnail', NS_MAP), 'url')
+
+        self._sort_formats(formats)
+
+        webpage = self._download_webpage(url, video_id, 'Downloading page', fatal=False)
+        if webpage:
             title = self._html_search_regex(
-            [r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
+                (r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
-             r"'title'\s*:\s*'([^']+)'"], webpage, 'title')
+                 r"'title'\s*:\s*'([^']+)'"),
+                webpage, 'title', default=None) or title
             description = self._html_search_regex(
                 r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>',
                 webpage, 'description', default=None) or self._html_search_meta(
                 'description', webpage, 'description')
-
+            thumbnail = thumbnail or self._og_search_thumbnail(webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
+            duration = int_or_none(self._html_search_meta(
-        duration = self._og_search_property(
+                'video:duration', webpage, 'video duration', fatal=False))
-            'video:duration', webpage,
+            upload_date = unified_strdate(self._html_search_meta(
-            'video duration', fatal=False)
+                'ya:ovs:upload_date', webpage, 'upload date', fatal=False))
-
-        like_count = self._html_search_regex(
-            r'title="Понравилось".*?/></label> \[(\d+)\]',
-            webpage, 'like count', default=None)
-        dislike_count = self._html_search_regex(
-            r'title="Не понравилось".*?/></label> \[(\d+)\]',
-            webpage, 'dislike count', default=None)
 
         return {
             'id': video_id,
-            'url': video_url,
             'thumbnail': thumbnail,
             'title': title,
             'description': description,
+            'upload_date': upload_date,
             'duration': int_or_none(duration),
-            'like_count': int_or_none(like_count),
+            'formats': formats
-            'dislike_count': int_or_none(dislike_count),
         }

youtube_dl/extractor/flickr.py

@@ -24,13 +24,28 @@ class FlickrIE(InfoExtractor):
             'upload_date': '20110423',
             'uploader_id': '10922353@N03',
             'uploader': 'Forest Wander',
+            'uploader_url': 'https://www.flickr.com/photos/forestwander-nature-pictures/',
             'comment_count': int,
             'view_count': int,
             'tags': list,
+            'license': 'Attribution-ShareAlike',
         }
     }
-
     _API_BASE_URL = 'https://api.flickr.com/services/rest?'
+    # https://help.yahoo.com/kb/flickr/SLN25525.html
+    _LICENSES = {
+        '0': 'All Rights Reserved',
+        '1': 'Attribution-NonCommercial-ShareAlike',
+        '2': 'Attribution-NonCommercial',
+        '3': 'Attribution-NonCommercial-NoDerivs',
+        '4': 'Attribution',
+        '5': 'Attribution-ShareAlike',
+        '6': 'Attribution-NoDerivs',
+        '7': 'No known copyright restrictions',
+        '8': 'United States government work',
+        '9': 'Public Domain Dedication (CC0)',
+        '10': 'Public Domain Work',
+    }
 
     def _call_api(self, method, video_id, api_key, note, secret=None):
         query = {
@@ -75,6 +90,9 @@ class FlickrIE(InfoExtractor):
             self._sort_formats(formats)
 
             owner = video_info.get('owner', {})
+            uploader_id = owner.get('nsid')
+            uploader_path = owner.get('path_alias') or uploader_id
+            uploader_url = 'https://www.flickr.com/photos/%s/' % uploader_path if uploader_path else None
 
             return {
                 'id': video_id,
@@ -83,11 +101,13 @@ class FlickrIE(InfoExtractor):
                 'formats': formats,
                 'timestamp': int_or_none(video_info.get('dateuploaded')),
                 'duration': int_or_none(video_info.get('video', {}).get('duration')),
-                'uploader_id': owner.get('nsid'),
+                'uploader_id': uploader_id,
                 'uploader': owner.get('realname'),
+                'uploader_url': uploader_url,
                 'comment_count': int_or_none(video_info.get('comments', {}).get('_content')),
                 'view_count': int_or_none(video_info.get('views')),
-                'tags': [tag.get('_content') for tag in video_info.get('tags', {}).get('tag', [])]
+                'tags': [tag.get('_content') for tag in video_info.get('tags', {}).get('tag', [])],
+                'license': self._LICENSES.get(video_info.get('license')),
             }
         else:
             raise ExtractorError('not a video', expected=True)

youtube_dl/extractor/funimation.py

@@ -2,6 +2,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_HTTPError,
+    compat_urllib_parse_unquote_plus,
+)
 from ..utils import (
     clean_html,
     determine_ext,
@@ -27,6 +31,7 @@ class FunimationIE(InfoExtractor):
             'description': 'md5:1769f43cd5fc130ace8fd87232207892',
             'thumbnail': 're:https?://.*\.jpg',
         },
+        'skip': 'Access without user interaction is forbidden by CloudFlare, and video removed',
     }, {
         'url': 'http://www.funimation.com/shows/hacksign/videos/official/role-play',
         'info_dict': {
@@ -37,6 +42,7 @@ class FunimationIE(InfoExtractor):
             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
             'thumbnail': 're:https?://.*\.jpg',
         },
+        'skip': 'Access without user interaction is forbidden by CloudFlare',
     }, {
         'url': 'http://www.funimation.com/shows/attack-on-titan-junior-high/videos/promotional/broadcast-dub-preview',
         'info_dict': {
@@ -47,8 +53,36 @@ class FunimationIE(InfoExtractor):
             'description': 'md5:f8ec49c0aff702a7832cd81b8a44f803',
             'thumbnail': 're:https?://.*\.(?:jpg|png)',
         },
+        'skip': 'Access without user interaction is forbidden by CloudFlare',
     }]
 
+    _LOGIN_URL = 'http://www.funimation.com/login'
+
+    def _download_webpage(self, *args, **kwargs):
+        try:
+            return super(FunimationIE, self)._download_webpage(*args, **kwargs)
+        except ExtractorError as ee:
+            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
+                response = ee.cause.read()
+                if b'>Please complete the security check to access<' in response:
+                    raise ExtractorError(
+                        'Access to funimation.com is blocked by CloudFlare. '
+                        'Please browse to http://www.funimation.com/, solve '
+                        'the reCAPTCHA, export browser cookies to a text file,'
+                        ' and then try again with --cookies YOUR_COOKIE_FILE.',
+                        expected=True)
+            raise
+
+    def _extract_cloudflare_session_ua(self, url):
+        ci_session_cookie = self._get_cookies(url).get('ci_session')
+        if ci_session_cookie:
+            ci_session = compat_urllib_parse_unquote_plus(ci_session_cookie.value)
+            # ci_session is a string serialized by PHP function serialize()
+            # This case is simple enough to use regular expressions only
+            return self._search_regex(
+                r'"user_agent";s:\d+:"([^"]+)"', ci_session, 'user agent',
+                default=None)
+
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
@@ -57,8 +91,11 @@ class FunimationIE(InfoExtractor):
             'email_field': username,
             'password_field': password,
         })
-        login_request = sanitized_Request('http://www.funimation.com/login', data, headers={
+        user_agent = self._extract_cloudflare_session_ua(self._LOGIN_URL)
-            'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0',
+        if not user_agent:
+            user_agent = 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'
+        login_request = sanitized_Request(self._LOGIN_URL, data, headers={
+            'User-Agent': user_agent,
             'Content-Type': 'application/x-www-form-urlencoded'
         })
         login_page = self._download_webpage(
@@ -103,11 +140,16 @@ class FunimationIE(InfoExtractor):
             ('mobile', 'Mozilla/5.0 (Linux; Android 4.4.2; Nexus 4 Build/KOT49H) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.114 Mobile Safari/537.36'),
         )
 
+        user_agent = self._extract_cloudflare_session_ua(url)
+        if user_agent:
+            USER_AGENTS = ((None, user_agent),)
+
         for kind, user_agent in USER_AGENTS:
             request = sanitized_Request(url)
             request.add_header('User-Agent', user_agent)
             webpage = self._download_webpage(
-                request, display_id, 'Downloading %s webpage' % kind)
+                request, display_id,
+                'Downloading %s webpage' % kind if kind else 'Downloading webpage')
 
             playlist = self._parse_json(
                 self._search_regex(

youtube_dl/extractor/funnyordie.py

@@ -46,8 +46,8 @@ class FunnyOrDieIE(InfoExtractor):
         links.sort(key=lambda link: 1 if link[1] == 'mp4' else 0)
 
         m3u8_url = self._search_regex(
-            r'<source[^>]+src=(["\'])(?P<url>.+?/master\.m3u8)\1',
+            r'<source[^>]+src=(["\'])(?P<url>.+?/master\.m3u8[^"\']*)\1',
-            webpage, 'm3u8 url', default=None, group='url')
+            webpage, 'm3u8 url', group='url')
 
         formats = []
 

youtube_dl/extractor/gazeta.py

@@ -7,7 +7,7 @@ from .common import InfoExtractor
 
 
 class GazetaIE(InfoExtractor):
-    _VALID_URL = r'(?P<url>https?://(?:www\.)?gazeta\.ru/(?:[^/]+/)?video/(?:(?:main|\d{4}/\d{2}/\d{2})/)?(?P<id>[A-Za-z0-9-_.]+)\.s?html)'
+    _VALID_URL = r'(?P<url>https?://(?:www\.)?gazeta\.ru/(?:[^/]+/)?video/(?:main/)*(?:\d{4}/\d{2}/\d{2}/)?(?P<id>[A-Za-z0-9-_.]+)\.s?html)'
     _TESTS = [{
         'url': 'http://www.gazeta.ru/video/main/zadaite_vopros_vladislavu_yurevichu.shtml',
         'md5': 'd49c9bdc6e5a7888f27475dc215ee789',
@@ -18,9 +18,19 @@ class GazetaIE(InfoExtractor):
             'description': 'md5:38617526050bd17b234728e7f9620a71',
             'thumbnail': 're:^https?://.*\.jpg',
         },
+        'skip': 'video not found',
     }, {
         'url': 'http://www.gazeta.ru/lifestyle/video/2015/03/08/master-klass_krasivoi_byt._delaem_vesennii_makiyazh.shtml',
         'only_matching': True,
+    }, {
+        'url': 'http://www.gazeta.ru/video/main/main/2015/06/22/platit_ili_ne_platit_po_isku_yukosa.shtml',
+        'md5': '37f19f78355eb2f4256ee1688359f24c',
+        'info_dict': {
+            'id': '252048',
+            'ext': 'mp4',
+            'title': '"Если по иску ЮКОСа придется платить, это будет большой удар по бюджету"',
+        },
+        'add_ie': ['EaglePlatform'],
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/gdcvault.py

@@ -4,7 +4,6 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    remove_end,
     HEADRequest,
     sanitized_Request,
     urlencode_postdata,
@@ -51,63 +50,33 @@ class GDCVaultIE(InfoExtractor):
         {
             'url': 'http://gdcvault.com/play/1020791/',
             'only_matching': True,
+        },
+        {
+            # Hard-coded hostname
+            'url': 'http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface',
+            'md5': 'a8efb6c31ed06ca8739294960b2dbabd',
+            'info_dict': {
+                'id': '1023460',
+                'ext': 'mp4',
+                'display_id': 'Tenacious-Design-and-The-Interface',
+                'title': 'Tenacious Design and The Interface of \'Destiny\'',
+            },
+        },
+        {
+            # Multiple audios
+            'url': 'http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC',
+            'info_dict': {
+                'id': '1014631',
+                'ext': 'flv',
+                'title': 'How to Create a Good Game - From My Experience of Designing Pac-Man',
+            },
+            'params': {
+                'skip_download': True,  # Requires rtmpdump
+                'format': 'jp',  # The japanese audio
             }
+        },
     ]
 
-    def _parse_mp4(self, xml_description):
-        video_formats = []
-        mp4_video = xml_description.find('./metadata/mp4video')
-        if mp4_video is None:
-            return None
-
-        mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video.text)
-        video_root = mobj.group('root')
-        formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
-        for format in formats:
-            mobj = re.match(r'mp4\:(?P<path>.*)', format.find('streamName').text)
-            url = video_root + mobj.group('path')
-            vbr = format.find('bitrate').text
-            video_formats.append({
-                'url': url,
-                'vbr': int(vbr),
-            })
-        return video_formats
-
-    def _parse_flv(self, xml_description):
-        formats = []
-        akamai_url = xml_description.find('./metadata/akamaiHost').text
-        audios = xml_description.find('./metadata/audios')
-        if audios is not None:
-            for audio in audios:
-                formats.append({
-                    'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
-                    'play_path': remove_end(audio.get('url'), '.flv'),
-                    'ext': 'flv',
-                    'vcodec': 'none',
-                    'format_id': audio.get('code'),
-                })
-        slide_video_path = xml_description.find('./metadata/slideVideo').text
-        formats.append({
-            'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
-            'play_path': remove_end(slide_video_path, '.flv'),
-            'ext': 'flv',
-            'format_note': 'slide deck video',
-            'quality': -2,
-            'preference': -2,
-            'format_id': 'slides',
-        })
-        speaker_video_path = xml_description.find('./metadata/speakerVideo').text
-        formats.append({
-            'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
-            'play_path': remove_end(speaker_video_path, '.flv'),
-            'ext': 'flv',
-            'format_note': 'speaker video',
-            'quality': -1,
-            'preference': -1,
-            'format_id': 'speaker',
-        })
-        return formats
-
     def _login(self, webpage_url, display_id):
         (username, password) = self._get_login_info()
         if username is None or password is None:
@@ -183,17 +152,10 @@ class GDCVaultIE(InfoExtractor):
                 r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>',
                 start_page, 'xml filename')
 
