release 2017.04.26

In some non-standard implementations, the server may respond AUTH_USER_PASS
even if's not listed in available authentication methods. (it should
respond AUTH_NO_ACCEPTABLE per standards)

.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 *2017.02.04.1*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.02.04.1**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.04.26*. 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 Issues with outdated version will be rejected.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 **2017.04.26**
 
 ### 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 2017.02.04.1
+[debug] youtube-dl version 2017.04.26
 [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

@@ -35,8 +35,8 @@ updates_key.pem
 *.mkv
 *.swf
 *.part
+*.ytdl
 *.swp
-test/testdata
 test/local_parameters.json
 .tox
 youtube-dl.zsh

.travis.yml

@@ -6,8 +6,12 @@ python:
   - "3.3"
   - "3.4"
   - "3.5"
+  - "3.6"
 sudo: false
-script: nosetests test --verbose
+env:
+  - YTDL_TEST_SET=core
+  - YTDL_TEST_SET=download
+script: ./devscripts/run_tests.sh
 notifications:
   email:
     - filippo.valsorda@gmail.com

AUTHORS

@@ -201,3 +201,14 @@ Stephen Chen
 Fabian Stahl
 Bagira
 Odd Stråbø
+Philip Herzog
+Thomas Christlieb
+Marek Rusinowski
+Tobias Gruetzmacher
+Olivier Bilodeau
+Lars Vierbergen
+Juanjo Benages
+Xiao Di Guan
+Thomas Winant
+Daniel Twardowski
+Jeremie Jarosh