-        xml_description = self._download_xml(
-            '%s/xml/%s' % (xml_root, xml_name), display_id)
-
-        video_title = xml_description.find('./metadata/title').text
-        video_formats = self._parse_mp4(xml_description)
-        if video_formats is None:
-            video_formats = self._parse_flv(xml_description)
-
         return {
+            '_type': 'url_transparent',
             'id': video_id,
             'display_id': display_id,
-            'title': video_title,
+            'url': '%s/xml/%s' % (xml_root, xml_name),
-            'formats': video_formats,
+            'ie_key': 'DigitallySpeaking',
         }

youtube_dl/extractor/generic.py

@@ -51,7 +51,7 @@ from .tnaflix import TNAFlixNetworkEmbedIE
 from .vimeo import VimeoIE
 from .dailymotion import DailymotionCloudIE
 from .onionstudios import OnionStudiosIE
-from .snagfilms import SnagFilmsEmbedIE
+from .viewlift import ViewLiftEmbedIE
 from .screenwavemedia import ScreenwaveMediaIE
 from .mtv import MTVServicesEmbeddedIE
 from .pladform import PladformIE
@@ -60,6 +60,7 @@ from .googledrive import GoogleDriveIE
 from .jwplatform import JWPlatformIE
 from .digiteka import DigitekaIE
 from .instagram import InstagramIE
+from .liveleak import LiveLeakIE
 
 
 class GenericIE(InfoExtractor):
@@ -104,7 +105,8 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,  # infinite live stream
             },
             'expected_warnings': [
-                r'501.*Not Implemented'
+                r'501.*Not Implemented',
+                r'400.*Bad Request',
             ],
         },
         # Direct link with incorrect MIME type
@@ -235,6 +237,7 @@ class GenericIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'car-20120827-manifest',
                 'formats': 'mincount:9',
+                'upload_date': '20130904',
             },
             'params': {
                 'format': 'bestvideo',
@@ -594,7 +597,11 @@ class GenericIE(InfoExtractor):
                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
                 'ext': 'mp4',
                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
+                'description': 'md5:faf028e48a461b8b7fad38f1e104b119',
                 'uploader': 'Spi0n',
+                'uploader_id': 'xgditw',
+                'upload_date': '20140425',
+                'timestamp': 1398441542,
             },
             'add_ie': ['Dailymotion'],
         },
@@ -727,8 +734,11 @@ class GenericIE(InfoExtractor):
                 'id': 'uxjb0lwrcz',
                 'ext': 'mp4',
                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
+                'description': 'a Martin Fowler video from ThoughtWorks',
                 'duration': 1715.0,
                 'uploader': 'thoughtworks.wistia.com',
+                'upload_date': '20140603',
+                'timestamp': 1401832161,
             },
         },
         # Soundcloud embed
@@ -877,6 +887,7 @@ class GenericIE(InfoExtractor):
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
+            # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
             'info_dict': {
                 'id': '227304',
                 'ext': 'mp4',
@@ -891,6 +902,7 @@ class GenericIE(InfoExtractor):
         # ClipYou (Eagle.Platform) embed (custom URL)
         {
             'url': 'http://muz-tv.ru/play/7129/',
+            # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
             'info_dict': {
                 'id': '12820',
                 'ext': 'mp4',
@@ -979,6 +991,9 @@ class GenericIE(InfoExtractor):
                 'ext': 'flv',
                 'title': "PFT Live: New leader in the 'new-look' defense",
                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
+                'uploader': 'NBCU-SPORTS',
+                'upload_date': '20140107',
+                'timestamp': 1389118457,
             },
         },
         # UDN embed
@@ -1031,6 +1046,9 @@ class GenericIE(InfoExtractor):
                 'title': 'SN Presents: Russell Martin, World Citizen',
                 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.',
                 'uploader': 'Rogers Sportsnet',
+                'uploader_id': '1704050871',
+                'upload_date': '20150525',
+                'timestamp': 1432570283,
             },
         },
         # Dailymotion Cloud video
@@ -1122,12 +1140,39 @@ class GenericIE(InfoExtractor):
                 'title': 'The Cardinal Pell Interview',
                 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ',
                 'uploader': 'GlobeCast Australia - GlobeStream',
+                'uploader_id': '2733773828001',
+                'upload_date': '20160304',
+                'timestamp': 1457083087,
             },
             'params': {
                 # m3u8 downloads
                 'skip_download': True,
             },
         },
+        # Another form of arte.tv embed
+        {
+            'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
+            'md5': '850bfe45417ddf221288c88a0cffe2e2',
+            'info_dict': {
+                'id': '030273-562_PLUS7-F',
+                'ext': 'mp4',
+                'title': 'ARTE Reportage - Nulle part, en France',
+                'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d',
+                'upload_date': '20160409',
+            },
+        },
+        # LiveLeak embed
+        {
+            'url': 'http://www.wykop.pl/link/3088787/',
+            'md5': 'ace83b9ed19b21f68e1b50e844fdf95d',
+            'info_dict': {
+                'id': '874_1459135191',
+                'ext': 'mp4',
+                'title': 'Man shows poor quality of new apartment building',
+                'description': 'The wall is like a sand pile.',
+                'uploader': 'Lake8737',
+            }
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -1702,7 +1747,7 @@ class GenericIE(InfoExtractor):
 
         # Look for embedded arte.tv player
         mobj = re.search(
-            r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
+            r'<(?:script|iframe) [^>]*?src="(?P<url>http://www\.arte\.tv/(?:playerv2/embed|arte_vp/index)[^"]+)"',
             webpage)
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
@@ -1879,10 +1924,10 @@ class GenericIE(InfoExtractor):
         if onionstudios_url:
             return self.url_result(onionstudios_url)
 
-        # Look for SnagFilms embeds
+        # Look for ViewLift embeds
-        snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
+        viewlift_url = ViewLiftEmbedIE._extract_url(webpage)
-        if snagfilms_url:
+        if viewlift_url:
-            return self.url_result(snagfilms_url)
+            return self.url_result(viewlift_url)
 
         # Look for JWPlatform embeds
         jwplatform_url = JWPlatformIE._extract_url(webpage)
@@ -1930,7 +1975,13 @@ class GenericIE(InfoExtractor):
         # Look for Instagram embeds
         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
         if instagram_embed_url is not None:
-            return self.url_result(instagram_embed_url, InstagramIE.ie_key())
+            return self.url_result(
+                self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key())
+
+        # Look for LiveLeak embeds
+        liveleak_url = LiveLeakIE._extract_url(webpage)
+        if liveleak_url:
+            return self.url_result(liveleak_url, 'LiveLeak')
 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
@@ -2013,6 +2064,7 @@ class GenericIE(InfoExtractor):
 
         entries = []
         for video_url in found:
+            video_url = unescapeHTML(video_url)
             video_url = video_url.replace('\\/', '/')
             video_url = compat_urlparse.urljoin(url, video_url)
             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))

youtube_dl/extractor/glide.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import unified_strdate
 
 
 class GlideIE(InfoExtractor):
@@ -15,26 +16,38 @@ class GlideIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Damon Timm\'s Glide message',
             'thumbnail': 're:^https?://.*?\.cloudfront\.net/.*\.jpg$',
+            'uploader': 'Damon Timm',
+            'upload_date': '20140919',
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
+
         title = self._html_search_regex(
-            r'<title>(.*?)</title>', webpage, 'title')
+            r'<title>(.+?)</title>', webpage, 'title')
-        video_url = self.http_scheme() + self._search_regex(
+        video_url = self._proto_relative_url(self._search_regex(
-            r'<source src="(.*?)" type="video/mp4">', webpage, 'video URL')
+            r'<source[^>]+src=(["\'])(?P<url>.+?)\1',
-        thumbnail_url = self._search_regex(
+            webpage, 'video URL', default=None,
-            r'<img id="video-thumbnail" src="(.*?)"',
+            group='url')) or self._og_search_video_url(webpage)
-            webpage, 'thumbnail url', fatal=False)
+        thumbnail = self._proto_relative_url(self._search_regex(
-        thumbnail = (
+            r'<img[^>]+id=["\']video-thumbnail["\'][^>]+src=(["\'])(?P<url>.+?)\1',
-            thumbnail_url if thumbnail_url is None
+            webpage, 'thumbnail url', default=None,
-            else self.http_scheme() + thumbnail_url)
+            group='url')) or self._og_search_thumbnail(webpage)
+        uploader = self._search_regex(
+            r'<div[^>]+class=["\']info-name["\'][^>]*>([^<]+)',
+            webpage, 'uploader', fatal=False)
+        upload_date = unified_strdate(self._search_regex(
+            r'<div[^>]+class="info-date"[^>]*>([^<]+)',
+            webpage, 'upload date', fatal=False))
 
         return {
             'id': video_id,
             'title': title,
             'url': video_url,
             'thumbnail': thumbnail,
+            'uploader': uploader,
+            'upload_date': upload_date,
         }

youtube_dl/extractor/goshgay.py

@@ -14,13 +14,13 @@ class GoshgayIE(InfoExtractor):
     _VALID_URL = r'https?://www\.goshgay\.com/video(?P<id>\d+?)($|/)'
     _TEST = {
         'url': 'http://www.goshgay.com/video299069/diesel_sfw_xxx_video',
-        'md5': '027fcc54459dff0feb0bc06a7aeda680',
+        'md5': '4b6db9a0a333142eb9f15913142b0ed1',
         'info_dict': {
             'id': '299069',
             'ext': 'flv',
             'title': 'DIESEL SFW XXX Video',
             'thumbnail': 're:^http://.*\.jpg$',
-            'duration': 79,
+            'duration': 80,
             'age_limit': 18,
         }
     }
@@ -47,5 +47,5 @@ class GoshgayIE(InfoExtractor):
             'title': title,
             'thumbnail': thumbnail,
             'duration': duration,
-            'age_limit': self._family_friendly_search(webpage),
+            'age_limit': 18,
         }

youtube_dl/extractor/gputechconf.py

@@ -2,12 +2,6 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import (
-    xpath_element,
-    xpath_text,
-    int_or_none,
-    parse_duration,
-)
 
 
 class GPUTechConfIE(InfoExtractor):
@@ -27,29 +21,15 @@ class GPUTechConfIE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        root_path = self._search_regex(r'var\s+rootPath\s*=\s*"([^"]+)', webpage, 'root path', 'http://evt.dispeak.com/nvidia/events/gtc15/')
+        root_path = self._search_regex(
-        xml_file_id = self._search_regex(r'var\s+xmlFileId\s*=\s*"([^"]+)', webpage, 'xml file id')
+            r'var\s+rootPath\s*=\s*"([^"]+)', webpage, 'root path',
-
+            default='http://evt.dispeak.com/nvidia/events/gtc15/')
-        doc = self._download_xml('%sxml/%s.xml' % (root_path, xml_file_id), video_id)
+        xml_file_id = self._search_regex(
-
+            r'var\s+xmlFileId\s*=\s*"([^"]+)', webpage, 'xml file id')
-        metadata = xpath_element(doc, 'metadata')
-        http_host = xpath_text(metadata, 'httpHost', 'http host', True)
-        mbr_videos = xpath_element(metadata, 'MBRVideos')
-
-        formats = []
-        for mbr_video in mbr_videos.findall('MBRVideo'):
-            stream_name = xpath_text(mbr_video, 'streamName')
-            if stream_name:
-                formats.append({
-                    'url': 'http://%s/%s' % (http_host, stream_name.replace('mp4:', '')),
-                    'tbr': int_or_none(xpath_text(mbr_video, 'bitrate')),
-                })
-        self._sort_formats(formats)
 
         return {
+            '_type': 'url_transparent',
             'id': video_id,
-            'title': xpath_text(metadata, 'title'),
+            'url': '%sxml/%s.xml' % (root_path, xml_file_id),
-            'duration': parse_duration(xpath_text(metadata, 'endTime')),
+            'ie_key': 'DigitallySpeaking',
-            'creator': xpath_text(metadata, 'speaker'),
-            'formats': formats,
         }

youtube_dl/extractor/groupon.py

@@ -16,14 +16,14 @@ class GrouponIE(InfoExtractor):
         'playlist': [{
             'info_dict': {
                 'id': 'tubGNycTo_9Uxg82uESj4i61EYX8nyuf',
-                'ext': 'mp4',
+                'ext': 'flv',
                 'title': 'Bikram Yoga Huntington Beach | Orange County',
                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
                 'duration': 44.961,
             },
         }],
         'params': {
-            'skip_download': 'HLS',
+            'skip_download': 'HDS',
         }
     }
 
@@ -32,7 +32,7 @@ class GrouponIE(InfoExtractor):
         webpage = self._download_webpage(url, playlist_id)
 
         payload = self._parse_json(self._search_regex(
-            r'var\s+payload\s*=\s*(.*?);\n', webpage, 'payload'), playlist_id)
+            r'(?:var\s+|window\.)payload\s*=\s*(.*?);\n', webpage, 'payload'), playlist_id)
         videos = payload['carousel'].get('dealVideos', [])
         entries = []
         for v in videos:

youtube_dl/extractor/howstuffworks.py

@@ -24,6 +24,7 @@ class HowStuffWorksIE(InfoExtractor):
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'duration': 161,
             },
+            'skip': 'Video broken',
         },
         {
             'url': 'http://adventure.howstuffworks.com/7199-survival-zone-food-and-water-in-the-savanna-video.htm',

youtube_dl/extractor/huffpost.py

@@ -4,6 +4,7 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    determine_ext,
     parse_duration,
     unified_strdate,
 )
@@ -29,7 +30,12 @@ class HuffPostIE(InfoExtractor):
             'description': 'This week on Legalese It, Mike talks to David Bosco about his new book on the ICC, "Rough Justice," he also discusses the Virginia AG\'s historic stance on gay marriage, the execution of Edgar Tamayo, the ICC\'s delay of Kenya\'s President and more.  ',
             'duration': 1549,
             'upload_date': '20140124',
-        }
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'expected_warnings': ['HTTP Error 404: Not Found'],
     }
 
     def _real_extract(self, url):
@@ -45,7 +51,7 @@ class HuffPostIE(InfoExtractor):
         description = data.get('description')
 
         thumbnails = []
-        for url in data['images'].values():
+        for url in filter(None, data['images'].values()):
             m = re.match('.*-([0-9]+x[0-9]+)\.', url)
             if not m:
                 continue
@@ -54,13 +60,25 @@ class HuffPostIE(InfoExtractor):
                 'resolution': m.group(1),
             })
 
-        formats = [{
+        formats = []
+        sources = data.get('sources', {})
+        live_sources = list(sources.get('live', {}).items()) + list(sources.get('live_again', {}).items())
+        for key, url in live_sources:
+            ext = determine_ext(url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    url, video_id, ext='mp4', m3u8_id='hls', fatal=False))
+            elif ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(
+                    url + '?hdcore=2.9.5', video_id, f4m_id='hds', fatal=False))
+            else:
+                formats.append({
                     'format': key,
                     'format_id': key.replace('/', '.'),
                     'ext': 'mp4',
                     'url': url,
                     'vcodec': 'none' if key.startswith('audio/') else None,
-        } for key, url in data.get('sources', {}).get('live', {}).items()]
+                })
 
         if not formats and data.get('fivemin_id'):
             return self.url_result('5min:%s' % data['fivemin_id'])

youtube_dl/extractor/imdb.py

@@ -1,10 +1,10 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
 from ..utils import (
+    mimetype2ext,
     qualities,
 )
 
@@ -12,9 +12,9 @@ from ..utils import (
 class ImdbIE(InfoExtractor):
     IE_NAME = 'imdb'
     IE_DESC = 'Internet Movie Database trailers'
-    _VALID_URL = r'https?://(?:www|m)\.imdb\.com/video/imdb/vi(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www|m)\.imdb\.com/video/[^/]+/vi(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.imdb.com/video/imdb/vi2524815897',
         'info_dict': {
             'id': '2524815897',
@@ -22,7 +22,10 @@ class ImdbIE(InfoExtractor):
             'title': 'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
             'description': 'md5:9061c2219254e5d14e03c25c98e96a81',
         }
-    }
+    }, {
+        'url': 'http://www.imdb.com/video/_/vi2524815897',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -48,13 +51,27 @@ class ImdbIE(InfoExtractor):
             json_data = self._search_regex(
                 r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>',
                 format_page, 'json data', flags=re.DOTALL)
-            info = json.loads(json_data)
+            info = self._parse_json(json_data, video_id, fatal=False)
-            format_info = info['videoPlayerObject']['video']
+            if not info:
-            f_id = format_info['ffname']
+                continue
+            format_info = info.get('videoPlayerObject', {}).get('video', {})
+            if not format_info:
+                continue
+            video_info_list = format_info.get('videoInfoList')
+            if not video_info_list or not isinstance(video_info_list, list):
+                continue
+            video_info = video_info_list[0]
+            if not video_info or not isinstance(video_info, dict):
+                continue
+            video_url = video_info.get('videoUrl')
+            if not video_url:
+                continue
+            format_id = format_info.get('ffname')
             formats.append({
-                'format_id': f_id,
+                'format_id': format_id,
-                'url': format_info['videoInfoList'][0]['videoUrl'],
+                'url': video_url,
-                'quality': quality(f_id),
+                'ext': mimetype2ext(video_info.get('videoMimeType')),
+                'quality': quality(format_id),
             })
         self._sort_formats(formats)
 

youtube_dl/extractor/instagram.py

@@ -12,7 +12,7 @@ from ..utils import (
 
 
 class InstagramIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?instagram\.com/p/(?P<id>[^/?#&]+)'
+    _VALID_URL = r'(?P<url>https?://(?:www\.)?instagram\.com/p/(?P<id>[^/?#&]+))'
     _TESTS = [{
         'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc',
         'md5': '0d2da106a9d2631273e192b372806516',
@@ -38,10 +38,19 @@ class InstagramIE(InfoExtractor):
     }, {
         'url': 'https://instagram.com/p/-Cmh1cukG2/',
         'only_matching': True,
+    }, {
+        'url': 'http://instagram.com/p/9o6LshA7zy/embed/',
+        'only_matching': True,
     }]
 
     @staticmethod
     def _extract_embed_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?instagram\.com/p/[^/]+/embed.*?)\1',
+            webpage)
+        if mobj:
+            return mobj.group('url')
+
         blockquote_el = get_element_by_attribute(
             'class', 'instagram-media', webpage)
         if blockquote_el is None:
@@ -53,7 +62,9 @@ class InstagramIE(InfoExtractor):
             return mobj.group('link')
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        url = mobj.group('url')
 
         webpage = self._download_webpage(url, video_id)
         uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',

youtube_dl/extractor/internetvideoarchive.py

@@ -1,93 +1,91 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..compat import (
+    compat_parse_qs,
     compat_urlparse,
-    compat_urllib_parse_urlencode,
 )
 from ..utils import (
-    xpath_with_ns,
+    determine_ext,
+    int_or_none,
+    xpath_text,
 )
 
 
 class InternetVideoArchiveIE(InfoExtractor):
-    _VALID_URL = r'https?://video\.internetvideoarchive\.net/flash/players/.*?\?.*?publishedid.*?'
+    _VALID_URL = r'https?://video\.internetvideoarchive\.net/(?:player|flash/players)/.*?\?.*?publishedid.*?'
 
     _TEST = {
-        'url': 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?customerid=69249&publishedid=452693&playerid=247',
+        'url': 'http://video.internetvideoarchive.net/player/6/configuration.ashx?customerid=69249&publishedid=194487&reporttag=vdbetatitle&playerid=641&autolist=0&domain=www.videodetective.com&maxrate=high&minrate=low&socialplayer=false',
         'info_dict': {
-            'id': '452693',
+            'id': '194487',
             'ext': 'mp4',
-            'title': 'SKYFALL',
+            'title': 'KICK-ASS 2',
-            'description': 'In SKYFALL, Bond\'s loyalty to M is tested as her past comes back to haunt her. As MI6 comes under attack, 007 must track down and destroy the threat, no matter how personal the cost.',
+            'description': 'md5:c189d5b7280400630a1d3dd17eaa8d8a',
-            'duration': 152,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
         },
     }
 
     @staticmethod
-    def _build_url(query):
+    def _build_json_url(query):
-        return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
+        return 'http://video.internetvideoarchive.net/player/6/configuration.ashx?' + query
 
     @staticmethod
-    def _clean_query(query):
+    def _build_xml_url(query):
-        NEEDED_ARGS = ['publishedid', 'customerid']
+        return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
-        query_dic = compat_urlparse.parse_qs(query)
-        cleaned_dic = dict((k, v[0]) for (k, v) in query_dic.items() if k in NEEDED_ARGS)
-        # Other player ids return m3u8 urls
-        cleaned_dic['playerid'] = '247'
-        cleaned_dic['videokbrate'] = '100000'
-        return compat_urllib_parse_urlencode(cleaned_dic)
 
     def _real_extract(self, url):
         query = compat_urlparse.urlparse(url).query
-        query_dic = compat_urlparse.parse_qs(query)
+        query_dic = compat_parse_qs(query)
         video_id = query_dic['publishedid'][0]
-        url = self._build_url(query)
 
-        flashconfiguration = self._download_xml(url, video_id,
+        if '/player/' in url:
-                                                'Downloading flash configuration')
+            configuration = self._download_json(url, video_id)
-        file_url = flashconfiguration.find('file').text
+
-        file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
+            # There are multiple videos in the playlist whlie only the first one
-        # Replace some of the parameters in the query to get the best quality
+            # matches the video played in browsers
-        # and http links (no m3u8 manifests)
+            video_info = configuration['playlist'][0]
-        file_url = re.sub(r'(?<=\?)(.+)$',
-                          lambda m: self._clean_query(m.group()),
-                          file_url)
-        info = self._download_xml(file_url, video_id,
-                                  'Downloading video info')
-        item = info.find('channel/item')
 
-        def _bp(p):
-            return xpath_with_ns(
-                p,
-                {
-                    'media': 'http://search.yahoo.com/mrss/',
-                    'jwplayer': 'http://developer.longtailvideo.com/trac/wiki/FlashFormats',
-                }
-            )
             formats = []
-        for content in item.findall(_bp('media:group/media:content')):
+            for source in video_info['sources']:
-            attr = content.attrib
+                file_url = source['file']
-            f_url = attr['url']
+                if determine_ext(file_url) == 'm3u8':
-            width = int(attr['width'])
+                    formats.extend(self._extract_m3u8_formats(
-            bitrate = int(attr['bitrate'])
+                        file_url, video_id, ext='mp4', m3u8_id='hls'))
-            format_id = '%d-%dk' % (width, bitrate)
+                else:
-            formats.append({
+                    a_format = {
-                'format_id': format_id,
+                        'url': file_url,
-                'url': f_url,
+                    }
-                'width': width,
+
-                'tbr': bitrate,
+                    if source.get('label') and source['label'][-4:] == ' kbs':
+                        tbr = int_or_none(source['label'][:-4])
+                        a_format.update({
+                            'tbr': tbr,
+                            'format_id': 'http-%d' % tbr,
                         })
+                        formats.append(a_format)
 
             self._sort_formats(formats)
 
+            title = video_info['title']
+            description = video_info.get('description')
+            thumbnail = video_info.get('image')
+        else:
+            configuration = self._download_xml(url, video_id)
+            formats = [{
+                'url': xpath_text(configuration, './file', 'file URL', fatal=True),
+            }]
+            thumbnail = xpath_text(configuration, './image', 'thumbnail')
+            title = 'InternetVideoArchive video %s' % video_id
+            description = None
+
         return {
             'id': video_id,
-            'title': item.find('title').text,
+            'title': title,
             'formats': formats,
-            'thumbnail': item.find(_bp('media:thumbnail')).attrib['url'],
+            'thumbnail': thumbnail,
-            'description': item.find('description').text,
+            'description': description,
-            'duration': int(attr['duration']),
         }

youtube_dl/extractor/iqiyi.py

@@ -165,7 +165,7 @@ class IqiyiIE(InfoExtractor):
     IE_NAME = 'iqiyi'
     IE_DESC = '爱奇艺'
 
-    _VALID_URL = r'https?://(?:[^.]+\.)?iqiyi\.com/.+\.html'
+    _VALID_URL = r'https?://(?:(?:[^.]+\.)?iqiyi\.com|www\.pps\.tv)/.+\.html'
 
     _NETRC_MACHINE = 'iqiyi'
 
@@ -273,6 +273,9 @@ class IqiyiIE(InfoExtractor):
             'title': '灌篮高手 国语版',
         },
         'playlist_count': 101,
+    }, {
+        'url': 'http://www.pps.tv/w_19rrbav0ph.html',
+        'only_matching': True,
     }]
 
     _FORMATS_MAP = [
@@ -284,6 +287,13 @@ class IqiyiIE(InfoExtractor):
         ('10', 'h1'),
     ]
 
+    AUTH_API_ERRORS = {
+        # No preview available (不允许试看鉴权失败)
+        'Q00505': 'This video requires a VIP account',
+        # End of preview time (试看结束鉴权失败)
+        'Q00506': 'Needs a VIP account for full video',
+    }
+
     def _real_initialize(self):
         self._login()
 
@@ -369,14 +379,18 @@ class IqiyiIE(InfoExtractor):
             note='Downloading video authentication JSON',
             errnote='Unable to download video authentication JSON')
 
-        if auth_result['code'] == 'Q00505':  # No preview available (不允许试看鉴权失败)
+        code = auth_result.get('code')
-            raise ExtractorError('This video requires a VIP account', expected=True)
+        msg = self.AUTH_API_ERRORS.get(code) or auth_result.get('msg') or code
-        if auth_result['code'] == 'Q00506':  # End of preview time (试看结束鉴权失败)
+        if code == 'Q00506':
             if do_report_warning:
-                self.report_warning('Needs a VIP account for full video')
+                self.report_warning(msg)
             return False
+        if 'data' not in auth_result:
+            if msg is not None:
+                raise ExtractorError('%s said: %s' % (self.IE_NAME, msg), expected=True)
+            raise ExtractorError('Unexpected error from Iqiyi auth API')
 
-        return auth_result
+        return auth_result['data']
 
     def construct_video_urls(self, data, video_id, _uuid, tvid):
         def do_xor(x, y):
@@ -452,11 +466,11 @@ class IqiyiIE(InfoExtractor):
                         need_vip_warning_report = False
                         break
                     param.update({
-                        't': auth_result['data']['t'],
+                        't': auth_result['t'],
                         # cid is hard-coded in com/qiyi/player/core/player/RuntimeData.as
                         'cid': 'afbe8fd3d73448c9',
                         'vid': video_id,
-                        'QY00001': auth_result['data']['u'],
+                        'QY00001': auth_result['u'],
                     })
                 api_video_url += '?' if '?' not in api_video_url else '&'
                 api_video_url += compat_urllib_parse_urlencode(param)
@@ -491,7 +505,10 @@ class IqiyiIE(InfoExtractor):
             'enc': md5_text(enc_key + tail),
             'qyid': _uuid,
             'tn': random.random(),
-            'um': 0,
+            # In iQiyi's flash player, um is set to 1 if there's a logged user
+            # Some 1080P formats are only available with a logged user.
+            # Here force um=1 to trick the iQiyi server
+            'um': 1,
             'authkey': md5_text(md5_text('') + tail),
             'k_tag': 1,
         }

youtube_dl/extractor/izlesene.py

@@ -29,7 +29,7 @@ class IzleseneIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Sevinçten Çıldırtan Doğum Günü Hediyesi',
                 'description': 'md5:253753e2655dde93f59f74b572454f6d',
-                'thumbnail': 're:^http://.*\.jpg',
+                'thumbnail': 're:^https?://.*\.jpg',
                 'uploader_id': 'pelikzzle',
                 'timestamp': int,
                 'upload_date': '20140702',
@@ -44,8 +44,7 @@ class IzleseneIE(InfoExtractor):
                 'id': '17997',
                 'ext': 'mp4',
                 'title': 'Tarkan Dortmund 2006 Konseri',
-                'description': 'Tarkan Dortmund 2006 Konseri',
+                'thumbnail': 're:^https://.*\.jpg',
-                'thumbnail': 're:^http://.*\.jpg',
                 'uploader_id': 'parlayankiz',
                 'timestamp': int,
                 'upload_date': '20061112',
@@ -62,7 +61,7 @@ class IzleseneIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         title = self._og_search_title(webpage)
-        description = self._og_search_description(webpage)
+        description = self._og_search_description(webpage, default=None)
         thumbnail = self._proto_relative_url(
             self._og_search_thumbnail(webpage), scheme='http:')
 

youtube_dl/extractor/jadorecettepub.py

@@ -1,47 +0,0 @@
-# coding: utf-8
-
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from .youtube import YoutubeIE
-
-
-class JadoreCettePubIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?jadorecettepub\.com/[0-9]{4}/[0-9]{2}/(?P<id>.*?)\.html'
-
-    _TEST = {
-        'url': 'http://www.jadorecettepub.com/2010/12/star-wars-massacre-par-les-japonais.html',
-        'md5': '401286a06067c70b44076044b66515de',
-        'info_dict': {
-            'id': 'jLMja3tr7a4',
-            'ext': 'mp4',
-            'title': 'La pire utilisation de Star Wars',
-            'description': "Jadorecettepub.com vous a gratifié de plusieurs pubs géniales utilisant Star Wars et Dark Vador plus particulièrement... Mais l'heure est venue de vous proposer une version totalement massacrée, venue du Japon.  Quand les Japonais détruisent l'image de Star Wars pour vendre du thon en boite, ça promet...",
-        },
-    }
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('id')
-
-        webpage = self._download_webpage(url, display_id)
-
-        title = self._html_search_regex(
-            r'<span style="font-size: x-large;"><b>(.*?)</b></span>',
-            webpage, 'title')
-        description = self._html_search_regex(
-            r'(?s)<div id="fb-root">(.*?)<script>', webpage, 'description',
-            fatal=False)
-        real_url = self._search_regex(
-            r'\[/postlink\](.*)endofvid', webpage, 'video URL')
-        video_id = YoutubeIE.extract_id(real_url)
-
-        return {
-            '_type': 'url_transparent',
-            'url': real_url,
-            'id': video_id,
-            'title': title,
-            'description': description,
-        }

youtube_dl/extractor/jwplatform.py

@@ -4,16 +4,15 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    float_or_none,
+    int_or_none,
+)
 
 
 class JWPlatformBaseIE(InfoExtractor):
     def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True):
         video_data = jwplayer_data['playlist'][0]
-        subtitles = {}
-        for track in video_data['tracks']:
-            if track['kind'] == 'captions':
-                subtitles[track['label']] = [{'url': self._proto_relative_url(track['file'])}]
 
         formats = []
         for source in video_data['sources']:
@@ -35,12 +34,22 @@ class JWPlatformBaseIE(InfoExtractor):
                 })
         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'])
+                    })
+
         return {
             '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,
         }

youtube_dl/extractor/karaoketv.py

@@ -2,39 +2,63 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote_plus
-from ..utils import (
-    js_to_json,
-)
 
 
 class KaraoketvIE(InfoExtractor):
-    _VALID_URL = r'https?://karaoketv\.co\.il/\?container=songs&id=(?P<id>[0-9]+)'
+    _VALID_URL = r'http://www.karaoketv.co.il/[^/]+/(?P<id>\d+)'
     _TEST = {
-        'url': 'http://karaoketv.co.il/?container=songs&id=171568',
+        'url': 'http://www.karaoketv.co.il/%D7%A9%D7%99%D7%A8%D7%99_%D7%A7%D7%A8%D7%99%D7%95%D7%A7%D7%99/58356/%D7%90%D7%99%D7%96%D7%95%D7%9F',
         'info_dict': {
-            'id': '171568',
+            'id': '58356',
-            'ext': 'mp4',
+            'ext': 'flv',
-            'title': 'אל העולם שלך - רותם כהן - שרים קריוקי',
+            'title': 'קריוקי של איזון',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
+        api_page_url = self._search_regex(
+            r'<iframe[^>]+src=(["\'])(?P<url>https?://www\.karaoke\.co\.il/api_play\.php\?.+?)\1',
+            webpage, 'API play URL', group='url')
 
-        page_video_url = self._og_search_video_url(webpage, video_id)
+        api_page = self._download_webpage(api_page_url, video_id)
-        config_json = compat_urllib_parse_unquote_plus(self._search_regex(
+        video_cdn_url = self._search_regex(
-            r'config=(.*)', page_video_url, 'configuration'))
+            r'<iframe[^>]+src=(["\'])(?P<url>https?://www\.video-cdn\.com/embed/iframe/.+?)\1',
+            api_page, 'video cdn URL', group='url')
 
-        urls_info_json = self._download_json(
+        video_cdn = self._download_webpage(video_cdn_url, video_id)
-            config_json, video_id, 'Downloading configuration',
+        play_path = self._parse_json(
-            transform_source=js_to_json)
+            self._search_regex(
+                r'var\s+options\s*=\s*({.+?});', video_cdn, 'options'),
+            video_id)['clip']['url']
 
-        url = urls_info_json['playlist'][0]['url']
+        settings = self._parse_json(
+            self._search_regex(
+                r'var\s+settings\s*=\s*({.+?});', video_cdn, 'servers', default='{}'),
+            video_id, fatal=False) or {}
+
+        servers = settings.get('servers')
+        if not servers or not isinstance(servers, list):
+            servers = ('wowzail.video-cdn.com:80/vodcdn', )
+
+        formats = [{
+            'url': 'rtmp://%s' % server if not server.startswith('rtmp') else server,
+            'play_path': play_path,
+            'app': 'vodcdn',
+            'page_url': video_cdn_url,
+            'player_url': 'http://www.video-cdn.com/assets/flowplayer/flowplayer.commercial-3.2.18.swf',
+            'rtmp_real_time': True,
+            'ext': 'flv',
+        } for server in servers]
 
         return {
             'id': video_id,
             'title': self._og_search_title(webpage),
-            'url': url,
+            'formats': formats,
         }

youtube_dl/extractor/karrierevideos.py

@@ -52,9 +52,12 @@ class KarriereVideosIE(InfoExtractor):
 
         video_id = self._search_regex(
             r'/config/video/(.+?)\.xml', webpage, 'video id')
+        # Server returns malformed headers
+        # Force Accept-Encoding: * to prevent gzipped results
         playlist = self._download_xml(
             'http://www.karrierevideos.at/player-playlist.xml.php?p=%s' % video_id,
-            video_id, transform_source=fix_xml_ampersands)
+            video_id, transform_source=fix_xml_ampersands,
+            headers={'Accept-Encoding': '*'})
 