ChangeLog

@@ -1,3 +1,557 @@
+version 2017.04.26
+
+Core
+* Introduce --keep-fragments for keeping fragments of fragmented download
+  on disk after download is finished
+* [YoutubeDL] Fix output template for missing timestamp (#12796)
+* [socks] Handle cases where credentials are required but missing
+* [extractor/common] Improve HLS extraction (#12211)
+    - Extract m3u8 parsing to separate method
+    - Improve rendition groups extraction
+    - Build stream name according stream GROUP-ID
+    - Ignore reference to AUDIO group without URI when stream has no CODECS
+    - Use float for scaled tbr in _parse_m3u8_formats
+* [utils] Add support for TTML styles in dfxp2srt
+* [downloader/hls] No need to download keys for fragments that have been
+  already downloaded
+* [downloader/fragment] Improve fragment downloading
+    - Resume immediately
+    - Don't concatenate fragments and decrypt them on every resume
+    - Optimize disk storage usage, don't store intermediate fragments on disk
+    - Store bookkeeping download state file
++ [extractor/common] Add support for multiple getters in try_get
++ [extractor/common] Add support for video of WebPage context in _json_ld
+  (#12778)
++ [extractor/common] Relax JWPlayer regular expression and remove
+  duplicate URLs (#12768)
+
+Extractors
+* [iqiyi] Fix extraction of Yule videos
+* [vidio] Improve extraction and sort formats
++ [brightcove] Match only video elements with data-video-id attribute
+* [iqiyi] Fix playlist detection (#12504)
+- [azubu] Remove extractor (#12813)
+* [porn91] Fix extraction (#12814)
+* [vidzi] Fix extraction (#12793)
++ [amp] Extract error message (#12795)
++ [xfileshare] Add support for gorillavid.com and daclips.com (#12776)
+* [instagram] Fix extraction (#12777)
++ [generic] Support Brightcove videos in <iframe> (#12482)
++ [brightcove] Support URLs with bcpid instead of playerID (#12482)
+* [brightcove] Fix _extract_url (#12782)
++ [odnoklassniki] Extract HLS formats
+
+
+version 2017.04.17
+
+Extractors
+* [limelight] Improve extraction LimelightEmbeddedPlayerFlash media embeds and
+  add support for channel and channelList embeds
+* [generic] Extract multiple Limelight embeds (#12761)
++ [itv] Extract series metadata
+* [itv] Fix RTMP formats downloading (#12759)
+* [itv] Use native HLS downloader by default
++ [go90] Extract subtitles (#12752)
++ [go90] Extract series metadata (#12752)
+
+
+version 2017.04.16
+
+Core
+* [YoutubeDL] Apply expand_path after output template substitution
++ [YoutubeDL] Propagate overridden meta fields to extraction results of type
+  url (#11163)
+
+Extractors
++ [generic] Extract RSS entries as url_transparent (#11163)
++ [streamango] Add support for streamango.com (#12643)
++ [wsj:article] Add support for articles (#12558)
+* [brightcove] Relax video tag embeds extraction and validate ambiguous embeds'
+  URLs (#9163, #12005, #12178, #12480)
++ [udemy] Add support for react rendition (#12744)
+
+
+version 2017.04.15
+
+Extractors
+* [youku] Fix fileid extraction (#12741, #12743)
+
+
+version 2017.04.14
+
+Core
++ [downloader/hls] Add basic support for EXT-X-BYTERANGE tag (#10955)
++ [adobepass] Improve Comcast and Verison login code (#10803)
++ [adobepass] Add support for Verizon (#10803)
+
+Extractors
++ [aenetworks] Add support for specials (#12723)
++ [hbo] Extract HLS formats
++ [go90] Add support for go90.com (#10127)
++ [tv2hu] Add support for tv2.hu (#10509)
++ [generic] Exclude URLs with xml ext from valid video URLs (#10768, #11654)
+* [youtube] Improve HLS formats extraction
+* [afreecatv] Fix extraction for videos with different key layout (#12718)
+- [youtube] Remove explicit preference for audio-only and video-only formats in
+  order not to break sorting when new formats appear
+* [canalplus] Bypass geo restriction
+
+
+version 2017.04.11
+
+Extractors
+* [afreecatv] Fix extraction (#12706)
++ [generic] Add support for <object> YouTube embeds (#12637)
+* [bbccouk] Treat bitrate as audio+video bitrate in media selector
++ [bbccouk] Skip unrecognized formats in media selector (#12701)
++ [bbccouk] Add support for https protocol in media selector (#12701)
+* [curiositystream] Fix extraction (#12638)
+* [adn] Update subtitle decryption key
+* [chaturbate] Fix extraction (#12665, #12688, #12690)
+
+
+version 2017.04.09
+
+Extractors
++ [medici] Add support for medici.tv (#3406)
++ [rbmaradio] Add support for redbullradio.com URLs (#12687)
++ [npo:live] Add support for default URL (#12555)
+* [mixcloud:playlist] Fix title, description and view count extraction (#12582)
++ [thesun] Add suport for thesun.co.uk (#11298, #12674)
++ [ceskateleveize:porady] Add support for porady (#7411, #12645)
+* [ceskateleveize] Improve extraction and remove URL replacement hacks
++ [kaltura] Add support for iframe embeds (#12679)
+* [airmozilla] Fix extraction (#12670)
+* [wshh] Extract html5 entries and delegate to generic extractor (12676)
++ [raiplay] Extract subtitles
++ [xfileshare] Add support for vidlo.us (#12660)
++ [xfileshare] Add support for vidbom.com (#12661)
++ [aenetworks] Add more video URL regular expressions (#12657)
++ [odnoklassniki] Fix format sorting for 1080p quality
++ [rtl2] Add support for you.rtl2.de (#10257)
++ [vshare] Add support for vshare.io (#12278)
+
+
+version 2017.04.03
+
+Core
++ [extractor/common] Add censorship check for TransTelekom ISP
+* [extractor/common] Move censorship checks to a separate method
+
+Extractors
++ [discoveryvr] Add support for discoveryvr.com (#12578)
++ [tv5mondeplus] Add support for tv5mondeplus.com (#11386)
++ [periscope] Add support for pscp.tv URLs (#12618, #12625)
+
+
+version 2017.04.02
+
+Core
+* [YoutubeDL] Return early when extraction of url_transparent fails
+
+Extractors
+* [rai] Fix and improve extraction (#11790)
++ [vrv] Add support for series pages
+* [limelight] Improve extraction for audio only formats
+* [funimation] Fix extraction (#10696, #11773)
++ [xfileshare] Add support for vidabc.com (#12589)
++ [xfileshare] Improve extraction and extract hls formats
++ [crunchyroll] Pass geo verifcation proxy
++ [cwtv] Extract ISM formats
++ [tvplay] Bypass geo restriction
++ [vrv] Add support for vrv.co
++ [packtpub] Add support for packtpub.com (#12610)
++ [generic] Pass base_url to _parse_jwplayer_data
++ [adn] Add support for animedigitalnetwork.fr (#4866)
++ [allocine] Extract more metadata
+* [allocine] Fix extraction (#12592)
+* [openload] Fix extraction
+
+
+version 2017.03.26
+
+Core
+* Don't raise an error if JWPlayer config data is not a Javascript object
+  literal. _find_jwplayer_data now returns a dict rather than an str. (#12307)
+* Expand environment variables for options representing paths (#12556)
++ [utils] Introduce expand_path
+* [downloader/hls] Delegate downloading to ffmpeg immediately for live streams
+
+Extractors
+* [afreecatv] Fix extraction (#12179)
++ [atvat] Add support for atv.at (#5325)
++ [fox] Add metadata extraction (#12391)
++ [atresplayer] Extract DASH formats
++ [atresplayer] Extract HD manifest (#12548)
+* [atresplayer] Fix login error detection (#12548)
+* [franceculture] Fix extraction (#12547)
+* [youtube] Improve URL regular expression (#12538)
+* [generic] Do not follow redirects to the same URL
+
+
+version 2017.03.24
+
+Extractors
+- [9c9media] Remove mp4 URL extraction request
++ [bellmedia] Add support for etalk.ca and space.ca (#12447)
+* [channel9] Fix extraction (#11323)
+* [cloudy] Fix extraction (#12525)
++ [hbo] Add support for free episode URLs and new formats extraction (#12519)
+* [condenast] Fix extraction and style (#12526)
+* [viu] Relax URL regular expression (#12529)
+
+
+version 2017.03.22
+
+Extractors
+- [pluralsight] Omit module title from video title (#12506)
+* [pornhub] Decode obfuscated video URL (#12470, #12515)
+* [senateisvp] Allow https URL scheme for embeds (#12512)
+
+
+version 2017.03.20
+
+Core
++ [YoutubeDL] Allow multiple input URLs to be used with stdout (-) as
+  output template
++ [adobepass] Detect and output error on authz token extraction (#12472)
+
+Extractors
++ [bostonglobe] Add extractor for bostonglobe.com (#12099)
++ [toongoggles] Add support for toongoggles.com (#12171)
++ [medialaan] Add support for Medialaan sites (#9974, #11912)
++ [discoverynetworks] Add support for more domains and bypass geo restiction
+* [openload] Fix extraction (#10408)
+
+
+version 2017.03.16
+
+Core
++ [postprocessor/ffmpeg] Add support for flac
++ [extractor/common] Extract SMIL formats from jwplayer
+
+Extractors
++ [generic] Add forgotten return for jwplayer formats
+* [redbulltv] Improve extraction
+
+
+version 2017.03.15
+
+Core
+* Fix missing subtitles if --add-metadata is used (#12423)
+
+Extractors
+* [facebook] Make title optional (#12443)
++ [mitele] Add support for ooyala videos (#12430)
+* [openload] Fix extraction (#12435, #12446)
+* [streamable] Update API URL (#12433)
++ [crunchyroll] Extract season name (#12428)
+* [discoverygo] Bypass geo restriction
++ [discoverygo:playlist] Add support for playlists (#12424)
+
+
+version 2017.03.10
+
+Extractors
+* [generic] Make title optional for jwplayer embeds (#12410)
+* [wdr:maus] Fix extraction (#12373)
+* [prosiebensat1] Improve title extraction (#12318, #12327)
+* [dplayit] Separate and rewrite extractor and bypass geo restriction (#12393)
+* [miomio] Fix extraction (#12291, #12388, #12402)
+* [telequebec] Fix description extraction (#12399)
+* [openload] Fix extraction (#12357)
+* [brightcove:legacy] Relax videoPlayer validation check (#12381)
+
+
+version 2017.03.07
+
+Core
+* Metadata are now added after conversion (#5594)
+
+Extractors
+* [soundcloud] Update client id (#12376)
+* [openload] Fix extraction (#10408, #12357)
+
+
+version 2017.03.06
+
+Core
++ [utils] Process bytestrings in urljoin (#12369)
+* [extractor/common] Improve height extraction and extract bitrate
+* [extractor/common] Move jwplayer formats extraction in separate method
++ [external:ffmpeg] Limit test download size to 10KiB (#12362)