         NS_MAP = {
             'jwplayer': 'http://developer.longtailvideo.com/trac/wiki/FlashFormats'

youtube_dl/extractor/kuwo.py

@@ -81,7 +81,7 @@ class KuwoIE(KuwoBaseIE):
             'id': '6446136',
             'ext': 'mp3',
             'title': '心',
-            'description': 'md5:b2ab6295d014005bfc607525bfc1e38a',
+            'description': 'md5:5d0e947b242c35dc0eb1d2fce9fbf02c',
             'creator': 'IU',
             'upload_date': '20150518',
         },
@@ -102,10 +102,10 @@ class KuwoIE(KuwoBaseIE):
             raise ExtractorError('this song has been offline because of copyright issues', expected=True)
 
         song_name = self._html_search_regex(
-            r'(?s)class="(?:[^"\s]+\s+)*title(?:\s+[^"\s]+)*".*?<h1[^>]+title="([^"]+)"', webpage, 'song name')
+            r'<p[^>]+id="lrcName">([^<]+)</p>', webpage, 'song name')
-        singer_name = self._html_search_regex(
+        singer_name = remove_start(self._html_search_regex(
-            r'<div[^>]+class="s_img">\s*<a[^>]+title="([^>]+)"',
+            r'<a[^>]+href="http://www\.kuwo\.cn/artist/content\?name=([^"]+)">',
-            webpage, 'singer name', fatal=False)
+            webpage, 'singer name', fatal=False), '歌手')
         lrc_content = clean_html(get_element_by_id('lrcContent', webpage))
         if lrc_content == '暂无':     # indicates no lyrics
             lrc_content = None
@@ -114,7 +114,7 @@ class KuwoIE(KuwoBaseIE):
         self._sort_formats(formats)
 
         album_id = self._html_search_regex(
-            r'<p[^>]+class="album"[^<]+<a[^>]+href="http://www\.kuwo\.cn/album/(\d+)/"',
+            r'<a[^>]+href="http://www\.kuwo\.cn/album/(\d+)/"',
             webpage, 'album id', fatal=False)
 
         publish_time = None
@@ -268,7 +268,7 @@ class KuwoCategoryIE(InfoExtractor):
             'title': '八十年代精选',
             'description': '这些都是属于八十年代的回忆！',
         },
-        'playlist_count': 30,
+        'playlist_mincount': 24,
     }
 
     def _real_extract(self, url):
@@ -283,6 +283,8 @@ class KuwoCategoryIE(InfoExtractor):
         category_desc = remove_start(
             get_element_by_id('intro', webpage).strip(),
             '%s简介：' % category_name)
+        if category_desc == '暂无':
+            category_desc = None
 
         jsonm = self._parse_json(self._html_search_regex(
             r'var\s+jsonm\s*=\s*([^;]+);', webpage, 'category songs'), category_id)

youtube_dl/extractor/laola1tv.py

@@ -63,6 +63,7 @@ class Laola1TvIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+        'skip': 'This live stream has already finished.',
     }]
 
     def _real_extract(self, url):
@@ -74,6 +75,9 @@ class Laola1TvIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
+        if 'Dieser Livestream ist bereits beendet.' in webpage:
+            raise ExtractorError('This live stream has already finished.', expected=True)
+
         iframe_url = self._search_regex(
             r'<iframe[^>]*?id="videoplayer"[^>]*?src="([^"]+)"',
             webpage, 'iframe url')

youtube_dl/extractor/lecture2go.py

@@ -6,6 +6,7 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
+    determine_protocol,
     parse_duration,
     int_or_none,
 )
@@ -18,10 +19,14 @@ class Lecture2GoIE(InfoExtractor):
         'md5': 'ac02b570883020d208d405d5a3fd2f7f',
         'info_dict': {
             'id': '17473',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': '2 - Endliche Automaten und reguläre Sprachen',
             'creator': 'Frank Heitmann',
             'duration': 5220,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
         }
     }
 
@@ -32,14 +37,18 @@ class Lecture2GoIE(InfoExtractor):
         title = self._html_search_regex(r'<em[^>]+class="title">(.+)</em>', webpage, 'title')
 
         formats = []
-        for url in set(re.findall(r'"src","([^"]+)"', webpage)):
+        for url in set(re.findall(r'var\s+playerUri\d+\s*=\s*"([^"]+)"', webpage)):
             ext = determine_ext(url)
+            protocol = determine_protocol({'url': url})
             if ext == 'f4m':
-                formats.extend(self._extract_f4m_formats(url, video_id))
+                formats.extend(self._extract_f4m_formats(url, video_id, f4m_id='hds'))
             elif ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(url, video_id))
+                formats.extend(self._extract_m3u8_formats(url, video_id, ext='mp4', m3u8_id='hls'))
             else:
+                if protocol == 'rtmp':
+                    continue  # XXX: currently broken
                 formats.append({
+                    'format_id': protocol,
                     'url': url,
                 })
 

youtube_dl/extractor/litv.py

@@ -0,0 +1,137 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    smuggle_url,
+    unsmuggle_url,
+)
+
+
+class LiTVIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.litv\.tv/vod/[^/]+/content\.do\?.*?\bid=(?P<id>[^&]+)'
+
+    _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?id=%s'
+
+    _TESTS = [{
+        'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'info_dict': {
+            'id': 'VOD00041606',
+            'title': '花千骨',
+        },
+        'playlist_count': 50,
+    }, {
+        'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'info_dict': {
+            'id': 'VOD00041610',
+            'ext': 'mp4',
+            'title': '花千骨第1集',
+            'thumbnail': 're:https?://.*\.jpg$',
+            'description': 'md5:c7017aa144c87467c4fb2909c4b05d6f',
+            'episode_number': 1,
+        },
+        'params': {
+            'noplaylist': True,
+            'skip_download': True,  # m3u8 download
+        },
+        'skip': 'Georestricted to Taiwan',
+    }]
+
+    def _extract_playlist(self, season_list, video_id, vod_data, view_data, prompt=True):
+        episode_title = view_data['title']
+        content_id = season_list['contentId']
+
+        if prompt:
+            self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (content_id, video_id))
+
+        all_episodes = [
+            self.url_result(smuggle_url(
+                self._URL_TEMPLATE % (view_data['contentType'], episode['contentId']),
+                {'force_noplaylist': True}))  # To prevent infinite recursion
+            for episode in season_list['episode']]
+
+        return self.playlist_result(all_episodes, content_id, episode_title)
+
+    def _real_extract(self, url):
+        url, data = unsmuggle_url(url, {})
+
+        video_id = self._match_id(url)
+
+        noplaylist = self._downloader.params.get('noplaylist')
+        noplaylist_prompt = True
+        if 'force_noplaylist' in data:
+            noplaylist = data['force_noplaylist']
+            noplaylist_prompt = False
+
+        webpage = self._download_webpage(url, video_id)
+
+        view_data = dict(map(lambda t: (t[0], t[2]), re.findall(
+            r'viewData\.([a-zA-Z]+)\s*=\s*(["\'])([^"\']+)\2',
+            webpage)))
+
+        vod_data = self._parse_json(self._search_regex(
+            'var\s+vod\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'),
+            video_id)
+
+        season_list = list(vod_data.get('seasonList', {}).values())
+        if season_list:
+            if not noplaylist:
+                return self._extract_playlist(
+                    season_list[0], video_id, vod_data, view_data,
+                    prompt=noplaylist_prompt)
+
+            if noplaylist_prompt:
+                self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
+
+        # In browsers `getMainUrl` request is always issued. Usually this
+        # endpoint gives the same result as the data embedded in the webpage.
+        # If georestricted, there are no embedded data, so an extra request is
+        # necessary to get the error code
+        video_data = self._parse_json(self._search_regex(
+            r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);',
+            webpage, 'video data', default='{}'), video_id)
+        if not video_data:
+            payload = {
+                'assetId': view_data['assetId'],
+                'watchDevices': vod_data['watchDevices'],
+                'contentType': view_data['contentType'],
+            }
+            video_data = self._download_json(
+                'https://www.litv.tv/vod/getMainUrl', video_id,
+                data=json.dumps(payload).encode('utf-8'),
+                headers={'Content-Type': 'application/json'})
+
+        if not video_data.get('fullpath'):
+            error_msg = video_data.get('errorMessage')
+            if error_msg == 'vod.error.outsideregionerror':
+                self.raise_geo_restricted('This video is available in Taiwan only')
+            if error_msg:
+                raise ExtractorError('%s said: %s' % (self.IE_NAME, error_msg), expected=True)
+            raise ExtractorError('Unexpected result from %s' % self.IE_NAME)
+
+        formats = self._extract_m3u8_formats(
+            video_data['fullpath'], video_id, ext='mp4', m3u8_id='hls')
+        for a_format in formats:
+            # LiTV HLS segments doesn't like compressions
+            a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = True
+
+        title = view_data['title'] + view_data.get('secondaryMark', '')
+        description = view_data.get('description')
+        thumbnail = view_data.get('imageFile')
+        categories = [item['name'] for item in vod_data.get('category', [])]
+        episode = int_or_none(view_data.get('episode'))
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'categories': categories,
+            'episode_number': episode,
+        }

youtube_dl/extractor/liveleak.py

@@ -17,7 +17,8 @@ class LiveLeakIE(InfoExtractor):
             'ext': 'flv',
             'description': 'extremely bad day for this guy..!',
             'uploader': 'ljfriel2',
-            'title': 'Most unlucky car accident'
+            'title': 'Most unlucky car accident',
+            'thumbnail': 're:^https?://.*\.jpg$'
         }
     }, {
         'url': 'http://www.liveleak.com/view?i=f93_1390833151',
@@ -28,6 +29,7 @@ class LiveLeakIE(InfoExtractor):
             'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.',
             'uploader': 'ARD_Stinkt',
             'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
+            'thumbnail': 're:^https?://.*\.jpg$'
         }
     }, {
         'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
@@ -49,10 +51,19 @@ class LiveLeakIE(InfoExtractor):
             'ext': 'mp4',
             'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
             'uploader': 'bony333',
-            'title': 'Crazy Hungarian tourist films close call waterspout in Croatia'
+            'title': 'Crazy Hungarian tourist films close call waterspout in Croatia',
+            'thumbnail': 're:^https?://.*\.jpg$'
         }
     }]
 
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+src="https?://(?:\w+\.)?liveleak\.com/ll_embed\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)',
+            webpage)
+        if mobj:
+            return 'http://www.liveleak.com/view?i=%s' % mobj.group('id')
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
@@ -64,6 +75,7 @@ class LiveLeakIE(InfoExtractor):
         age_limit = int_or_none(self._search_regex(
             r'you confirm that you are ([0-9]+) years and over.',
             webpage, 'age limit', default=None))
+        video_thumbnail = self._og_search_thumbnail(webpage)
 
         sources_raw = self._search_regex(
             r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None)
@@ -116,4 +128,5 @@ class LiveLeakIE(InfoExtractor):
             'uploader': video_uploader,
             'formats': formats,
             'age_limit': age_limit,
+            'thumbnail': video_thumbnail,
         }

youtube_dl/extractor/malemotion.py

@@ -1,46 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
-
-
-class MalemotionIE(InfoExtractor):
-    _VALID_URL = r'https?://malemotion\.com/video/(.+?)\.(?P<id>.+?)(#|$)'
-    _TEST = {
-        'url': 'http://malemotion.com/video/bete-de-concours.ltc',
-        'md5': '3013e53a0afbde2878bc39998c33e8a5',
-        'info_dict': {
-            'id': 'ltc',
-            'ext': 'mp4',
-            'title': 'Bête de Concours',
-            'age_limit': 18,
-        },
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        video_url = compat_urllib_parse_unquote(self._search_regex(
-            r'<source type="video/mp4" src="(.+?)"', webpage, 'video URL'))
-        video_title = self._html_search_regex(
-            r'<title>(.*?)</title', webpage, 'title')
-        video_thumbnail = self._search_regex(
-            r'<video .+?poster="(.+?)"', webpage, 'thumbnail', fatal=False)
-
-        formats = [{
-            'url': video_url,
-            'ext': 'mp4',
-            'format_id': 'mp4',
-            'preference': 1,
-        }]
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'formats': formats,
-            'title': video_title,
-            'thumbnail': video_thumbnail,
-            'age_limit': 18,
-        }

youtube_dl/extractor/mdr.py

@@ -49,8 +49,8 @@ class MDRIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Beutolomäus und der geheime Weihnachtswunsch',
             'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
-            'timestamp': 1419047100,
+            'timestamp': 1450950000,
-            'upload_date': '20141220',
+            'upload_date': '20151224',
             'duration': 4628,
             'uploader': 'KIKA',
         },
@@ -71,8 +71,8 @@ class MDRIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         data_url = self._search_regex(
-            r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>\\?/.+/(?:video|audio)-?[0-9]+-avCustom\.xml)\1',
+            r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>.+/(?:video|audio)-?[0-9]+-avCustom\.xml)\1',
-            webpage, 'data url', default=None, group='url').replace('\/', '/')
+            webpage, 'data url', group='url').replace('\/', '/')
 
         doc = self._download_xml(
             compat_urlparse.urljoin(url, data_url), video_id)

youtube_dl/extractor/metacafe.py

@@ -81,6 +81,9 @@ class MetacafeIE(InfoExtractor):
                 'title': 'Open: This is Face the Nation, February 9',
                 'description': 'md5:8a9ceec26d1f7ed6eab610834cc1a476',
                 'duration': 96,
+                'uploader': 'CBSI-NEW',
+                'upload_date': '20140209',
+                'timestamp': 1391959800,
             },
             'params': {
                 # rtmp download

youtube_dl/extractor/metacritic.py

@@ -11,7 +11,7 @@ from ..utils import (
 class MetacriticIE(InfoExtractor):
     _VALID_URL = r'https?://www\.metacritic\.com/.+?/trailers/(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.metacritic.com/game/playstation-4/infamous-second-son/trailers/3698222',
         'info_dict': {
             'id': '3698222',
@@ -20,7 +20,17 @@ class MetacriticIE(InfoExtractor):
             'description': 'Take a peak behind-the-scenes to see how Sucker Punch brings smoke into the universe of inFAMOUS Second Son on the PS4.',
             'duration': 221,
         },
-    }
+        'skip': 'Not providing trailers anymore',
+    }, {
+        'url': 'http://www.metacritic.com/game/playstation-4/tales-from-the-borderlands-a-telltale-game-series/trailers/5740315',
+        'info_dict': {
+            'id': '5740315',
+            'ext': 'mp4',
+            'title': 'Tales from the Borderlands - Finale: The Vault of the Traveler',
+            'description': 'In the final episode of the season, all hell breaks loose. Jack is now in control of Helios\' systems, and he\'s ready to reclaim his rightful place as king of Hyperion (with or without you).',
+            'duration': 114,
+        },
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/mgtv.py

@@ -0,0 +1,64 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import int_or_none
+
+
+class MGTVIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.mgtv\.com/v/(?:[^/]+/)*(?P<id>\d+)\.html'
+    IE_DESC = '芒果TV'
+
+    _TEST = {
+        'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
+        'md5': '1bdadcf760a0b90946ca68ee9a2db41a',
+        'info_dict': {
+            'id': '3116640',
+            'ext': 'mp4',
+            'title': '我是歌手第四季双年巅峰会：韩红李玟“双王”领军对抗',
+            'description': '我是歌手第四季双年巅峰会',
+            'duration': 7461,
+            'thumbnail': 're:^https?://.*\.jpg$',
+        },
+    }
+
+    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']
+        info = api_data['info']
+
+        formats = []
+        for idx, stream in enumerate(api_data['stream']):
+            stream_url = stream.get('url')
+            if not stream_url:
+                continue
+            tbr = int_or_none(self._search_regex(
+                r'(\d+)\.mp4', stream_url, 'tbr', default=None))
+
+            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)
+                return {
+                    'format_id': format_id,
+                    'url': format_info['info'],
+                    'ext': 'mp4',
+                    'tbr': tbr,
+                }
+
+            formats.append(extract_format(
+                stream_url, 'hls-%d' % tbr if tbr else None, idx * 2))
+            formats.append(extract_format(stream_url.replace(
+                '/playlist.m3u8', ''), 'http-%d' % tbr if tbr else None, idx * 2 + 1, {'pno': 1031}))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': info['title'].strip(),
+            'formats': formats,
+            'description': info.get('desc'),
+            'duration': int_or_none(info.get('duration')),
+            'thumbnail': info.get('thumb'),
+        }

youtube_dl/extractor/microsoftvirtualacademy.py

@@ -0,0 +1,192 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_xpath,
+)
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    smuggle_url,
+    unsmuggle_url,
+    xpath_text,
+)
+
+
+class MicrosoftVirtualAcademyBaseIE(InfoExtractor):
+    def _extract_base_url(self, course_id, display_id):
+        return self._download_json(
+            'https://api-mlxprod.microsoft.com/services/products/anonymous/%s' % course_id,
+            display_id, 'Downloading course base URL')
+
+    def _extract_chapter_and_title(self, title):
+        if not title:
+            return None, None
+        m = re.search(r'(?P<chapter>\d+)\s*\|\s*(?P<title>.+)', title)
+        return (int(m.group('chapter')), m.group('title')) if m else (None, title)
+
+
+class MicrosoftVirtualAcademyIE(MicrosoftVirtualAcademyBaseIE):
+    IE_NAME = 'mva'
+    IE_DESC = 'Microsoft Virtual Academy videos'
+    _VALID_URL = r'(?:%s:|https?://(?:mva\.microsoft|(?:www\.)?microsoftvirtualacademy)\.com/[^/]+/training-courses/[^/?#&]+-)(?P<course_id>\d+)(?::|\?l=)(?P<id>[\da-zA-Z]+_\d+)' % IE_NAME
+
+    _TESTS = [{
+        'url': 'https://mva.microsoft.com/en-US/training-courses/microsoft-azure-fundamentals-virtual-machines-11788?l=gfVXISmEB_6804984382',
+        'md5': '7826c44fc31678b12ad8db11f6b5abb9',
+        'info_dict': {
+            'id': 'gfVXISmEB_6804984382',
+            'ext': 'mp4',
+            'title': 'Course Introduction',
+            'formats': 'mincount:3',
+            'subtitles': {
+                'en': [{
+                    'ext': 'ttml',
+                }],
+            },
+        }
+    }, {
+        'url': 'mva:11788:gfVXISmEB_6804984382',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
+
+        mobj = re.match(self._VALID_URL, url)
+        course_id = mobj.group('course_id')
+        video_id = mobj.group('id')
+
+        base_url = smuggled_data.get('base_url') or self._extract_base_url(course_id, video_id)
+
+        settings = self._download_xml(
+            '%s/content/content_%s/videosettings.xml?v=1' % (base_url, video_id),
+            video_id, 'Downloading video settings XML')
+
+        _, title = self._extract_chapter_and_title(xpath_text(
+            settings, './/Title', 'title', fatal=True))
+
+        formats = []
+
+        for sources in settings.findall(compat_xpath('.//MediaSources')):
+            if sources.get('videoType') == 'smoothstreaming':
+                continue
+            for source in sources.findall(compat_xpath('./MediaSource')):
+                video_url = source.text
+                if not video_url or not video_url.startswith('http'):
+                    continue
+                video_mode = source.get('videoMode')
+                height = int_or_none(self._search_regex(
+                    r'^(\d+)[pP]$', video_mode or '', 'height', default=None))
+                codec = source.get('codec')
+                acodec, vcodec = [None] * 2
+                if codec:
+                    codecs = codec.split(',')
+                    if len(codecs) == 2:
+                        acodec, vcodec = codecs
+                    elif len(codecs) == 1:
+                        vcodec = codecs[0]
+                formats.append({
+                    'url': video_url,
+                    'format_id': video_mode,
+                    'height': height,
+                    'acodec': acodec,
+                    'vcodec': vcodec,
+                })
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for source in settings.findall(compat_xpath('.//MarkerResourceSource')):
+            subtitle_url = source.text
+            if not subtitle_url:
+                continue
+            subtitles.setdefault('en', []).append({
+                'url': '%s/%s' % (base_url, subtitle_url),
+                'ext': source.get('type'),
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'subtitles': subtitles,
+            'formats': formats
+        }
+
+
+class MicrosoftVirtualAcademyCourseIE(MicrosoftVirtualAcademyBaseIE):
+    IE_NAME = 'mva:course'
+    IE_DESC = 'Microsoft Virtual Academy courses'
+    _VALID_URL = r'(?:%s:|https?://(?:mva\.microsoft|(?:www\.)?microsoftvirtualacademy)\.com/[^/]+/training-courses/(?P<display_id>[^/?#&]+)-)(?P<id>\d+)' % IE_NAME
+
+    _TESTS = [{
+        'url': 'https://mva.microsoft.com/en-US/training-courses/microsoft-azure-fundamentals-virtual-machines-11788',
+        'info_dict': {
+            'id': '11788',
+            'title': 'Microsoft Azure Fundamentals: Virtual Machines',
+        },
+        'playlist_count': 36,
+    }, {
+        # with emphasized chapters
+        'url': 'https://mva.microsoft.com/en-US/training-courses/developing-windows-10-games-with-construct-2-16335',
+        'info_dict': {
+            'id': '16335',
+            'title': 'Developing Windows 10 Games with Construct 2',
+        },
+        'playlist_count': 10,
+    }, {
+        'url': 'https://www.microsoftvirtualacademy.com/en-US/training-courses/microsoft-azure-fundamentals-virtual-machines-11788',
+        'only_matching': True,
+    }, {
+        'url': 'mva:course:11788',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if MicrosoftVirtualAcademyIE.suitable(url) else super(
+            MicrosoftVirtualAcademyCourseIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        course_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        base_url = self._extract_base_url(course_id, display_id)
+
+        manifest = self._download_json(
+            '%s/imsmanifestlite.json' % base_url,
+            display_id, 'Downloading course manifest JSON')['manifest']
+
+        organization = manifest['organizations']['organization'][0]
+
+        entries = []
+        for chapter in organization['item']:
+            chapter_number, chapter_title = self._extract_chapter_and_title(chapter.get('title'))
+            chapter_id = chapter.get('@identifier')
+            for item in chapter.get('item', []):
+                item_id = item.get('@identifier')
+                if not item_id:
+                    continue
+                metadata = item.get('resource', {}).get('metadata') or {}
+                if metadata.get('learningresourcetype') != 'Video':
+                    continue
+                _, title = self._extract_chapter_and_title(item.get('title'))
+                duration = parse_duration(metadata.get('duration'))
+                description = metadata.get('description')
+                entries.append({
+                    '_type': 'url_transparent',
+                    'url': smuggle_url(
+                        'mva:%s:%s' % (course_id, item_id), {'base_url': base_url}),
+                    'title': title,
+                    'description': description,
+                    'duration': duration,
+                    'chapter': chapter_title,
+                    'chapter_number': chapter_number,
+                    'chapter_id': chapter_id,
+                })
+
+        title = organization.get('title') or manifest.get('metadata', {}).get('title')
+
+        return self.playlist_result(entries, course_id, title)

youtube_dl/extractor/ministrygrid.py

@@ -1,8 +1,5 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
@@ -20,21 +17,28 @@ class MinistryGridIE(InfoExtractor):
             'id': '3453494717001',
             'ext': 'mp4',
             'title': 'The Gospel by Numbers',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'upload_date': '20140410',
             'description': 'Coming soon from T4G 2014!',
-            'uploader': 'LifeWay Christian Resources (MG)',
+            'uploader_id': '2034960640001',
+            'timestamp': 1397145591,
         },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'add_ie': ['TDSLifeway'],
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id)
-        portlets_json = self._search_regex(
+        portlets = self._parse_json(self._search_regex(
-            r'Liferay\.Portlet\.list=(\[.+?\])', webpage, 'portlet list')
+            r'Liferay\.Portlet\.list=(\[.+?\])', webpage, 'portlet list'),
-        portlets = json.loads(portlets_json)
+            video_id)
         pl_id = self._search_regex(
-            r'<!--\s*p_l_id - ([0-9]+)<br>', webpage, 'p_l_id')
+            r'getPlid:function\(\){return"(\d+)"}', webpage, 'p_l_id')
 
         for i, portlet in enumerate(portlets):
             portlet_url = 'http://www.ministrygrid.com/c/portal/render_portlet?p_l_id=%s&p_p_id=%s' % (pl_id, portlet)
@@ -46,12 +50,8 @@ class MinistryGridIE(InfoExtractor):
                 r'<iframe.*?src="([^"]+)"', portlet_code, 'video iframe',
                 default=None)
             if video_iframe_url:
-                surl = smuggle_url(
+                return self.url_result(
-                    video_iframe_url, {'force_videoid': video_id})
+                    smuggle_url(video_iframe_url, {'force_videoid': video_id}),
-                return {
+                    video_id=video_id)
-                    '_type': 'url',
-                    'id': video_id,
-                    'url': surl,
-                }
 
         raise ExtractorError('Could not find video iframe in any portlets')

youtube_dl/extractor/mitele.py

@@ -15,9 +15,9 @@ class MiTeleIE(InfoExtractor):
     IE_DESC = 'mitele.es'
     _VALID_URL = r'https?://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
 
-    _TESTS = [{
+    _TEST = {
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
-        'md5': '0ff1a13aebb35d9bc14081ff633dd324',
+        # MD5 is unstable
         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',
@@ -27,7 +27,7 @@ class MiTeleIE(InfoExtractor):
             'thumbnail': 're:(?i)^https?://.*\.jpg$',
             'duration': 2913,
         },
-    }]
+    }
 
     def _real_extract(self, url):
         display_id = self._match_id(url)

youtube_dl/extractor/mixcloud.py

@@ -1,26 +1,35 @@
 from __future__ import unicode_literals
 
+import base64
+import functools
+import itertools
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
+from ..compat import (
+    compat_chr,
+    compat_ord,
+    compat_urllib_parse_unquote,
+    compat_urlparse,
+)
 from ..utils import (
+    clean_html,
     ExtractorError,
-    HEADRequest,
+    OnDemandPagedList,
     parse_count,
     str_to_int,
 )
 
 
 class MixcloudIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/([^/]+)'
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)'
     IE_NAME = 'mixcloud'
 
     _TESTS = [{
         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
         'info_dict': {
             'id': 'dholbach-cryptkeeper',
-            'ext': 'mp3',
+            'ext': 'm4a',
             'title': 'Cryptkeeper',
             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
             'uploader': 'Daniel Holbach',
@@ -38,22 +47,22 @@ class MixcloudIE(InfoExtractor):
             'description': 'md5:2b8aec6adce69f9d41724647c65875e8',
             'uploader': 'Gilles Peterson Worldwide',
             'uploader_id': 'gillespeterson',
-            'thumbnail': 're:https?://.*/images/',
+            'thumbnail': 're:https?://.*',
             'view_count': int,
             'like_count': int,
         },
     }]
 
-    def _check_url(self, url, track_id, ext):
+    # See https://www.mixcloud.com/media/js2/www_js_2.9e23256562c080482435196ca3975ab5.js
-        try:
+    @staticmethod
-            # We only want to know if the request succeed
+    def _decrypt_play_info(play_info):
-            # don't download the whole file
+        KEY = 'pleasedontdownloadourmusictheartistswontgetpaid'
-            self._request_webpage(
+
-                HEADRequest(url), track_id,
+        play_info = base64.b64decode(play_info.encode('ascii'))
-                'Trying %s URL' % ext)
+
-            return True
+        return ''.join([
-        except ExtractorError:
+            compat_chr(compat_ord(ch) ^ compat_ord(KEY[idx % len(KEY)]))
-            return False
+            for idx, ch in enumerate(play_info)])
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -63,14 +72,19 @@ class MixcloudIE(InfoExtractor):
 
         webpage = self._download_webpage(url, track_id)
 
-        preview_url = self._search_regex(
+        message = self._html_search_regex(
-            r'\s(?:data-preview-url|m-preview)="([^"]+)"', webpage, 'preview url')
+            r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
-        song_url = re.sub(r'audiocdn(\d+)', r'stream\1', preview_url)
+            webpage, 'error message', default=None)
-        song_url = song_url.replace('/previews/', '/c/originals/')
+
-        if not self._check_url(song_url, track_id, 'mp3'):
+        encrypted_play_info = self._search_regex(
-            song_url = song_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/')
+            r'm-play-info="([^"]+)"', webpage, 'play info')
-            if not self._check_url(song_url, track_id, 'm4a'):
+        play_info = self._parse_json(
-                raise ExtractorError('Unable to extract track url')
+            self._decrypt_play_info(encrypted_play_info), track_id)
+
+        if message and 'stream_url' not in play_info:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+
+        song_url = play_info['stream_url']
 