+
+Extractors
++ [drtv] Add geo countries to GeoRestrictedError
++ [drtv:live] Bypass geo restriction
++ [tunepk] Add extractor (#12197, #12243)
+
+
+version 2017.03.05
+
+Extractors
++ [twitch] Add basic support for two-factor authentication (#11974)
++ [vier] Add support for vijf.be (#12304)
++ [redbulltv] Add support for redbull.tv (#3919, #11948)
+* [douyutv] Switch to the PC API to escape the 5-min limitation (#12316)
++ [generic] Add support for rutube embeds
++ [rutube] Relax URL regular expression
++ [vrak] Add support for vrak.tv (#11452)
++ [brightcove:new] Add ability to smuggle geo_countries into URL
++ [brightcove:new] Raise GeoRestrictedError
+* [go] Relax URL regular expression (#12341)
+* [24video] Use original host for requests (#12339)
+* [ruutu] Disable DASH formats (#12322)
+
+
+version 2017.03.02
+
+Core
++ [adobepass] Add support for Charter Spectrum (#11465)
+* [YoutubeDL] Don't sanitize identifiers in output template (#12317)
+
+Extractors
+* [facebook] Fix extraction (#12323, #12330)
+* [youtube] Mark errors about rental videos as expected (#12324)
++ [npo] Add support for audio
+* [npo] Adapt to app.php API (#12311, #12320)
+
+
+version 2017.02.28
+
+Core
++ [utils] Add bytes_to_long and long_to_bytes
++ [utils] Add pkcs1pad
++ [aes] Add aes_cbc_encrypt
+
+Extractors
++ [azmedien:showplaylist] Add support for show playlists (#12160)
++ [youtube:playlist] Recognize another playlist pattern (#11928, #12286)
++ [daisuki] Add support for daisuki.net (#2486, #3186, #4738, #6175, #7776,
+  #10060)
+* [douyu] Fix extraction (#12301)
+
+
+version 2017.02.27
+
+Core
+* [downloader/common] Limit displaying 2 digits after decimal point in sleep
+  interval message (#12183)
++ [extractor/common] Add preference to _parse_html5_media_entries
+
+Extractors
++ [npo] Add support for zapp.nl
++ [npo] Add support for hetklokhuis.nl (#12293)
+- [scivee] Remove extractor (#9315)
++ [cda] Decode download URL (#12255)
++ [crunchyroll] Improve uploader extraction (#12267)
++ [youtube] Raise GeoRestrictedError
++ [dailymotion] Raise GeoRestrictedError
++ [mdr] Recognize more URL patterns (#12169)
++ [tvigle] Raise GeoRestrictedError
+* [vevo] Fix extraction for videos with the new streams/streamsV3 format
+  (#11719)
++ [freshlive] Add support for freshlive.tv (#12175)
++ [xhamster] Capture and output videoClosed error (#12263)
++ [etonline] Add support for etonline.com (#12236)
++ [njpwworld] Add support for njpwworld.com (#11561)
+* [amcnetworks] Relax URL regular expression (#12127)
+
+
+version 2017.02.24.1
+
+Extractors
+* [noco] Modernize
+* [noco] Switch login URL to https (#12246)
++ [thescene] Extract more metadata
+* [thescene] Fix extraction (#12235)
++ [tubitv] Use geo bypass mechanism
+* [openload] Fix extraction (#10408)
++ [ivi] Raise GeoRestrictedError
+
+
+version 2017.02.24
+
+Core
+* [options] Hide deprecated options from --help
+* [options] Deprecate --autonumber-size
++ [YoutubeDL] Add support for string formatting operations in output template
+  (#5185, #5748, #6841, #9929, #9966 #9978, #12189)
+
+Extractors
++ [lynda:course] Add webpage extraction fallback (#12238)
+* [go] Sign all uplynk URLs and use geo bypass only for free videos
+  (#12087, #12210)
++ [skylinewebcams] Add support for skylinewebcams.com (#12221)
++ [instagram] Add support for multi video posts (#12226)
++ [crunchyroll] Extract playlist entries ids
+* [mgtv] Fix extraction
++ [sohu] Raise GeoRestrictedError
++ [leeco] Raise GeoRestrictedError and use geo bypass mechanism
+
+
+version 2017.02.22
+
+Extractors
+* [crunchyroll] Fix descriptions with double quotes (#12124)
+* [dailymotion] Make comment count optional (#12209)
++ [vidzi] Add support for vidzi.cc (#12213)
++ [24video] Add support for 24video.tube (#12217)
++ [crackle] Use geo bypass mechanism
++ [viewster] Use geo verification headers
++ [tfo] Improve geo restriction detection and use geo bypass mechanism
++ [telequebec] Use geo bypass mechanism
++ [limelight] Extract PlaylistService errors and improve geo restriction
+  detection
+
+
+version 2017.02.21
+
+Core
+* [extractor/common] Allow calling _initialize_geo_bypass from extractors
+  (#11970)
++ [adobepass] Add support for Time Warner Cable (#12191)
++ [travis] Run tests in parallel
++ [downloader/ism] Honor HTTP headers when downloading fragments
++ [downloader/dash] Honor HTTP headers when downloading fragments
++ [utils] Add GeoUtils class for working with geo tools and GeoUtils.random_ipv4
++ Add option --geo-bypass-country for explicit geo bypass on behalf of
+  specified country
++ Add options to control geo bypass mechanism --geo-bypass and --no-geo-bypass
++ Add experimental geo restriction bypass mechanism based on faking
+  X-Forwarded-For HTTP header
++ [utils] Introduce GeoRestrictedError for geo restricted videos
++ [utils] Introduce YoutubeDLError base class for all youtube-dl exceptions
+
+Extractors
++ [ninecninemedia] Use geo bypass mechanism
+* [spankbang] Make uploader optional (#12193)
++ [iprima] Improve geo restriction detection and disable geo bypass
+* [iprima] Modernize
+* [commonmistakes] Disable UnicodeBOM extractor test for python 3.2
++ [prosiebensat1] Throw ExtractionError on unsupported page type (#12180)
+* [nrk] Update _API_HOST and relax _VALID_URL
++ [tv4] Bypass geo restriction and improve detection
+* [tv4] Switch to hls3 protocol (#12177)
++ [viki] Improve geo restriction detection
++ [vgtv] Improve geo restriction detection
++ [srgssr] Improve geo restriction detection
++ [vbox7] Improve geo restriction detection and use geo bypass mechanism
++ [svt] Improve geo restriction detection and use geo bypass mechanism
++ [pbs] Improve geo restriction detection and use geo bypass mechanism
++ [ondemandkorea] Improve geo restriction detection and use geo bypass mechanism
++ [nrk] Improve geo restriction detection and use geo bypass mechanism
++ [itv] Improve geo restriction detection and use geo bypass mechanism
++ [go] Improve geo restriction detection and use geo bypass mechanism
++ [dramafever] Improve geo restriction detection and use geo bypass mechanism
+* [brightcove:legacy] Restrict videoPlayer value (#12040)
++ [tvn24] Add support for tvn24.pl and tvn24bis.pl (#11679)
++ [thisav] Add support for HTML5 media (#11771)
+* [metacafe] Bypass family filter (#10371)
+* [viceland] Improve info extraction
+
+
+version 2017.02.17
+
+Extractors
+* [heise] Improve extraction (#9725)
+* [ellentv] Improve (#11653)
+* [openload] Fix extraction (#10408, #12002)
++ [theplatform] Recognize URLs with whitespaces (#12044)
+* [einthusan] Relax URL regular expression (#12141, #12159)
++ [generic] Support complex JWPlayer embedded videos (#12030)
+* [elpais] Improve extraction (#12139)
+
+
+version 2017.02.16
+
+Core
++ [utils] Add support for quoted string literals in --match-filter (#8050,
+  #12142, #12144)
+
+Extractors
+* [ceskatelevize] Lower priority for audio description sources (#12119)
+* [amcnetworks] Fix extraction (#12127)
+* [pinkbike] Fix uploader extraction (#12054)
++ [onetpl] Add support for businessinsider.com.pl and plejada.pl
++ [onetpl] Add support for onet.pl (#10507)
++ [onetmvp] Add shortcut extractor
++ [vodpl] Add support for vod.pl (#12122)
++ [pornhub] Extract video URL from tv platform site (#12007, #12129)
++ [ceskatelevize] Extract DASH formats (#12119, #12133)
+
+
+version 2017.02.14
+
+Core
+* TypeError is fixed with Python 2.7.13 on Windows (#11540, #12085)
+
+Extractor
+* [zdf] Fix extraction (#12117)
+* [xtube] Fix extraction for both kinds of video id (#12088)
+* [xtube] Improve title extraction (#12088)
++ [lemonde] Fallback delegate extraction to generic extractor (#12115, #12116)
+* [bellmedia] Allow video id longer than 6 characters (#12114)
++ [limelight] Add support for referer protected videos
+* [disney] Improve extraction (#4975, #11000, #11882, #11936)
+* [hotstar] Improve extraction (#12096)
+* [einthusan] Fix extraction (#11416)
++ [aenetworks] Add support for lifetimemovieclub.com (#12097)
+* [youtube] Fix parsing codecs (#12091)
+
+
+version 2017.02.11
+
+Core
++ [utils] Introduce get_elements_by_class and get_elements_by_attribute
+  utility functions
++ [extractor/common] Skip m3u8 manifests protected with Adobe Flash Access
+
+Extractor
+* [pluralsight:course] Fix extraction (#12075)
++ [bbc] Extract m3u8 formats with 320k audio
+* [facebook] Relax video id matching (#11017, #12055, #12056)
++ [corus] Add support for Corus Entertainment sites (#12060, #9164)
++ [pluralsight] Detect blocked account error message (#12070)
++ [bloomberg] Add another video id pattern (#12062)
+* [extractor/commonmistakes] Restrict URL regular expression (#12050)
++ [tvplayer] Add support for tvplayer.com
+
+
+version 2017.02.10
+
+Extractors
+* [xtube] Fix extraction (#12023)
+* [pornhub] Fix extraction (#12007, #12018)
+* [facebook] Improve JS data regular expression (#12042)
+* [kaltura] Improve embed partner id extraction (#12041)
++ [sprout] Add support for sproutonline.com
+* [6play] Improve extraction
++ [scrippsnetworks:watch] Add support for Scripps Networks sites (#10765)
++ [go] Add support for Adobe Pass authentication (#11468, #10831)
+* [6play] Fix extraction (#12011)
++ [nbc] Add support for Adobe Pass authentication (#12006)
+
+
+version 2017.02.07
+
+Core
+* [extractor/common] Fix audio only with audio group in m3u8 (#11995)
++ [downloader/fragment] Respect --no-part
+* [extractor/common] Speed-up HTML5 media entries extraction (#11979)
+
+Extractors
+* [pornhub] Fix extraction (#11997)
++ [canalplus] Add support for cstar.fr (#11990)
++ [extractor/generic] Improve RTMP support (#11993)
++ [gaskrank] Add support for gaskrank.tv (#11685)
+* [bandcamp] Fix extraction for incomplete albums (#11727)
+* [iwara] Fix extraction (#11781)
+* [googledrive] Fix extraction on Python 3.6
++ [videopress] Add support for videopress.com
++ [afreecatv] Extract RTMP formats
+
+
 version 2017.02.04.1
 
 Extractors

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 *.m4a *.flv *.mp3 *.avi *.mkv *.webm *.3gp *.wav *.ape *.swf *.jpg *.png 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* *.ytdl *.info.json *.mp4 *.m4a *.flv *.mp3 *.avi *.mkv *.webm *.3gp *.wav *.ape *.swf *.jpg *.png CONTRIBUTING.md.tmp ISSUE_TEMPLATE.md.tmp youtube-dl youtube-dl.exe
 	find . -name "*.pyc" -delete
 	find . -name "*.class" -delete
 

README.md

@@ -99,11 +99,21 @@ Alternatively, refer to the [developer instructions](#developer-instructions) fo
     --source-address IP              Client-side IP address to bind to
     -4, --force-ipv4                 Make all connections via IPv4
     -6, --force-ipv6                 Make all connections via IPv6
+
+## Geo Restriction:
     --geo-verification-proxy URL     Use this proxy to verify the IP address for
                                      some geo-restricted sites. The default
                                      proxy specified by --proxy (or none, if the
                                      options is not present) is used for the
                                      actual downloading.
+    --geo-bypass                     Bypass geographic restriction via faking
+                                     X-Forwarded-For HTTP header (experimental)
+    --no-geo-bypass                  Do not bypass geographic restriction via
+                                     faking X-Forwarded-For HTTP header
+                                     (experimental)
+    --geo-bypass-country CODE        Force bypass geographic restriction with
+                                     explicitly provided two-letter ISO 3166-2
+                                     country code (experimental)
 