         PREFIX = (
             r'm-play-on-spacebar[^>]+'
@@ -105,3 +119,201 @@ class MixcloudIE(InfoExtractor):
             'view_count': view_count,
             'like_count': like_count,
         }
+
+
+class MixcloudPlaylistBaseIE(InfoExtractor):
+    _PAGE_SIZE = 24
+
+    def _find_urls_in_page(self, page):
+        for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', page):
+            yield self.url_result(
+                compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)),
+                MixcloudIE.ie_key())
+
+    def _fetch_tracks_page(self, path, video_id, page_name, current_page, real_page_number=None):
+        real_page_number = real_page_number or current_page + 1
+        return self._download_webpage(
+            'https://www.mixcloud.com/%s/' % path, video_id,
+            note='Download %s (page %d)' % (page_name, current_page + 1),
+            errnote='Unable to download %s' % page_name,
+            query={'page': real_page_number, 'list': 'main', '_ajax': '1'},
+            headers={'X-Requested-With': 'XMLHttpRequest'})
+
+    def _tracks_page_func(self, page, video_id, page_name, current_page):
+        resp = self._fetch_tracks_page(page, video_id, page_name, current_page)
+
+        for item in self._find_urls_in_page(resp):
+            yield item
+
+    def _get_user_description(self, page_content):
+        return self._html_search_regex(
+            r'<div[^>]+class="description-text"[^>]*>(.+?)</div>',
+            page_content, 'user description', fatal=False)
+
+
+class MixcloudUserIE(MixcloudPlaylistBaseIE):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
+    IE_NAME = 'mixcloud:user'
+
+    _TESTS = [{
+        'url': 'http://www.mixcloud.com/dholbach/',
+        'info_dict': {
+            'id': 'dholbach_uploads',
+            'title': 'Daniel Holbach (uploads)',
+            'description': 'md5:327af72d1efeb404a8216c27240d1370',
+        },
+        'playlist_mincount': 11,
+    }, {
+        'url': 'http://www.mixcloud.com/dholbach/uploads/',
+        'info_dict': {
+            'id': 'dholbach_uploads',
+            'title': 'Daniel Holbach (uploads)',
+            'description': 'md5:327af72d1efeb404a8216c27240d1370',
+        },
+        'playlist_mincount': 11,
+    }, {
+        'url': 'http://www.mixcloud.com/dholbach/favorites/',
+        'info_dict': {
+            'id': 'dholbach_favorites',
+            'title': 'Daniel Holbach (favorites)',
+            'description': 'md5:327af72d1efeb404a8216c27240d1370',
+        },
+        'params': {
+            'playlist_items': '1-100',
+        },
+        'playlist_mincount': 100,
+    }, {
+        'url': 'http://www.mixcloud.com/dholbach/listens/',
+        'info_dict': {
+            'id': 'dholbach_listens',
+            'title': 'Daniel Holbach (listens)',
+            'description': 'md5:327af72d1efeb404a8216c27240d1370',
+        },
+        'params': {
+            'playlist_items': '1-100',
+        },
+        'playlist_mincount': 100,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        user_id = mobj.group('user')
+        list_type = mobj.group('type')
+
+        # if only a profile URL was supplied, default to download all uploads
+        if list_type is None:
+            list_type = 'uploads'
+
+        video_id = '%s_%s' % (user_id, list_type)
+
+        profile = self._download_webpage(
+            'https://www.mixcloud.com/%s/' % user_id, video_id,
+            note='Downloading user profile',
+            errnote='Unable to download user profile')
+
+        username = self._og_search_title(profile)
+        description = self._get_user_description(profile)
+
+        entries = OnDemandPagedList(
+            functools.partial(
+                self._tracks_page_func,
+                '%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type),
+            self._PAGE_SIZE, use_cache=True)
+
+        return self.playlist_result(
+            entries, video_id, '%s (%s)' % (username, list_type), description)
+
+
+class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
+    IE_NAME = 'mixcloud:playlist'
+
+    _TESTS = [{
+        'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/',
+        'info_dict': {
+            'id': 'RedBullThre3style_tokyo-finalists-2015',
+            'title': 'National Champions 2015',
+            'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3',
+        },
+        'playlist_mincount': 16,
+    }, {
+        'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
+        'info_dict': {
+            'id': 'maxvibes_jazzcat-on-ness-radio',
+            'title': 'Jazzcat on Ness Radio',
+            'description': 'md5:7bbbf0d6359a0b8cda85224be0f8f263',
+        },
+        'playlist_mincount': 23
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        user_id = mobj.group('user')
+        playlist_id = mobj.group('playlist')
+        video_id = '%s_%s' % (user_id, playlist_id)
+
+        profile = self._download_webpage(
+            url, user_id,
+            note='Downloading playlist page',
+            errnote='Unable to download playlist page')
+
+        description = self._get_user_description(profile)
+        playlist_title = self._html_search_regex(
+            r'<span[^>]+class="[^"]*list-playlist-title[^"]*"[^>]*>(.*?)</span>',
+            profile, 'playlist title')
+
+        entries = OnDemandPagedList(
+            functools.partial(
+                self._tracks_page_func,
+                '%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'),
+            self._PAGE_SIZE)
+
+        return self.playlist_result(entries, video_id, playlist_title, description)
+
+
+class MixcloudStreamIE(MixcloudPlaylistBaseIE):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
+    IE_NAME = 'mixcloud:stream'
+
+    _TEST = {
+        'url': 'https://www.mixcloud.com/FirstEar/stream/',
+        'info_dict': {
+            'id': 'FirstEar',
+            'title': 'First Ear',
+            'description': 'Curators of good music\nfirstearmusic.com',
+        },
+        'playlist_mincount': 192,
+    }
+
+    def _real_extract(self, url):
+        user_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, user_id)
+
+        entries = []
+        prev_page_url = None
+
+        def _handle_page(page):
+            entries.extend(self._find_urls_in_page(page))
+            return self._search_regex(
+                r'm-next-page-url="([^"]+)"', page,
+                'next page URL', default=None)
+
+        next_page_url = _handle_page(webpage)
+
+        for idx in itertools.count(0):
+            if not next_page_url or prev_page_url == next_page_url:
+                break
+
+            prev_page_url = next_page_url
+            current_page = int(self._search_regex(
+                r'\?page=(\d+)', next_page_url, 'next page number'))
+
+            next_page_url = _handle_page(self._fetch_tracks_page(
+                '%s/stream' % user_id, user_id, 'stream', idx,
+                real_page_number=current_page))
+
+        username = self._og_search_title(webpage)
+        description = self._get_user_description(webpage)
+
+        return self.playlist_result(entries, user_id, username, description)

youtube_dl/extractor/mooshare.py

@@ -1,110 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    sanitized_Request,
-    urlencode_postdata,
-)
-
-
-class MooshareIE(InfoExtractor):
-    IE_NAME = 'mooshare'
-    IE_DESC = 'Mooshare.biz'
-    _VALID_URL = r'https?://(?:www\.)?mooshare\.biz/(?P<id>[\da-z]{12})'
-
-    _TESTS = [
-        {
-            'url': 'http://mooshare.biz/8dqtk4bjbp8g',
-            'md5': '4e14f9562928aecd2e42c6f341c8feba',
-            'info_dict': {
-                'id': '8dqtk4bjbp8g',
-                'ext': 'mp4',
-                'title': 'Comedy Football 2011 - (part 1-2)',
-                'duration': 893,
-            },
-        },
-        {
-            'url': 'http://mooshare.biz/aipjtoc4g95j',
-            'info_dict': {
-                'id': 'aipjtoc4g95j',
-                'ext': 'mp4',
-                'title': 'Orange Caramel  Dashing Through the Snow',
-                'duration': 212,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            }
-        }
-    ]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        page = self._download_webpage(url, video_id, 'Downloading page')
-
-        if re.search(r'>Video Not Found or Deleted<', page) is not None:
-            raise ExtractorError('Video %s does not exist' % video_id, expected=True)
-
-        hash_key = self._html_search_regex(r'<input type="hidden" name="hash" value="([^"]+)">', page, 'hash')
-        title = self._html_search_regex(r'(?m)<div class="blockTitle">\s*<h2>Watch ([^<]+)</h2>', page, 'title')
-
-        download_form = {
-            'op': 'download1',
-            'id': video_id,
-            'hash': hash_key,
-        }
-
-        request = sanitized_Request(
-            'http://mooshare.biz/%s' % video_id, urlencode_postdata(download_form))
-        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-
-        self._sleep(5, video_id)
-
-        video_page = self._download_webpage(request, video_id, 'Downloading video page')
-
-        thumbnail = self._html_search_regex(r'image:\s*"([^"]+)",', video_page, 'thumbnail', fatal=False)
-        duration_str = self._html_search_regex(r'duration:\s*"(\d+)",', video_page, 'duration', fatal=False)
-        duration = int(duration_str) if duration_str is not None else None
-
-        formats = []
-
-        # SD video
-        mobj = re.search(r'(?m)file:\s*"(?P<url>[^"]+)",\s*provider:', video_page)
-        if mobj is not None:
-            formats.append({
-                'url': mobj.group('url'),
-                'format_id': 'sd',
-                'format': 'SD',
-            })
-
-        # HD video
-        mobj = re.search(r'\'hd-2\': { file: \'(?P<url>[^\']+)\' },', video_page)
-        if mobj is not None:
-            formats.append({
-                'url': mobj.group('url'),
-                'format_id': 'hd',
-                'format': 'HD',
-            })
-
-        # rtmp video
-        mobj = re.search(r'(?m)file: "(?P<playpath>[^"]+)",\s*streamer: "(?P<rtmpurl>rtmp://[^"]+)",', video_page)
-        if mobj is not None:
-            formats.append({
-                'url': mobj.group('rtmpurl'),
-                'play_path': mobj.group('playpath'),
-                'rtmp_live': False,
-                'ext': 'mp4',
-                'format_id': 'rtmp',
-                'format': 'HD',
-            })
-
-        return {
-            'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-        }

youtube_dl/extractor/musicplayon.py

@@ -1,17 +1,21 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..compat import compat_urlparse
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    mimetype2ext,
+)
 
 
 class MusicPlayOnIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:.+?\.)?musicplayon\.com/play(?:-touch)?\?(?:v|pl=100&play)=(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:.+?\.)?musicplayon\.com/play(?:-touch)?\?(?:v|pl=\d+&play)=(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://en.musicplayon.com/play?v=433377',
+        'md5': '00cdcdea1726abdf500d1e7fd6dd59bb',
         'info_dict': {
             'id': '433377',
             'ext': 'mp4',
@@ -20,15 +24,16 @@ class MusicPlayOnIE(InfoExtractor):
             'duration': 342,
             'uploader': 'ultrafish',
         },
-        'params': {
+    }, {
-            # m3u8 download
+        'url': 'http://en.musicplayon.com/play?pl=102&play=442629',
-            'skip_download': True,
+        'only_matching': True,
-        },
+    }]
-    }
+
+    _URL_TEMPLATE = 'http://en.musicplayon.com/play?v=%s'
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
+        url = self._URL_TEMPLATE % video_id
 
         page = self._download_webpage(url, video_id)
 
@@ -40,28 +45,14 @@ class MusicPlayOnIE(InfoExtractor):
         uploader = self._html_search_regex(
             r'<div>by&nbsp;<a href="[^"]+" class="purple">([^<]+)</a></div>', page, 'uploader', fatal=False)
 
-        formats = [
+        sources = self._parse_json(
-            {
+            self._search_regex(r'setup\[\'_sources\'\]\s*=\s*([^;]+);', page, 'video sources'),
-                'url': 'http://media0-eu-nl.musicplayon.com/stream-mobile?id=%s&type=.mp4' % video_id,
+            video_id, transform_source=js_to_json)
-                'ext': 'mp4',
+        formats = [{
-            }
+            'url': compat_urlparse.urljoin(url, source['src']),
-        ]
+            'ext': mimetype2ext(source.get('type')),
-
+            'format_note': source.get('data-res'),
-        manifest = self._download_webpage(
+        } for source in sources]
-            'http://en.musicplayon.com/manifest.m3u8?v=%s' % video_id, video_id, 'Downloading manifest')
-
-        for entry in manifest.split('#')[1:]:
-            if entry.startswith('EXT-X-STREAM-INF:'):
-                meta, url, _ = entry.split('\n')
-                params = dict(param.split('=') for param in meta.split(',')[1:])
-                formats.append({
-                    'url': url,
-                    'ext': 'mp4',
-                    'tbr': int(params['BANDWIDTH']),
-                    'width': int(params['RESOLUTION'].split('x')[1]),
-                    'height': int(params['RESOLUTION'].split('x')[-1]),
-                    'format_note': params['NAME'].replace('"', '').strip(),
-                })
 
         return {
             'id': video_id,

youtube_dl/extractor/muzu.py

@@ -1,63 +0,0 @@
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlencode
-
-
-class MuzuTVIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.muzu\.tv/(.+?)/(.+?)/(?P<id>\d+)'
-    IE_NAME = 'muzu.tv'
-
-    _TEST = {
-        'url': 'http://www.muzu.tv/defected/marcashken-featuring-sos-cat-walk-original-mix-music-video/1981454/',
-        'md5': '98f8b2c7bc50578d6a0364fff2bfb000',
-        'info_dict': {
-            'id': '1981454',
-            'ext': 'mp4',
-            'title': 'Cat Walk (Original Mix)',
-            'description': 'md5:90e868994de201b2570e4e5854e19420',
-            'uploader': 'MarcAshken featuring SOS',
-        },
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        info_data = compat_urllib_parse_urlencode({
-            'format': 'json',
-            'url': url,
-        })
-        info = self._download_json(
-            'http://www.muzu.tv/api/oembed/?%s' % info_data,
-            video_id, 'Downloading video info')
-
-        player_info = self._download_json(
-            'http://player.muzu.tv/player/playerInit?ai=%s' % video_id,
-            video_id, 'Downloading player info')
-        video_info = player_info['videos'][0]
-        for quality in ['1080', '720', '480', '360']:
-            if video_info.get('v%s' % quality):
-                break
-
-        data = compat_urllib_parse_urlencode({
-            'ai': video_id,
-            # Even if each time you watch a video the hash changes,
-            # it seems to work for different videos, and it will work
-            # even if you use any non empty string as a hash
-            'viewhash': 'VBNff6djeV4HV5TRPW5kOHub2k',
-            'device': 'web',
-            'qv': quality,
-        })
-        video_url_info = self._download_json(
-            'http://player.muzu.tv/player/requestVideo?%s' % data,
-            video_id, 'Downloading video url')
-        video_url = video_url_info['url']
-
-        return {
-            'id': video_id,
-            'title': info['title'],
-            'url': video_url,
-            'thumbnail': info['thumbnail_url'],
-            'description': info['description'],
-            'uploader': info['author_name'],
-        }

youtube_dl/extractor/mwave.py

@@ -10,9 +10,10 @@ from ..utils import (
 
 class MwaveIE(InfoExtractor):
     _VALID_URL = r'https?://mwave\.interest\.me/mnettv/videodetail\.m\?searchVideoDetailVO\.clip_id=(?P<id>[0-9]+)'
+    _URL_TEMPLATE = 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=%s'
     _TEST = {
         'url': 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=168859',
-        'md5': 'c930e27b7720aaa3c9d0018dfc8ff6cc',
+        # md5 is unstable
         'info_dict': {
             'id': '168859',
             'ext': 'flv',
@@ -56,3 +57,28 @@ class MwaveIE(InfoExtractor):
             'view_count': int_or_none(vod_info.get('hit')),
             'formats': formats,
         }
+
+
+class MwaveMeetGreetIE(InfoExtractor):
+    _VALID_URL = r'https?://mwave\.interest\.me/meetgreet/view/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://mwave.interest.me/meetgreet/view/256',
+        'info_dict': {
+            'id': '173294',
+            'ext': 'flv',
+            'title': '[MEET&GREET] Park BoRam',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Mwave',
+            'duration': 3634,
+            'view_count': int,
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        clip_id = self._html_search_regex(
+            r'<iframe[^>]+src="/mnettv/ifr_clip\.m\?searchVideoDetailVO\.clip_id=(\d+)',
+            webpage, 'clip ID')
+        clip_url = MwaveIE._URL_TEMPLATE % clip_id
+        return self.url_result(clip_url, 'Mwave', clip_id)

youtube_dl/extractor/nbc.py

@@ -134,6 +134,9 @@ class NBCSportsIE(InfoExtractor):
             'ext': 'flv',
             'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
             'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
+            'uploader': 'NBCU-SPORTS',
+            'upload_date': '20150330',
+            'timestamp': 1427726529,
         }
     }
 