 ## Video Selection:
     --playlist-start NUMBER          Playlist video to start at (default is 1)
@@ -137,20 +147,22 @@ Alternatively, refer to the [developer instructions](#developer-instructions) fo
     --match-filter FILTER            Generic video filter. Specify any key (see
                                      help for -o for a list of available keys)
                                      to match if the key is present, !key to
-                                     check if the key is not present,key >
+                                     check if the key is not present, key >
                                      NUMBER (like "comment_count > 12", also
                                      works with >=, <, <=, !=, =) to compare
-                                     against a number, and & to require multiple
-                                     matches. Values which are not known are
-                                     excluded unless you put a question mark (?)
-                                     after the operator.For example, to only
-                                     match videos that have been liked more than
-                                     100 times and disliked less than 50 times
-                                     (or the dislike functionality is not
-                                     available at the given service), but who
-                                     also have a description, use --match-filter
-                                     "like_count > 100 & dislike_count <? 50 &
-                                     description" .
+                                     against a number, key = 'LITERAL' (like
+                                     "uploader = 'Mike Smith'", also works with
+                                     !=) to match against a string literal and &
+                                     to require multiple matches. Values which
+                                     are not known are excluded unless you put a
+                                     question mark (?) after the operator. For
+                                     example, to only match videos that have
+                                     been liked more than 100 times and disliked
+                                     less than 50 times (or the dislike
+                                     functionality is not available at the given
+                                     service), but who also have a description,
+                                     use --match-filter "like_count > 100 &
+                                     dislike_count <? 50 & description" .
     --no-playlist                    Download only the video, if the URL refers
                                      to a video and a playlist.
     --yes-playlist                   Download the playlist, if the URL refers to
@@ -169,12 +181,15 @@ Alternatively, refer to the [developer instructions](#developer-instructions) fo
     -R, --retries RETRIES            Number of retries (default is 10), or
                                      "infinite".
     --fragment-retries RETRIES       Number of retries for a fragment (default
-                                     is 10), or "infinite" (DASH and hlsnative
-                                     only)
-    --skip-unavailable-fragments     Skip unavailable fragments (DASH and
-                                     hlsnative only)
+                                     is 10), or "infinite" (DASH, hlsnative and
+                                     ISM)
+    --skip-unavailable-fragments     Skip unavailable fragments (DASH, hlsnative
+                                     and ISM)
     --abort-on-unavailable-fragment  Abort downloading when some fragment is not
                                      available
+    --keep-fragments                 Keep downloaded fragments on disk after
+                                     downloading is finished; fragments are
+                                     erased by default
     --buffer-size SIZE               Size of download buffer (e.g. 1024 or 16K)
                                      (default is 1024)
     --no-resize-buffer               Do not automatically adjust the buffer
@@ -205,21 +220,11 @@ Alternatively, refer to the [developer instructions](#developer-instructions) fo
     --id                             Use only video ID in file name
     -o, --output TEMPLATE            Output filename template, see the "OUTPUT
                                      TEMPLATE" for all the info
-    --autonumber-size NUMBER         Specify the number of digits in
-                                     %(autonumber)s when it is present in output
-                                     filename template or --auto-number option
-                                     is given (default is 5)
     --autonumber-start NUMBER        Specify the start value for %(autonumber)s
                                      (default is 1)
     --restrict-filenames             Restrict filenames to only ASCII
                                      characters, and avoid "&" and spaces in
                                      filenames
-    -A, --auto-number                [deprecated; use -o
-                                     "%(autonumber)s-%(title)s.%(ext)s" ] Number
-                                     downloaded files starting from 00000
-    -t, --title                      [deprecated] Use title in file name
-                                     (default)
-    -l, --literal                    [deprecated] Alias of --title
     -w, --no-overwrites              Do not overwrite files
     -c, --continue                   Force resume of partially downloaded files.
                                      By default, youtube-dl will resume
@@ -373,8 +378,9 @@ Alternatively, refer to the [developer instructions](#developer-instructions) fo
                                      (requires ffmpeg or avconv and ffprobe or
                                      avprobe)
     --audio-format FORMAT            Specify audio format: "best", "aac",
-                                     "vorbis", "mp3", "m4a", "opus", or "wav";
-                                     "best" by default; No effect without -x
+                                     "flac", "mp3", "m4a", "opus", "vorbis", or
+                                     "wav"; "best" by default; No effect without
+                                     -x
     --audio-quality QUALITY          Specify ffmpeg/avconv audio quality, insert
                                      a value between 0 (better) and 9 (worse)
                                      for VBR or a specific bitrate like 128K
@@ -474,87 +480,89 @@ The `-o` option allows users to indicate a template for the output file names.
 
 **tl;dr:** [navigate me to examples](#output-template-examples).
 
-The basic usage is not to set any template arguments when downloading a single file, like in `youtube-dl -o funny_video.flv "http://some/video"`. However, it may contain special sequences that will be replaced when downloading each video. The special sequences have the format `%(NAME)s`. To clarify, that is a percent symbol followed by a name in parentheses, followed by a lowercase S. Allowed names are:
+The basic usage is not to set any template arguments when downloading a single file, like in `youtube-dl -o funny_video.flv "http://some/video"`. However, it may contain special sequences that will be replaced when downloading each video. The special sequences may be formatted according to [python string formatting operations](https://docs.python.org/2/library/stdtypes.html#string-formatting). For example, `%(NAME)s` or `%(NAME)05d`. To clarify, that is a percent symbol followed by a name in parentheses, followed by a formatting operations. Allowed names along with sequence type are:
 
- - `id`: Video identifier
- - `title`: Video title
- - `url`: Video URL
- - `ext`: Video filename extension
- - `alt_title`: A secondary title of the video
- - `display_id`: An alternative identifier for the video
- - `uploader`: Full name of the video uploader
- - `license`: License name the video is licensed under
- - `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)
- - `uploader_id`: Nickname or id of the video uploader
- - `location`: Physical location where the video was filmed
- - `duration`: Length of the video in seconds
- - `view_count`: How many users have watched the video on the platform
- - `like_count`: Number of positive ratings of the video
- - `dislike_count`: Number of negative ratings of the video
- - `repost_count`: Number of reposts of the video
- - `average_rating`: Average rating give by users, the scale used depends on the webpage
- - `comment_count`: Number of comments on the video
- - `age_limit`: Age restriction for the video (years)
- - `format`: A human-readable description of the format 
- - `format_id`: Format code specified by `--format`
- - `format_note`: Additional info about the format
- - `width`: Width of the video
- - `height`: Height of the video
- - `resolution`: Textual description of width and height
- - `tbr`: Average bitrate of audio and video in KBit/s
- - `abr`: Average audio bitrate in KBit/s
- - `acodec`: Name of the audio codec in use
- - `asr`: Audio sampling rate in Hertz
- - `vbr`: Average video bitrate in KBit/s
- - `fps`: Frame rate
- - `vcodec`: Name of the video codec in use
- - `container`: Name of the container format
- - `filesize`: The number of bytes, if known in advance
- - `filesize_approx`: An estimate for the number of bytes
- - `protocol`: The protocol that will be used for the actual download
- - `extractor`: Name of the extractor
- - `extractor_key`: Key name of the extractor
- - `epoch`: Unix epoch when creating the file
- - `autonumber`: Five-digit number that will be increased with each download, starting at zero
- - `playlist`: Name or id of the playlist that contains the video
- - `playlist_index`: Index of the video in the playlist padded with leading zeros according to the total length of the playlist
- - `playlist_id`: Playlist identifier
- - `playlist_title`: Playlist title
+ - `id` (string): Video identifier
+ - `title` (string): Video title
+ - `url` (string): Video URL
+ - `ext` (string): Video filename extension
+ - `alt_title` (string): A secondary title of the video
+ - `display_id` (string): An alternative identifier for the video
+ - `uploader` (string): Full name of the video uploader
+ - `license` (string): License name the video is licensed under
+ - `creator` (string): The creator of the video
+ - `release_date` (string): The date (YYYYMMDD) when the video was released
+ - `timestamp` (numeric): UNIX timestamp of the moment the video became available
+ - `upload_date` (string): Video upload date (YYYYMMDD)
+ - `uploader_id` (string): Nickname or id of the video uploader
+ - `location` (string): Physical location where the video was filmed
+ - `duration` (numeric): Length of the video in seconds
+ - `view_count` (numeric): How many users have watched the video on the platform
+ - `like_count` (numeric): Number of positive ratings of the video
+ - `dislike_count` (numeric): Number of negative ratings of the video
+ - `repost_count` (numeric): Number of reposts of the video
+ - `average_rating` (numeric): Average rating give by users, the scale used depends on the webpage
+ - `comment_count` (numeric): Number of comments on the video
+ - `age_limit` (numeric): Age restriction for the video (years)
+ - `format` (string): A human-readable description of the format 
+ - `format_id` (string): Format code specified by `--format`
+ - `format_note` (string): Additional info about the format
+ - `width` (numeric): Width of the video
+ - `height` (numeric): Height of the video
+ - `resolution` (string): Textual description of width and height
+ - `tbr` (numeric): Average bitrate of audio and video in KBit/s
+ - `abr` (numeric): Average audio bitrate in KBit/s
+ - `acodec` (string): Name of the audio codec in use
+ - `asr` (numeric): Audio sampling rate in Hertz
+ - `vbr` (numeric): Average video bitrate in KBit/s
+ - `fps` (numeric): Frame rate
+ - `vcodec` (string): Name of the video codec in use
+ - `container` (string): Name of the container format
+ - `filesize` (numeric): The number of bytes, if known in advance
+ - `filesize_approx` (numeric): An estimate for the number of bytes
+ - `protocol` (string): The protocol that will be used for the actual download
+ - `extractor` (string): Name of the extractor
+ - `extractor_key` (string): Key name of the extractor
+ - `epoch` (numeric): Unix epoch when creating the file
+ - `autonumber` (numeric): Five-digit number that will be increased with each download, starting at zero
+ - `playlist` (string): Name or id of the playlist that contains the video
+ - `playlist_index` (numeric): Index of the video in the playlist padded with leading zeros according to the total length of the playlist
+ - `playlist_id` (string): Playlist identifier
+ - `playlist_title` (string): Playlist title
 