@@ -172,7 +175,7 @@ class CSNNEIE(InfoExtractor):
 
 
 class NBCNewsIE(ThePlatformIE):
-    _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
+    _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today)\.com/
         (?:video/.+?/(?P<id>\d+)|
         ([^/]+/)*(?P<display_id>[^/?]+))
         '''
@@ -230,6 +233,18 @@ class NBCNewsIE(ThePlatformIE):
             },
             'expected_warnings': ['http-6000 is not available']
         },
+        {
+            'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
+            'md5': '118d7ca3f0bea6534f119c68ef539f71',
+            'info_dict': {
+                'id': '669831235788',
+                'ext': 'mp4',
+                'title': 'See the aurora borealis from space in stunning new NASA video',
+                'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
+                'upload_date': '20160420',
+                'timestamp': 1461152093,
+            },
+        },
         {
             'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
             'only_matching': True,
@@ -264,7 +279,10 @@ class NBCNewsIE(ThePlatformIE):
                 info = bootstrap['results'][0]['video']
             else:
                 player_instance_json = self._search_regex(
-                    r'videoObj\s*:\s*({.+})', webpage, 'player instance')
+                    r'videoObj\s*:\s*({.+})', webpage, 'player instance', default=None)
+                if not player_instance_json:
+                    player_instance_json = self._html_search_regex(
+                        r'data-video="([^"]+)"', webpage, 'video json')
                 info = self._parse_json(player_instance_json, display_id)
             video_id = info['mpxId']
             title = info['title']
@@ -295,7 +313,7 @@ class NBCNewsIE(ThePlatformIE):
                     formats.extend(tp_formats)
                     subtitles = self._merge_subtitles(subtitles, tp_subtitles)
                 else:
-                    tbr = int_or_none(video_asset.get('bitRate'), 1000)
+                    tbr = int_or_none(video_asset.get('bitRate') or video_asset.get('bitrate'), 1000)
                     format_id = 'http%s' % ('-%d' % tbr if tbr else '')
                     video_url = update_url_query(
                         video_url, {'format': 'redirect'})
@@ -321,10 +339,9 @@ class NBCNewsIE(ThePlatformIE):
                 'id': video_id,
                 'title': title,
                 'description': info.get('description'),
-                'thumbnail': info.get('description'),
                 'thumbnail': info.get('thumbnail'),
                 'duration': int_or_none(info.get('duration')),
-                'timestamp': parse_iso8601(info.get('pubDate')),
+                'timestamp': parse_iso8601(info.get('pubDate') or info.get('pub_date')),
                 'formats': formats,
                 'subtitles': subtitles,
             }

youtube_dl/extractor/nerdist.py

@@ -1,80 +0,0 @@
-# encoding: utf-8
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-
-from ..utils import (
-    determine_ext,
-    parse_iso8601,
-    xpath_text,
-)
-
-
-class NerdistIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?nerdist\.com/vepisode/(?P<id>[^/?#]+)'
-    _TEST = {
-        'url': 'http://www.nerdist.com/vepisode/exclusive-which-dc-characters-w',
-        'md5': '3698ed582931b90d9e81e02e26e89f23',
-        'info_dict': {
-            'display_id': 'exclusive-which-dc-characters-w',
-            'id': 'RPHpvJyr',
-            'ext': 'mp4',
-            'title': 'Your TEEN TITANS Revealed! Who\'s on the show?',
-            'thumbnail': 're:^https?://.*/thumbs/.*\.jpg$',
-            'description': 'Exclusive: Find out which DC Comics superheroes will star in TEEN TITANS Live-Action TV Show on Nerdist News with Jessica Chobot!',
-            'uploader': 'Eric Diaz',
-            'upload_date': '20150202',
-            'timestamp': 1422892808,
-        }
-    }
-
-    def _real_extract(self, url):
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
-
-        video_id = self._search_regex(
-            r'''(?x)<script\s+(?:type="text/javascript"\s+)?
-                src="https?://content\.nerdist\.com/players/([a-zA-Z0-9_]+)-''',
-            webpage, 'video ID')
-        timestamp = parse_iso8601(self._html_search_meta(
-            'shareaholic:article_published_time', webpage, 'upload date'))
-        uploader = self._html_search_meta(
-            'shareaholic:article_author_name', webpage, 'article author')
-
-        doc = self._download_xml(
-            'http://content.nerdist.com/jw6/%s.xml' % video_id, video_id)
-        video_info = doc.find('.//item')
-        title = xpath_text(video_info, './title', fatal=True)
-        description = xpath_text(video_info, './description')
-        thumbnail = xpath_text(
-            video_info, './{http://rss.jwpcdn.com/}image', 'thumbnail')
-
-        formats = []
-        for source in video_info.findall('./{http://rss.jwpcdn.com/}source'):
-            vurl = source.attrib['file']
-            ext = determine_ext(vurl)
-            if ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    vurl, video_id, entry_protocol='m3u8_native', ext='mp4',
-                    preference=0))
-            elif ext == 'smil':
-                formats.extend(self._extract_smil_formats(
-                    vurl, video_id, fatal=False
-                ))
-            else:
-                formats.append({
-                    'format_id': ext,
-                    'url': vurl,
-                })
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'display_id': display_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'timestamp': timestamp,
-            'formats': formats,
-            'uploader': uploader,
-        }

youtube_dl/extractor/neteasemusic.py

@@ -89,6 +89,7 @@ class NetEaseMusicIE(NetEaseMusicBaseIE):
             'timestamp': 1431878400,
             'description': 'md5:a10a54589c2860300d02e1de821eb2ef',
         },
+        'skip': 'Blocked outside Mainland China',
     }, {
         'note': 'No lyrics translation.',
         'url': 'http://music.163.com/#/song?id=29822014',
@@ -101,6 +102,7 @@ class NetEaseMusicIE(NetEaseMusicBaseIE):
             'timestamp': 1419523200,
             'description': 'md5:a4d8d89f44656af206b7b2555c0bce6c',
         },
+        'skip': 'Blocked outside Mainland China',
     }, {
         'note': 'No lyrics.',
         'url': 'http://music.163.com/song?id=17241424',
@@ -112,6 +114,7 @@ class NetEaseMusicIE(NetEaseMusicBaseIE):
             'upload_date': '20080211',
             'timestamp': 1202745600,
         },
+        'skip': 'Blocked outside Mainland China',
     }, {
         'note': 'Has translated name.',
         'url': 'http://music.163.com/#/song?id=22735043',
@@ -124,7 +127,8 @@ class NetEaseMusicIE(NetEaseMusicBaseIE):
             'upload_date': '20100127',
             'timestamp': 1264608000,
             'alt_title': '说出愿望吧(Genie)',
-        }
+        },
+        'skip': 'Blocked outside Mainland China',
     }]
 
     def _process_lyrics(self, lyrics_info):
@@ -192,6 +196,7 @@ class NetEaseMusicAlbumIE(NetEaseMusicBaseIE):
             'title': 'B\'day',
         },
         'playlist_count': 23,
+        'skip': 'Blocked outside Mainland China',
     }
 
     def _real_extract(self, url):
@@ -223,6 +228,7 @@ class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
             'title': '张惠妹 - aMEI;阿密特',
         },
         'playlist_count': 50,
+        'skip': 'Blocked outside Mainland China',
     }, {
         'note': 'Singer has translated name.',
         'url': 'http://music.163.com/#/artist?id=124098',
@@ -231,6 +237,7 @@ class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
             'title': '李昇基 - 이승기',
         },
         'playlist_count': 50,
+        'skip': 'Blocked outside Mainland China',
     }]
 
     def _real_extract(self, url):
@@ -266,6 +273,7 @@ class NetEaseMusicListIE(NetEaseMusicBaseIE):
             'description': 'md5:12fd0819cab2965b9583ace0f8b7b022'
         },
         'playlist_count': 99,
+        'skip': 'Blocked outside Mainland China',
     }, {
         'note': 'Toplist/Charts sample',
         'url': 'http://music.163.com/#/discover/toplist?id=3733003',
@@ -275,6 +283,7 @@ class NetEaseMusicListIE(NetEaseMusicBaseIE):
             'description': 'md5:73ec782a612711cadc7872d9c1e134fc',
         },
         'playlist_count': 50,
+        'skip': 'Blocked outside Mainland China',
     }]
 
     def _real_extract(self, url):
@@ -314,6 +323,7 @@ class NetEaseMusicMvIE(NetEaseMusicBaseIE):
             'creator': '白雅言',
             'upload_date': '20150520',
         },
+        'skip': 'Blocked outside Mainland China',
     }
 
     def _real_extract(self, url):
@@ -357,6 +367,7 @@ class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
             'upload_date': '20150613',
             'duration': 900,
         },
+        'skip': 'Blocked outside Mainland China',
     }, {
         'note': 'This program has accompanying songs.',
         'url': 'http://music.163.com/#/program?id=10141022',
@@ -366,6 +377,7 @@ class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
             'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
         },
         'playlist_count': 4,
+        'skip': 'Blocked outside Mainland China',
     }, {
         'note': 'This program has accompanying songs.',
         'url': 'http://music.163.com/#/program?id=10141022',
@@ -379,7 +391,8 @@ class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
         },
         'params': {
             'noplaylist': True
-        }
+        },
+        'skip': 'Blocked outside Mainland China',
     }]
 
     def _real_extract(self, url):
@@ -438,6 +451,7 @@ class NetEaseMusicDjRadioIE(NetEaseMusicBaseIE):
             'description': 'md5:766220985cbd16fdd552f64c578a6b15'
         },
         'playlist_mincount': 40,
+        'skip': 'Blocked outside Mainland China',
     }
     _PAGE_SIZE = 1000
 

youtube_dl/extractor/newgrounds.py

@@ -7,8 +7,8 @@ from .common import InfoExtractor
 
 
 class NewgroundsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/audio/listen/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/(?:audio/listen|portal/view)/(?P<id>[0-9]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.newgrounds.com/audio/listen/549479',
         'md5': 'fe6033d297591288fa1c1f780386f07a',
         'info_dict': {
@@ -17,7 +17,16 @@ class NewgroundsIE(InfoExtractor):
             'title': 'B7 - BusMode',
             'uploader': 'Burn7',
         }
-    }
+    }, {
+        'url': 'http://www.newgrounds.com/portal/view/673111',
+        'md5': '3394735822aab2478c31b1004fe5e5bc',
+        'info_dict': {
+            'id': '673111',
+            'ext': 'mp4',
+            'title': 'Dancin',
+            'uploader': 'Squirrelman82',
+        },
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -25,9 +34,11 @@ class NewgroundsIE(InfoExtractor):
         webpage = self._download_webpage(url, music_id)
 
         title = self._html_search_regex(
-            r',"name":"([^"]+)",', webpage, 'music title')
+            r'<title>([^>]+)</title>', webpage, 'title')
+
         uploader = self._html_search_regex(
-            r',"artist":"([^"]+)",', webpage, 'music uploader')
+            [r',"artist":"([^"]+)",', r'[\'"]owner[\'"]\s*:\s*[\'"]([^\'"]+)[\'"],'],
+            webpage, 'uploader')
 
         music_url_json_string = self._html_search_regex(
             r'({"url":"[^"]+"),', webpage, 'music url') + '}'

youtube_dl/extractor/newstube.py

@@ -4,24 +4,24 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+)
 
 
 class NewstubeIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)'
     _TEST = {
         'url': 'http://www.newstube.ru/media/telekanal-cnn-peremestil-gorod-slavyansk-v-krym',
+        'md5': '801eef0c2a9f4089fa04e4fe3533abdc',
         'info_dict': {
             'id': '728e0ef2-e187-4012-bac0-5a081fdcb1f6',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Телеканал CNN переместил город Славянск в Крым',
             'description': 'md5:419a8c9f03442bc0b0a794d689360335',
             'duration': 31.05,
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
     }
 
     def _real_extract(self, url):
@@ -62,7 +62,6 @@ class NewstubeIE(InfoExtractor):
             server = media_location.find(ns('./Server')).text
             app = media_location.find(ns('./App')).text
             media_id = stream_info.find(ns('./Id')).text
-            quality_id = stream_info.find(ns('./QualityId')).text
             name = stream_info.find(ns('./Name')).text
             width = int(stream_info.find(ns('./Width')).text)
             height = int(stream_info.find(ns('./Height')).text)
@@ -74,12 +73,38 @@ class NewstubeIE(InfoExtractor):
                 'rtmp_conn': ['S:%s' % session_id, 'S:%s' % media_id, 'S:n2'],
                 'page_url': url,
                 'ext': 'flv',
-                'format_id': quality_id,
+                'format_id': 'rtmp' + ('-%s' % name if name else ''),
-                'format_note': name,
                 'width': width,
                 'height': height,
             })
 
+        sources_data = self._download_json(
+            'http://www.newstube.ru/player2/getsources?guid=%s' % video_guid,
+            video_guid, fatal=False)
+        if sources_data:
+            for source in sources_data.get('Sources', []):
+                source_url = source.get('Src')
+                if not source_url:
+                    continue
+                height = int_or_none(source.get('Height'))
+                f = {
+                    'format_id': 'http' + ('-%dp' % height if height else ''),
+                    'url': source_url,
+                    'width': int_or_none(source.get('Width')),
+                    'height': height,
+                }
+                source_type = source.get('Type')
+                if source_type:
+                    mobj = re.search(r'codecs="([^,]+),\s*([^"]+)"', source_type)
+                    if mobj:
+                        vcodec, acodec = mobj.groups()
+                        f.update({
+                            'vcodec': vcodec,
+                            'acodec': acodec,
+                        })
+                formats.append(f)
+
+        self._check_formats(formats, video_guid)
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/nhl.py

@@ -8,10 +8,15 @@ from .common import InfoExtractor
 from ..compat import (
     compat_urlparse,
     compat_urllib_parse_urlencode,
-    compat_urllib_parse_urlparse
+    compat_urllib_parse_urlparse,
+    compat_str,
 )
 from ..utils import (
     unified_strdate,
+    determine_ext,
+    int_or_none,
+    parse_iso8601,
+    parse_duration,
 )
 
 
@@ -70,8 +75,8 @@ class NHLBaseInfoExtractor(InfoExtractor):
         return ret
 
 
-class NHLIE(NHLBaseInfoExtractor):
+class NHLVideocenterIE(NHLBaseInfoExtractor):
-    IE_NAME = 'nhl.com'
+    IE_NAME = 'nhl.com:videocenter'
     _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console|embed)?(?:\?(?:.*?[?&])?)(?:id|hlg|playlist)=(?P<id>[-0-9a-zA-Z,]+)'
 
     _TESTS = [{
@@ -186,8 +191,8 @@ class NHLNewsIE(NHLBaseInfoExtractor):
         return self._real_extract_video(video_id)
 
 
-class NHLVideocenterIE(NHLBaseInfoExtractor):
+class NHLVideocenterCategoryIE(NHLBaseInfoExtractor):
-    IE_NAME = 'nhl.com:videocenter'
+    IE_NAME = 'nhl.com:videocenter:category'
     IE_DESC = 'NHL videocenter category'
     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?[^(id=)]*catid=(?P<catid>[0-9]+)(?![&?]id=).*?)?$'
     _TEST = {
@@ -236,3 +241,86 @@ class NHLVideocenterIE(NHLBaseInfoExtractor):
             'id': cat_id,
             'entries': [self._extract_video(v) for v in videos],
         }
+
+
+class NHLIE(InfoExtractor):
+    IE_NAME = 'nhl.com'
+    _VALID_URL = r'https?://(?:www\.)?nhl\.com/([^/]+/)*c-(?P<id>\d+)'
+    _TESTS = [{
+        # type=video
+        'url': 'https://www.nhl.com/video/anisimov-cleans-up-mess/t-277752844/c-43663503',
+        'md5': '0f7b9a8f986fb4b4eeeece9a56416eaf',
+        'info_dict': {
+            'id': '43663503',
+            'ext': 'mp4',
+            'title': 'Anisimov cleans up mess',
+            'description': 'md5:a02354acdfe900e940ce40706939ca63',
+            'timestamp': 1461288600,
+            'upload_date': '20160422',
+        },
+    }, {
+        # type=article
+        'url': 'https://www.nhl.com/news/dennis-wideman-suspended/c-278258934',
+        'md5': '1f39f4ea74c1394dea110699a25b366c',
+        'info_dict': {
+            'id': '40784403',
+            'ext': 'mp4',
+            'title': 'Wideman suspended by NHL',
+            'description': 'Flames defenseman Dennis Wideman was banned 20 games for violation of Rule 40 (Physical Abuse of Officials)',
+            'upload_date': '20160204',
+            'timestamp': 1454544904,
+        },
+    }]
+
+    def _real_extract(self, url):
+        tmp_id = self._match_id(url)
+        video_data = self._download_json(
+            'https://nhl.bamcontent.com/nhl/id/v1/%s/details/web-v1.json' % tmp_id,
+            tmp_id)
+        if video_data.get('type') == 'article':
+            video_data = video_data['media']
+
+        video_id = compat_str(video_data['id'])
+        title = video_data['title']
+
+        formats = []
+        for playback in video_data.get('playbacks', []):
+            playback_url = playback.get('url')
+            if not playback_url:
+                continue
+            ext = determine_ext(playback_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    playback_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id=playback.get('name', 'hls'), fatal=False))
+            else:
+                height = int_or_none(playback.get('height'))
+                formats.append({
+                    'format_id': playback.get('name', 'http' + ('-%dp' % height if height else '')),
+                    'url': playback_url,
+                    'width': int_or_none(playback.get('width')),
+                    'height': height,
+                })
+        self._sort_formats(formats, ('preference', 'width', 'height', 'tbr', 'format_id'))
+
+        thumbnails = []
+        for thumbnail_id, thumbnail_data in video_data.get('image', {}).get('cuts', {}).items():
+            thumbnail_url = thumbnail_data.get('src')
+            if not thumbnail_url:
+                continue
+            thumbnails.append({
+                'id': thumbnail_id,
+                'url': thumbnail_url,
+                'width': int_or_none(thumbnail_data.get('width')),
+                'height': int_or_none(thumbnail_data.get('height')),
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': video_data.get('description'),
+            'timestamp': parse_iso8601(video_data.get('date')),
+            'duration': parse_duration(video_data.get('duration')),
+            'thumbnails': thumbnails,
+            'formats': formats,
+        }

youtube_dl/extractor/normalboots.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from .screenwavemedia import ScreenwaveMediaIE
 
 from ..utils import (
     unified_strdate,
@@ -12,7 +13,6 @@ class NormalbootsIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?normalboots\.com/video/(?P<id>[0-9a-z-]*)/?$'
     _TEST = {
         'url': 'http://normalboots.com/video/home-alone-games-jontron/',
-        'md5': '8bf6de238915dd501105b44ef5f1e0f6',
         'info_dict': {
             'id': 'home-alone-games-jontron',
             'ext': 'mp4',
@@ -22,9 +22,10 @@ class NormalbootsIE(InfoExtractor):
             'upload_date': '20140125',
         },
         'params': {
-            # rtmp download
+            # m3u8 download
             'skip_download': True,
         },
+        'add_ie': ['ScreenwaveMedia'],
     }
 
     def _real_extract(self, url):
@@ -38,16 +39,15 @@ class NormalbootsIE(InfoExtractor):
             r'<span style="text-transform:uppercase; font-size:inherit;">[A-Za-z]+, (?P<date>.*)</span>',
             webpage, 'date', fatal=False))
 