 
 Available for the video that belongs to some logical chapter or section:
- - `chapter`: Name or title of the chapter the video belongs to
- - `chapter_number`: Number of the chapter the video belongs to
- - `chapter_id`: Id of the chapter the video belongs to
+ - `chapter` (string): Name or title of the chapter the video belongs to
+ - `chapter_number` (numeric): Number of the chapter the video belongs to
+ - `chapter_id` (string): Id of the chapter the video belongs to
 
 Available for the video that is an episode of some series or programme:
- - `series`: Title of the series or programme the video episode belongs to
- - `season`: Title of the season the video episode belongs to
- - `season_number`: Number of the season the video episode belongs to
- - `season_id`: Id of the season the video episode belongs to
- - `episode`: Title of the video episode
- - `episode_number`: Number of the video episode within a season
- - `episode_id`: Id of the video episode
+ - `series` (string): Title of the series or programme the video episode belongs to
+ - `season` (string): Title of the season the video episode belongs to
+ - `season_number` (numeric): Number of the season the video episode belongs to
+ - `season_id` (string): Id of the season the video episode belongs to
+ - `episode` (string): Title of the video episode
+ - `episode_number` (numeric): Number of the video episode within a season
+ - `episode_id` (string): 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
+ - `track` (string): Title of the track
+ - `track_number` (numeric): Number of the track within an album or a disc
+ - `track_id` (string): Id of the track
+ - `artist` (string): Artist(s) of the track
+ - `genre` (string): Genre(s) of the track
+ - `album` (string): Title of the album the track belongs to
+ - `album_type` (string): Type of the album
+ - `album_artist` (string): List of all artists appeared on the album
+ - `disc_number` (numeric): Number of the disc or other physical medium the track belongs to
+ - `release_year` (numeric): Year (YYYY) when the album was released
 
 Each aforementioned sequence when referenced in an 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 a particular extractor. Such sequences will be replaced with `NA`.
 
 For example for `-o %(title)s-%(id)s.%(ext)s` and an 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.
 
+For numeric sequences you can use numeric related formatting, for example, `%(view_count)05d` will result in a string with view count padded with zeros up to 5 characters, like in `00042`.
+
 Output templates can also contain arbitrary hierarchical path, e.g. `-o '%(playlist)s/%(playlist_index)s - %(title)s.%(ext)s'` which will result in downloading each video in a directory corresponding to this path template. Any missing directory will be automatically created for you.
 
 To use percent literals in an output template use `%%`. To output to stdout use `-o -`.

devscripts/make_lazy_extractors.py

@@ -1,6 +1,7 @@
 from __future__ import unicode_literals, print_function
 
 from inspect import getsource
+import io
 import os
 from os.path import dirname as dirn
 import sys
@@ -95,5 +96,5 @@ module_contents.append(
 
 module_src = '\n'.join(module_contents) + '\n'
 
-with open(lazy_extractors_filename, 'wt') as f:
+with io.open(lazy_extractors_filename, 'wt', encoding='utf-8') as f:
     f.write(module_src)

devscripts/run_tests.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+
+DOWNLOAD_TESTS="age_restriction|download|subtitles|write_annotations|iqiyi_sdk_interpreter|youtube_lists"
+
+test_set=""
+multiprocess_args=""
+
+case "$YTDL_TEST_SET" in
+    core)
+        test_set="-I test_($DOWNLOAD_TESTS)\.py"
+    ;;
+    download)
+        test_set="-I test_(?!$DOWNLOAD_TESTS).+\.py"
+        multiprocess_args="--processes=4 --process-timeout=540"
+    ;;
+    *)
+        break
+    ;;
+esac
+
+nosetests test --verbose $test_set $multiprocess_args

docs/supportedsites.md

@@ -11,6 +11,7 @@
  - **4tube**
  - **56.com**
  - **5min**
+ - **6play**
  - **8tracks**
  - **91porn**
  - **9c9media**
@@ -27,6 +28,7 @@
  - **acast**
  - **acast:channel**
  - **AddAnime**
+ - **ADN**: Anime Digital Network
  - **AdobeTV**
  - **AdobeTVChannel**
  - **AdobeTVShow**
@@ -66,6 +68,7 @@
  - **arte.tv:playlist**
  - **AtresPlayer**
  - **ATTTechChannel**
+ - **ATVAt**
  - **AudiMedia**
  - **AudioBoom**
  - **audiomack**
@@ -77,8 +80,7 @@
  - **awaan:video**
  - **AZMedien**: AZ Medien videos
  - **AZMedienPlaylist**: AZ Medien playlists
- - **Azubu**
- - **AzubuLive**
+ - **AZMedienShowPlaylist**: AZ Medien show playlists
  - **BaiduVideo**: 百度视频
  - **bambuser**
  - **bambuser:channel**
@@ -106,6 +108,7 @@
  - **blinkx**
  - **Bloomberg**
  - **BokeCC**
+ - **BostonGlobe**
  - **Bpb**: Bundeszentrale für politische Bildung
  - **BR**: Bayerischer Rundfunk Mediathek
  - **BravoTV**
@@ -122,7 +125,7 @@
  - **CamWithHer**
  - **canalc2.tv**
  - **Canalplus**: canalplus.fr, piwiplus.fr and d8.tv
- - **Canvas**
+ - **Canvas**: canvas.be and een.be
  - **CarambaTV**
  - **CarambaTVPage**
  - **CartoonNetwork**
@@ -140,6 +143,7 @@
  - **CCTV**: 央视网
  - **CDA**
  - **CeskaTelevize**
+ - **CeskaTelevizePorady**
  - **channel9**: Channel 9
  - **CharlieRose**
  - **Chaturbate**
@@ -168,6 +172,7 @@
  - **ComedyCentralShortname**
  - **ComedyCentralTV**
  - **CondeNast**: Condé Nast media group: Allure, Architectural Digest, Ars Technica, Bon Appétit, Brides, Condé Nast, Condé Nast Traveler, Details, Epicurious, GQ, Glamour, Golf Digest, SELF, Teen Vogue, The New Yorker, Vanity Fair, Vogue, W Magazine, WIRED
+ - **Corus**
  - **Coub**
  - **Cracked**
  - **Crackle**
@@ -189,6 +194,8 @@
  - **dailymotion:playlist**
  - **dailymotion:user**
  - **DailymotionCloud**
+ - **Daisuki**
+ - **DaisukiPlaylist**
  - **daum.net**
  - **daum.net:clip**
  - **daum.net:playlist**
@@ -203,10 +210,14 @@
  - **Digiteka**
  - **Discovery**
  - **DiscoveryGo**
+ - **DiscoveryGoPlaylist**
+ - **DiscoveryNetworksDe**
+ - **DiscoveryVR**
  - **Disney**
  - **Dotsub**
  - **DouyuTV**: 斗鱼
  - **DPlay**
+ - **DPlayIt**
  - **dramafever**
  - **dramafever:series**
  - **DRBonanza**
@@ -237,6 +248,7 @@
  - **ESPN**
  - **ESPNArticle**
  - **EsriVideo**
+ - **ETOnline**
  - **Europa**
  - **EveryonesMixtape**
  - **ExpoTV**
@@ -272,6 +284,7 @@
  - **francetvinfo.fr**
  - **Freesound**
  - **freespeech.org**
+ - **FreshLive**
  - **Funimation**
  - **FunnyOrDie**
  - **Fusion**
@@ -282,6 +295,7 @@
  - **Gamersyde**
  - **GameSpot**
  - **GameStar**
+ - **Gaskrank**
  - **Gazeta**
  - **GDCVault**
  - **generic**: Generic downloader that works on some sites
@@ -292,6 +306,7 @@
  - **Globo**
  - **GloboArticle**
  - **Go**
+ - **Go90**
  - **GodTube**
  - **GodTV**
  - **Golem**
@@ -300,14 +315,14 @@
  - **GPUTechConf**
  - **Groupon**
  - **Hark**
- - **HBO**
- - **HBOEpisode**
+ - **hbo**
+ - **hbo:episode**
  - **HearThisAt**
  - **Heise**
  - **HellPorno**
  - **Helsinki**: helsinki.fi
  - **HentaiStigma**
- - **HGTV**
+ - **hetklokhuis**
  - **hgtv.com:show**
  - **HistoricFilms**
  - **history:topic**: History.com Topic
@@ -415,6 +430,8 @@
  - **MatchTV**
  - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
+ - **Medialaan**
+ - **Medici**
  - **Meipai**: 美拍
  - **MelonVOD**
  - **META**
@@ -509,6 +526,7 @@
  - **Nintendo**
  - **njoy**: N-JOY
  - **njoy:embed**
+ - **NJPWWorld**: 新日本プロレスワールド
  - **NobelPrize**
  - **Noco**
  - **Normalboots**
@@ -544,8 +562,10 @@
  - **OktoberfestTV**
  - **on.aol.com**
  - **OnDemandKorea**
+ - **onet.pl**
  - **onet.tv**
  - **onet.tv:channel**
+ - **OnetMVP**
  - **OnionStudios**
  - **Ooyala**
  - **OoyalaExternal**
@@ -555,6 +575,8 @@
  - **orf:iptv**: iptv.ORF.at
  - **orf:oe1**: Radio Österreich 1
  - **orf:tvthek**: ORF TVthek
+ - **PacktPub**
+ - **PacktPubCourse**
  - **PandaTV**: 熊猫TV
  - **pandora.tv**: 판도라TV
  - **parliamentlive.tv**: UK parliament videos
@@ -612,9 +634,10 @@
  - **radiofrance**
  - **RadioJavan**
  - **Rai**
- - **RaiTV**
+ - **RaiPlay**
  - **RBMARadio**
  - **RDS**: RDS.ca
+ - **RedBullTV**
  - **RedTube**
  - **RegioTV**
  - **RENTV**
@@ -636,7 +659,9 @@
  - **rte**: Raidió Teilifís Éireann TV
  - **rte:radio**: Raidió Teilifís Éireann radio
  - **rtl.nl**: rtl.nl and rtlxl.nl
- - **RTL2**
+ - **rtl2**
+ - **rtl2:you**
+ - **rtl2:you:series**
  - **RTP**
  - **RTS**: RTS.ch
  - **rtve.es:alacarta**: RTVE a la carta
@@ -662,10 +687,10 @@
  - **savefrom.net**
  - **SBS**: sbs.com.au
  - **schooltv**
- - **SciVee**
  - **screen.yahoo:search**: Yahoo screen search
  - **Screencast**
  - **ScreencastOMatic**
+ - **scrippsnetworks:watch**
  - **Seeker**
  - **SenateISVP**
  - **SendtoNews**
@@ -675,7 +700,7 @@
  - **Shared**: shared.sx
  - **ShowRoomLive**
  - **Sina**
- - **SixPlay**
+ - **SkylineWebcams**
  - **skynewsarabia:article**
  - **skynewsarabia:video**
  - **SkySports**
@@ -710,6 +735,7 @@
  - **SportBoxEmbed**
  - **SportDeutschland**
  - **Sportschau**
+ - **Sprout**
  - **sr:mediathek**: Saarländischer Rundfunk
  - **SRGSSR**
  - **SRGSSRPlay**: srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites
@@ -717,6 +743,7 @@
  - **Steam**
  - **Stitcher**
  - **Streamable**
+ - **Streamango**
  - **streamcloud.eu**
  - **StreamCZ**
  - **StreetVoice**
@@ -757,17 +784,18 @@
  - **TheScene**
  - **TheSixtyOne**
  - **TheStar**
+ - **TheSun**
  - **TheWeatherChannel**
  - **ThisAmericanLife**
  - **ThisAV**
  - **ThisOldHouse**
  - **tinypic**: tinypic.com videos
- - **tlc.de**
  - **TMZ**
  - **TMZArticle**
  - **TNAFlix**
  - **TNAFlixNetworkEmbed**
  - **toggle**
+ - **ToonGoggles**
  - **Tosh**: Tosh.0
  - **tou.tv**
  - **Toypics**: Toypics user profile
@@ -785,13 +813,16 @@
  - **tunein:program**
  - **tunein:station**
  - **tunein:topic**
+ - **TunePk**
  - **Turbo**
  - **Tutv**
  - **tv.dfb.de**
  - **TV2**
+ - **tv2.hu**
  - **TV2Article**
  - **TV3**
  - **TV4**: tv4.se and tv4play.se
+ - **TV5MondePlus**: TV5MONDE+
  - **TVA**
  - **TVANouvelles**
  - **TVANouvellesArticle**
@@ -799,10 +830,12 @@
  - **TVCArticle**
  - **tvigle**: Интернет-телевидение Tvigle.ru
  - **tvland.com**
+ - **TVN24**
  - **TVNoe**
  - **tvp**: Telewizja Polska
  - **tvp:embed**: Telewizja Polska
  - **tvp:series**
+ - **TVPlayer**
  - **Tweakers**
  - **twitch:chapter**
  - **twitch:clips**
@@ -859,13 +892,14 @@
  - **videomore:season**
  - **videomore:video**
  - **VideoPremium**
+ - **VideoPress**
  - **videoweed**: VideoWeed
  - **Vidio**
  - **vidme**
  - **vidme:user**
  - **vidme:user:likes**
  - **Vidzi**
- - **vier**
+ - **vier**: vier.be and vijf.be
  - **vier:videos**
  - **ViewLift**
  - **ViewLiftEmbed**
@@ -895,12 +929,17 @@
  - **vlive**
  - **vlive:channel**
  - **Vodlocker**
+ - **VODPl**
  - **VODPlatform**
  - **VoiceRepublic**
  - **VoxMedia**
  - **Vporn**
  - **vpro**: npo.nl and ntr.nl
- - **VRT**
+ - **Vrak**
+ - **VRT**: deredactie.be, sporza.be, cobra.be and cobra.canvas.be
+ - **vrv**
+ - **vrv:series**
+ - **VShare**
  - **vube**: Vube.com
  - **VuClip**
  - **VVVVID**
@@ -926,9 +965,10 @@
  - **wrzuta.pl**
  - **wrzuta.pl:playlist**
  - **WSJ**: Wall Street Journal
+ - **WSJArticle**
  - **XBef**
  - **XboxClips**
- - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To, XVIDSTAGE
+ - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To, XVIDSTAGE, Vid ABC, VidBom, vidlo
  - **XHamster**
  - **XHamsterEmbed**
  - **xiami:album**: 虾米音乐 - 专辑

setup.py

@@ -107,8 +107,8 @@ setup(
     url='https://github.com/rg3/youtube-dl',
     author='Ricardo Garcia',
     author_email='ytdl@yt-dl.org',
-    maintainer='Philipp Hagemeister',
-    maintainer_email='phihag@phihag.de',
+    maintainer='Sergey M.',
+    maintainer_email='dstftw@gmail.com',
     packages=[
         'youtube_dl',
         'youtube_dl.extractor', 'youtube_dl.downloader',
@@ -130,6 +130,7 @@ setup(
         'Programming Language :: Python :: 3.3',
         'Programming Language :: Python :: 3.4',
         'Programming Language :: Python :: 3.5',
+        'Programming Language :: Python :: 3.6',
     ],
 
     cmdclass={'build_lazy_extractors': build_lazy_extractors},

test/test_InfoExtractor.py

@@ -3,12 +3,13 @@
 from __future__ import unicode_literals
 
 # Allow direct execution
+import io
 import os
 import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL
+from test.helper import FakeYDL, expect_dict, expect_value
 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, RegexNotFoundError
@@ -84,6 +85,445 @@ class TestInfoExtractor(unittest.TestCase):
         self.assertRaises(ExtractorError, self.ie._download_json, uri, None)
         self.assertEqual(self.ie._download_json(uri, None, fatal=False), None)
 