-        player_url = self._html_search_regex(
+        screenwavemedia_url = self._html_search_regex(
-            r'<iframe\swidth="[0-9]+"\sheight="[0-9]+"\ssrc="(?P<url>[\S]+)"',
+            ScreenwaveMediaIE.EMBED_PATTERN, webpage, 'screenwave URL',
-            webpage, 'player url')
+            group='url')
-        player_page = self._download_webpage(player_url, video_id)
-        video_url = self._html_search_regex(
-            r"file:\s'(?P<file>[^']+\.mp4)'", player_page, 'file')
 
         return {
+            '_type': 'url_transparent',
             'id': video_id,
-            'url': video_url,
+            'url': screenwavemedia_url,
+            'ie_key': ScreenwaveMediaIE.ie_key(),
             'title': self._og_search_title(webpage),
             'description': self._og_search_description(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),

youtube_dl/extractor/nrk.py

@@ -4,37 +4,140 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
+from ..compat import compat_urllib_parse_unquote
-    compat_urlparse,
-    compat_urllib_parse_unquote,
-)
 from ..utils import (
-    determine_ext,
     ExtractorError,
-    float_or_none,
+    int_or_none,
+    parse_age_limit,
     parse_duration,
-    unified_strdate,
 )
 
 
-class NRKIE(InfoExtractor):
+class NRKBaseIE(InfoExtractor):
-    _VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
+    def _extract_formats(self, manifest_url, video_id, fatal=True):
+        formats = []
+        formats.extend(self._extract_f4m_formats(
+            manifest_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81',
+            video_id, f4m_id='hds', fatal=fatal))
+        formats.extend(self._extract_m3u8_formats(manifest_url.replace(
+            'akamaihd.net/z/', 'akamaihd.net/i/').replace('/manifest.f4m', '/master.m3u8'),
+            video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=fatal))
+        return formats
 
-    _TESTS = [
+    def _real_extract(self, url):
-        {
+        video_id = self._match_id(url)
+
+        data = self._download_json(
+            'http://%s/mediaelement/%s' % (self._API_HOST, video_id),
+            video_id, 'Downloading mediaelement JSON')
+
+        title = data.get('fullTitle') or data.get('mainTitle') or data['title']
+        video_id = data.get('id') or video_id
+
+        entries = []
+
+        media_assets = data.get('mediaAssets')
+        if media_assets and isinstance(media_assets, list):
+            def video_id_and_title(idx):
+                return ((video_id, title) if len(media_assets) == 1
+                        else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
+            for num, asset in enumerate(media_assets, 1):
+                asset_url = asset.get('url')
+                if not asset_url:
+                    continue
+                formats = self._extract_formats(asset_url, video_id, fatal=False)
+                if not formats:
+                    continue
+                self._sort_formats(formats)
+                entry_id, entry_title = video_id_and_title(num)
+                duration = parse_duration(asset.get('duration'))
+                subtitles = {}
+                for subtitle in ('webVtt', 'timedText'):
+                    subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
+                    if subtitle_url:
+                        subtitles.setdefault('no', []).append({'url': subtitle_url})
+                entries.append({
+                    'id': asset.get('carrierId') or entry_id,
+                    'title': entry_title,
+                    'duration': duration,
+                    'subtitles': subtitles,
+                    'formats': formats,
+                })
+
+        if not entries:
+            media_url = data.get('mediaUrl')
+            if media_url:
+                formats = self._extract_formats(media_url, video_id)
+                self._sort_formats(formats)
+                duration = parse_duration(data.get('duration'))
+                entries = [{
+                    'id': video_id,
+                    'title': title,
+                    'duration': duration,
+                    'formats': formats,
+                }]
+
+        if not entries:
+            if data.get('usageRights', {}).get('isGeoBlocked'):
+                raise ExtractorError(
+                    'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
+                    expected=True)
+
+        conviva = data.get('convivaStatistics') or {}
+        series = conviva.get('seriesName') or data.get('seriesTitle')
+        episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
+
+        thumbnails = None
+        images = data.get('images')
+        if images and isinstance(images, dict):
+            web_images = images.get('webImages')
+            if isinstance(web_images, list):
+                thumbnails = [{
+                    'url': image['imageUrl'],
+                    'width': int_or_none(image.get('width')),
+                    'height': int_or_none(image.get('height')),
+                } for image in web_images if image.get('imageUrl')]
+
+        description = data.get('description')
+
+        common_info = {
+            'description': description,
+            'series': series,
+            'episode': episode,
+            'age_limit': parse_age_limit(data.get('legalAge')),
+            'thumbnails': thumbnails,
+        }
+
+        vcodec = 'none' if data.get('mediaType') == 'Audio' else None
+
+        # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged
+
+        for entry in entries:
+            entry.update(common_info)
+            for f in entry['formats']:
+                f['vcodec'] = vcodec
+
+        return self.playlist_result(entries, video_id, title, description)
+
+
+class NRKIE(NRKBaseIE):
+    _VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
+    _API_HOST = 'v8.psapi.nrk.no'
+    _TESTS = [{
+        # video
         'url': 'http://www.nrk.no/video/PS*150533',
-            'md5': 'bccd850baebefe23b56d708a113229c2',
+        'md5': '2f7f6eeb2aacdd99885f355428715cfa',
         'info_dict': {
             'id': '150533',
-                'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Dompap og andre fugler i Piip-Show',
             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
             'duration': 263,
         }
-        },
+    }, {
-        {
+        # audio
         'url': 'http://www.nrk.no/video/PS*154915',
-            'md5': '0b1493ba1aae7d9579a5ad5531bc395a',
+        # MD5 is unstable
         'info_dict': {
             'id': '154915',
             'ext': 'flv',
@@ -42,52 +145,75 @@ class NRKIE(InfoExtractor):
             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
             'duration': 20,
         }
-        },
-    ]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        data = self._download_json(
-            'http://v8.psapi.nrk.no/mediaelement/%s' % video_id,
-            video_id, 'Downloading media JSON')
-
-        media_url = data.get('mediaUrl')
-
-        if not media_url:
-            if data['usageRights']['isGeoBlocked']:
-                raise ExtractorError(
-                    'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
-                    expected=True)
-
-        if determine_ext(media_url) == 'f4m':
-            formats = self._extract_f4m_formats(
-                media_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', video_id, f4m_id='hds')
-            self._sort_formats(formats)
-        else:
-            formats = [{
-                'url': media_url,
-                'ext': 'flv',
     }]
 
-        duration = parse_duration(data.get('duration'))
 
-        images = data.get('images')
+class NRKTVIE(NRKBaseIE):
-        if images:
+    IE_DESC = 'NRK TV and NRK Radio'
-            thumbnails = images['webImages']
+    _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
-            thumbnails.sort(key=lambda image: image['pixelWidth'])
+    _API_HOST = 'psapi-we.nrk.no'
-            thumbnail = thumbnails[-1]['imageUrl']
-        else:
-            thumbnail = None
 
-        return {
+    _TESTS = [{
-            'id': video_id,
+        'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
-            'title': data['title'],
+        'md5': '4e9ca6629f09e588ed240fb11619922a',
-            'description': data['description'],
+        'info_dict': {
-            'duration': duration,
+            'id': 'MUHH48000314AA',
-            'thumbnail': thumbnail,
+            'ext': 'mp4',
-            'formats': formats,
+            'title': '20 spørsmål 23.05.2014',
-        }
+            'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
+            'duration': 1741.52,
+        },
+    }, {
+        'url': 'https://tv.nrk.no/program/mdfp15000514',
+        'md5': '43d0be26663d380603a9cf0c24366531',
+        'info_dict': {
+            'id': 'MDFP15000514CA',
+            'ext': 'mp4',
+            'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
+            'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
+            'duration': 4605.08,
+        },
+    }, {
+        # single playlist video
+        'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
+        'md5': 'adbd1dbd813edaf532b0a253780719c2',
+        'info_dict': {
+            'id': 'MSPO40010515-part2',
+            'ext': 'flv',
+            'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
+            'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
+        },
+        'skip': 'Only works from Norway',
+    }, {
+        'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
+        'playlist': [{
+            'md5': '9480285eff92d64f06e02a5367970a7a',
+            'info_dict': {
+                'id': 'MSPO40010515-part1',
+                'ext': 'flv',
+                'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
+                'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
+            },
+        }, {
+            'md5': 'adbd1dbd813edaf532b0a253780719c2',
+            'info_dict': {
+                'id': 'MSPO40010515-part2',
+                'ext': 'flv',
+                'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
+                'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
+            },
+        }],
+        'info_dict': {
+            'id': 'MSPO40010515',
+            'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
+            'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
+            'duration': 6947.52,
+        },
+        'skip': 'Only works from Norway',
+    }, {
+        'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
+        'only_matching': True,
+    }]
 
 
 class NRKPlaylistIE(InfoExtractor):
@@ -159,179 +285,3 @@ class NRKSkoleIE(InfoExtractor):
 
         nrk_id = self._search_regex(r'data-nrk-id=["\'](\d+)', webpage, 'nrk id')
         return self.url_result('nrk:%s' % nrk_id)
-
-
-class NRKTVIE(InfoExtractor):
-    IE_DESC = 'NRK TV and NRK Radio'
-    _VALID_URL = r'(?P<baseurl>https?://(?:tv|radio)\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
-
-    _TESTS = [
-        {
-            'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
-            'info_dict': {
-                'id': 'MUHH48000314',
-                'ext': 'mp4',
-                'title': '20 spørsmål',
-                'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
-                'upload_date': '20140523',
-                'duration': 1741.52,
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        },
-        {
-            'url': 'https://tv.nrk.no/program/mdfp15000514',
-            'info_dict': {
-                'id': 'mdfp15000514',
-                'ext': 'mp4',
-                'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting',
-                'description': 'md5:654c12511f035aed1e42bdf5db3b206a',
-                'upload_date': '20140524',
-                'duration': 4605.08,
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        },
-        {
-            # single playlist video
-            'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
-            'md5': 'adbd1dbd813edaf532b0a253780719c2',
-            'info_dict': {
-                'id': 'MSPO40010515-part2',
-                'ext': 'flv',
-                'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
-                'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
-                'upload_date': '20150106',
-            },
-            'skip': 'Only works from Norway',
-        },
-        {
-            'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
-            'playlist': [
-                {
-                    'md5': '9480285eff92d64f06e02a5367970a7a',
-                    'info_dict': {
-                        'id': 'MSPO40010515-part1',
-                        'ext': 'flv',
-                        'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
-                        'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
-                        'upload_date': '20150106',
-                    },
-                },
-                {
-                    'md5': 'adbd1dbd813edaf532b0a253780719c2',
-                    'info_dict': {
-                        'id': 'MSPO40010515-part2',
-                        'ext': 'flv',
-                        'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
-                        'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
-                        'upload_date': '20150106',
-                    },
-                },
-            ],
-            'info_dict': {
-                'id': 'MSPO40010515',
-                'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
-                'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
-                'upload_date': '20150106',
-                'duration': 6947.5199999999995,
-            },
-            'skip': 'Only works from Norway',
-        },
-        {
-            'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
-            'only_matching': True,
-        }
-    ]
-
-    def _extract_f4m(self, manifest_url, video_id):
-        return self._extract_f4m_formats(
-            manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id, f4m_id='hds')
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        part_id = mobj.group('part_id')
-        base_url = mobj.group('baseurl')
-
-        webpage = self._download_webpage(url, video_id)
-
-        title = self._html_search_meta(
-            'title', webpage, 'title')
-        description = self._html_search_meta(
-            'description', webpage, 'description')
-
-        thumbnail = self._html_search_regex(
-            r'data-posterimage="([^"]+)"',
-            webpage, 'thumbnail', fatal=False)
-        upload_date = unified_strdate(self._html_search_meta(
-            'rightsfrom', webpage, 'upload date', fatal=False))
-        duration = float_or_none(self._html_search_regex(
-            r'data-duration="([^"]+)"',
-            webpage, 'duration', fatal=False))
-
-        # playlist
-        parts = re.findall(
-            r'<a href="#del=(\d+)"[^>]+data-argument="([^"]+)">([^<]+)</a>', webpage)
-        if parts:
-            entries = []
-            for current_part_id, stream_url, part_title in parts:
-                if part_id and current_part_id != part_id:
-                    continue
-                video_part_id = '%s-part%s' % (video_id, current_part_id)
-                formats = self._extract_f4m(stream_url, video_part_id)
-                entries.append({
-                    'id': video_part_id,
-                    'title': part_title,
-                    'description': description,
-                    'thumbnail': thumbnail,
-                    'upload_date': upload_date,
-                    'formats': formats,
-                })
-            if part_id:
-                if entries:
-                    return entries[0]
-            else:
-                playlist = self.playlist_result(entries, video_id, title, description)
-                playlist.update({
-                    'thumbnail': thumbnail,
-                    'upload_date': upload_date,
-                    'duration': duration,
-                })
-                return playlist
-
-        formats = []
-
-        f4m_url = re.search(r'data-media="([^"]+)"', webpage)
-        if f4m_url:
-            formats.extend(self._extract_f4m(f4m_url.group(1), video_id))
-
-        m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
-        if m3u8_url:
-            formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4', m3u8_id='hls'))
-        self._sort_formats(formats)
-
-        subtitles_url = self._html_search_regex(
-            r'data-subtitlesurl\s*=\s*(["\'])(?P<url>.+?)\1',
-            webpage, 'subtitle URL', default=None, group='url')
-        subtitles = {}
-        if subtitles_url:
-            subtitles['no'] = [{
-                'ext': 'ttml',
-                'url': compat_urlparse.urljoin(base_url, subtitles_url),
-            }]
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'upload_date': upload_date,
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,
-        }