+    def test_extract_jwplayer_data_realworld(self):
+        # from http://www.suffolk.edu/sjc/
+        expect_dict(
+            self,
+            self.ie._extract_jwplayer_data(r'''
+                <script type='text/javascript'>
+                    jwplayer('my-video').setup({
+                        file: 'rtmp://192.138.214.154/live/sjclive',
+                        fallback: 'true',
+                        width: '95%',
+                      aspectratio: '16:9',
+                      primary: 'flash',
+                      mediaid:'XEgvuql4'
+                    });
+                </script>
+                ''', None, require_title=False),
+            {
+                'id': 'XEgvuql4',
+                'formats': [{
+                    'url': 'rtmp://192.138.214.154/live/sjclive',
+                    'ext': 'flv'
+                }]
+            })
+
+        # from https://www.pornoxo.com/videos/7564/striptease-from-sexy-secretary/
+        expect_dict(
+            self,
+            self.ie._extract_jwplayer_data(r'''
+<script type="text/javascript">
+    jwplayer("mediaplayer").setup({
+        'videoid': "7564",
+        'width': "100%",
+        'aspectratio': "16:9",
+        'stretching': "exactfit",
+        'autostart': 'false',
+        'flashplayer': "https://t04.vipstreamservice.com/jwplayer/v5.10/player.swf",
+        'file': "https://cdn.pornoxo.com/key=MF+oEbaxqTKb50P-w9G3nA,end=1489689259,ip=104.199.146.27/ip=104.199.146.27/speed=6573765/buffer=3.0/2009-12/4b2157147afe5efa93ce1978e0265289c193874e02597.flv",
+        'image': "https://t03.vipstreamservice.com/thumbs/pxo-full/2009-12/14/a4b2157147afe5efa93ce1978e0265289c193874e02597.flv-full-13.jpg",
+        'filefallback': "https://cdn.pornoxo.com/key=9ZPsTR5EvPLQrBaak2MUGA,end=1489689259,ip=104.199.146.27/ip=104.199.146.27/speed=6573765/buffer=3.0/2009-12/m_4b2157147afe5efa93ce1978e0265289c193874e02597.mp4",
+        'logo.hide': true,
+        'skin': "https://t04.vipstreamservice.com/jwplayer/skin/modieus-blk.zip",
+        'plugins': "https://t04.vipstreamservice.com/jwplayer/dock/dockableskinnableplugin.swf",
+        'dockableskinnableplugin.piclink': "/index.php?key=ajax-videothumbsn&vid=7564&data=2009-12--14--4b2157147afe5efa93ce1978e0265289c193874e02597.flv--17370",
+        'controlbar': 'bottom',
+        'modes': [
+            {type: 'flash', src: 'https://t04.vipstreamservice.com/jwplayer/v5.10/player.swf'}
+        ],
+        'provider': 'http'
+    });
+    //noinspection JSAnnotator
+    invideo.setup({
+        adsUrl: "/banner-iframe/?zoneId=32",
+        adsUrl2: "",
+        autostart: false
+    });
+</script>
+            ''', 'dummy', require_title=False),
+            {
+                'thumbnail': 'https://t03.vipstreamservice.com/thumbs/pxo-full/2009-12/14/a4b2157147afe5efa93ce1978e0265289c193874e02597.flv-full-13.jpg',
+                'formats': [{
+                    'url': 'https://cdn.pornoxo.com/key=MF+oEbaxqTKb50P-w9G3nA,end=1489689259,ip=104.199.146.27/ip=104.199.146.27/speed=6573765/buffer=3.0/2009-12/4b2157147afe5efa93ce1978e0265289c193874e02597.flv',
+                    'ext': 'flv'
+                }]
+            })
+
+        # from http://www.indiedb.com/games/king-machine/videos
+        expect_dict(
+            self,
+            self.ie._extract_jwplayer_data(r'''
+<script>
+jwplayer("mediaplayer").setup({"abouttext":"Visit Indie DB","aboutlink":"http:\/\/www.indiedb.com\/","displaytitle":false,"autostart":false,"repeat":false,"title":"king machine trailer 1","sharing":{"link":"http:\/\/www.indiedb.com\/games\/king-machine\/videos\/king-machine-trailer-1","code":"<iframe width=\"560\" height=\"315\" src=\"http:\/\/www.indiedb.com\/media\/iframe\/1522983\" frameborder=\"0\" allowfullscreen><\/iframe><br><a href=\"http:\/\/www.indiedb.com\/games\/king-machine\/videos\/king-machine-trailer-1\">king machine trailer 1 - Indie DB<\/a>"},"related":{"file":"http:\/\/rss.indiedb.com\/media\/recommended\/1522983\/feed\/rss.xml","dimensions":"160x120","onclick":"link"},"sources":[{"file":"http:\/\/cdn.dbolical.com\/cache\/videos\/games\/1\/50\/49678\/encode_mp4\/king-machine-trailer.mp4","label":"360p SD","default":"true"},{"file":"http:\/\/cdn.dbolical.com\/cache\/videos\/games\/1\/50\/49678\/encode720p_mp4\/king-machine-trailer.mp4","label":"720p HD"}],"image":"http:\/\/media.indiedb.com\/cache\/images\/games\/1\/50\/49678\/thumb_620x2000\/king-machine-trailer.mp4.jpg","advertising":{"client":"vast","tag":"http:\/\/ads.intergi.com\/adrawdata\/3.0\/5205\/4251742\/0\/1013\/ADTECH;cors=yes;width=560;height=315;referring_url=http:\/\/www.indiedb.com\/games\/king-machine\/videos\/king-machine-trailer-1;content_url=http:\/\/www.indiedb.com\/games\/king-machine\/videos\/king-machine-trailer-1;media_id=1522983;title=king+machine+trailer+1;device=__DEVICE__;model=__MODEL__;os=Windows+OS;osversion=__OSVERSION__;ua=__UA__;ip=109.171.17.81;uniqueid=1522983;tags=__TAGS__;number=58cac25928151;time=1489683033"},"width":620,"height":349}).once("play", function(event) {
+            videoAnalytics("play");
+}).once("complete", function(event) {
+    videoAnalytics("completed");
+});
+</script>
+                ''', 'dummy'),
+            {
+                'title': 'king machine trailer 1',
+                'thumbnail': 'http://media.indiedb.com/cache/images/games/1/50/49678/thumb_620x2000/king-machine-trailer.mp4.jpg',
+                'formats': [{
+                    'url': 'http://cdn.dbolical.com/cache/videos/games/1/50/49678/encode_mp4/king-machine-trailer.mp4',
+                    'height': 360,
+                    'ext': 'mp4'
+                }, {
+                    'url': 'http://cdn.dbolical.com/cache/videos/games/1/50/49678/encode720p_mp4/king-machine-trailer.mp4',
+                    'height': 720,
+                    'ext': 'mp4'
+                }]
+            })
+
+    def test_parse_m3u8_formats(self):
+        _TEST_CASES = [
+            (
+                # https://github.com/rg3/youtube-dl/issues/11507
+                # http://pluzz.francetv.fr/videos/le_ministere.html
+                'pluzz_francetv_11507',
+                'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/master.m3u8?caption=2017%2F16%2F156589847-1492488987.m3u8%3Afra%3AFrancais&audiotrack=0%3Afra%3AFrancais',
+                [{
+                    'url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/master.m3u8?caption=2017%2F16%2F156589847-1492488987.m3u8%3Afra%3AFrancais&audiotrack=0%3Afra%3AFrancais',
+                    'ext': 'mp4',
+                    'format_id': 'meta',
+                    'format_note': 'Quality selection URL',
+                    'protocol': 'm3u8',
+                    'preference': -100,
+                    'resolution': 'multiple'
+                }, {
+                    'url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_0_av.m3u8?null=0',
+                    'manifest_url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_0_av.m3u8?null=0',
+                    'ext': 'mp4',
+                    'format_id': '180',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.66.30',
+                    'tbr': 180,
+                    'width': 256,
+                    'height': 144,
+                }, {
+                    'url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_1_av.m3u8?null=0',
+                    'manifest_url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_1_av.m3u8?null=0',
+                    'ext': 'mp4',
+                    'format_id': '303',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.66.30',
+                    'tbr': 303,
+                    'width': 320,
+                    'height': 180,
+                }, {
+                    'url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_2_av.m3u8?null=0',
+                    'manifest_url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_2_av.m3u8?null=0',
+                    'ext': 'mp4',
+                    'format_id': '575',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.66.30',
+                    'tbr': 575,
+                    'width': 512,
+                    'height': 288,
+                }, {
+                    'url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_3_av.m3u8?null=0',
+                    'manifest_url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_3_av.m3u8?null=0',
+                    'ext': 'mp4',
+                    'format_id': '831',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.77.30',
+                    'tbr': 831,
+                    'width': 704,
+                    'height': 396,
+                }, {
+                    'url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_4_av.m3u8?null=0',
+                    'manifest_url': 'http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_4_av.m3u8?null=0',
+                    'ext': 'mp4',
+                    'protocol': 'm3u8',
+                    'format_id': '1467',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.77.30',
+                    'tbr': 1467,
+                    'width': 1024,
+                    'height': 576,
+                }]
+            ),
+            (
+                # https://github.com/rg3/youtube-dl/issues/11995
+                # http://teamcoco.com/video/clueless-gamer-super-bowl-for-honor
+                'teamcoco_11995',
+                'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/main.m3u8',
+                [{
+                    'url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/main.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'meta',
+                    'format_note': 'Quality selection URL',
+                    'protocol': 'm3u8',
+                    'preference': -100,
+                    'resolution': 'multiple',
+                }, {
+                    'url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-audio-160k_v4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'audio-0-Default',
+                    'protocol': 'm3u8',
+                    'vcodec': 'none',
+                }, {
+                    'url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-audio-64k_v4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'audio-1-Default',
+                    'protocol': 'm3u8',
+                    'vcodec': 'none',
+                }, {
+                    'url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-audio-64k_v4.m3u8',
+                    'manifest_url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-audio-64k_v4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '71',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.5',
+                    'vcodec': 'none',
+                    'tbr': 71,
+                }, {
+                    'url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-400k_v4.m3u8',
+                    'manifest_url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-400k_v4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '413',
+                    'protocol': 'm3u8',
+                    'acodec': 'none',
+                    'vcodec': 'avc1.42001e',
+                    'tbr': 413,
+                    'width': 400,
+                    'height': 224,
+                }, {
+                    'url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-400k_v4.m3u8',
+                    'manifest_url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-400k_v4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '522',
+                    'protocol': 'm3u8',
+                    'acodec': 'none',
+                    'vcodec': 'avc1.42001e',
+                    'tbr': 522,
+                    'width': 400,
+                    'height': 224,
+                }, {
+                    'url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-1m_v4.m3u8',
+                    'manifest_url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-1m_v4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '1205',
+                    'protocol': 'm3u8',
+                    'acodec': 'none',
+                    'vcodec': 'avc1.4d001e',
+                    'tbr': 1205,
+                    'width': 640,
+                    'height': 360,
+                }, {
+                    'url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-2m_v4.m3u8',
+                    'manifest_url': 'http://ak.storage-w.teamcococdn.com/cdn/2017-02/98599/ed8f/hls/CONAN_020217_Highlight_show-2m_v4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '2374',
+                    'protocol': 'm3u8',
+                    'acodec': 'none',
+                    'vcodec': 'avc1.4d001f',
+                    'tbr': 2374,
+                    'width': 1024,
+                    'height': 576,
+                }]
+            ),
+            (
+                # https://github.com/rg3/youtube-dl/issues/12211
+                # http://video.toggle.sg/en/series/whoopie-s-world/ep3/478601
+                'toggle_mobile_12211',
+                'http://cdnapi.kaltura.com/p/2082311/sp/208231100/playManifest/protocol/http/entryId/0_89q6e8ku/format/applehttp/tags/mobile_sd/f/a.m3u8',
+                [{
+                    'url': 'http://cdnapi.kaltura.com/p/2082311/sp/208231100/playManifest/protocol/http/entryId/0_89q6e8ku/format/applehttp/tags/mobile_sd/f/a.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'meta',
+                    'format_note': 'Quality selection URL',
+                    'protocol': 'm3u8',
+                    'preference': -100,
+                    'resolution': 'multiple'
+                }, {
+                    'url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_sa2ntrdg/name/a.mp4/index.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'audio-English',
+                    'protocol': 'm3u8',
+                    'language': 'eng',
+                    'vcodec': 'none',
+                }, {
+                    'url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_r7y0nitg/name/a.mp4/index.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'audio-Undefined',
+                    'protocol': 'm3u8',
+                    'language': 'und',
+                    'vcodec': 'none',
+                }, {
+                    'url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_qlk9hlzr/name/a.mp4/index.m3u8',
+                    'manifest_url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_qlk9hlzr/name/a.mp4/index.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '155',
+                    'protocol': 'm3u8',
+                    'tbr': 155.648,
+                    'width': 320,
+                    'height': 180,
+                }, {
+                    'url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_oefackmi/name/a.mp4/index.m3u8',
+                    'manifest_url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_oefackmi/name/a.mp4/index.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '502',
+                    'protocol': 'm3u8',
+                    'tbr': 502.784,
+                    'width': 480,
+                    'height': 270,
+                }, {
+                    'url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/12/pv/1/flavorId/0_vyg9pj7k/name/a.mp4/index.m3u8',
+                    'manifest_url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/12/pv/1/flavorId/0_vyg9pj7k/name/a.mp4/index.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '827',
+                    'protocol': 'm3u8',
+                    'tbr': 827.392,
+                    'width': 640,
+                    'height': 360,
+                }, {
+                    'url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/12/pv/1/flavorId/0_50n4psvx/name/a.mp4/index.m3u8',
+                    'manifest_url': 'http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/12/pv/1/flavorId/0_50n4psvx/name/a.mp4/index.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '1396',
+                    'protocol': 'm3u8',
+                    'tbr': 1396.736,
+                    'width': 854,
+                    'height': 480,
+                }]
+            ),
+            (
+                # http://www.twitch.tv/riotgames/v/6528877
+                'twitch_vod',
+                'https://usher.ttvnw.net/vod/6528877?allow_source=true&allow_audio_only=true&allow_spectre=true&player=twitchweb&nauth=%7B%22user_id%22%3Anull%2C%22vod_id%22%3A6528877%2C%22expires%22%3A1492887874%2C%22chansub%22%3A%7B%22restricted_bitrates%22%3A%5B%5D%7D%2C%22privileged%22%3Afalse%2C%22https_required%22%3Afalse%7D&nauthsig=3e29296a6824a0f48f9e731383f77a614fc79bee',
+                [{
+                    'url': 'https://usher.ttvnw.net/vod/6528877?allow_source=true&allow_audio_only=true&allow_spectre=true&player=twitchweb&nauth=%7B%22user_id%22%3Anull%2C%22vod_id%22%3A6528877%2C%22expires%22%3A1492887874%2C%22chansub%22%3A%7B%22restricted_bitrates%22%3A%5B%5D%7D%2C%22privileged%22%3Afalse%2C%22https_required%22%3Afalse%7D&nauthsig=3e29296a6824a0f48f9e731383f77a614fc79bee',
+                    'ext': 'mp4',
+                    'format_id': 'meta',
+                    'format_note': 'Quality selection URL',
+                    'protocol': 'm3u8',
+                    'preference': -100,
+                    'resolution': 'multiple'
+                }, {
+                    'url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/audio_only/index-muted-HM49I092CC.m3u8',
+                    'manifest_url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/audio_only/index-muted-HM49I092CC.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'Audio Only',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'none',
+                    'tbr': 182.725,
+                }, {
+                    'url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/mobile/index-muted-HM49I092CC.m3u8',
+                    'manifest_url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/mobile/index-muted-HM49I092CC.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'Mobile',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.42C00D',
+                    'tbr': 280.474,
+                    'width': 400,
+                    'height': 226,
+                }, {
+                    'url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/low/index-muted-HM49I092CC.m3u8',
+                    'manifest_url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/low/index-muted-HM49I092CC.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'Low',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.42C01E',
+                    'tbr': 628.347,
+                    'width': 640,
+                    'height': 360,
+                }, {
+                    'url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/medium/index-muted-HM49I092CC.m3u8',
+                    'manifest_url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/medium/index-muted-HM49I092CC.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'Medium',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.42C01E',
+                    'tbr': 893.387,
+                    'width': 852,
+                    'height': 480,
+                }, {
+                    'url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/high/index-muted-HM49I092CC.m3u8',
+                    'manifest_url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/high/index-muted-HM49I092CC.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'High',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.42C01F',
+                    'tbr': 1603.789,
+                    'width': 1280,
+                    'height': 720,
+                }, {
+                    'url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/chunked/index-muted-HM49I092CC.m3u8',
+                    'manifest_url': 'https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/chunked/index-muted-HM49I092CC.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'Source',
+                    'protocol': 'm3u8',
+                    'acodec': 'mp4a.40.2',
+                    'vcodec': 'avc1.100.31',
+                    'tbr': 3214.134,
+                    'width': 1280,
+                    'height': 720,
+                }]
+            ),
+            (
+                # http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015
+                # EXT-X-STREAM-INF tag with NAME attribute that is not defined
+                # in HLS specification
+                'vidio',
+                'https://www.vidio.com/videos/165683/playlist.m3u8',
+                [{
+                    'url': 'https://www.vidio.com/videos/165683/playlist.m3u8',
+                    'ext': 'mp4',
+                    'format_id': 'meta',
+                    'format_note': 'Quality selection URL',
+                    'protocol': 'm3u8',
+                    'preference': -100,
+                    'resolution': 'multiple'
+                }, {
+                    'url': 'https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b300.mp4.m3u8',
+                    'manifest_url': 'https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b300.mp4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '270p 3G',
+                    'protocol': 'm3u8',
+                    'tbr': 300,
+                    'width': 480,
+                    'height': 270,
+                }, {
+                    'url': 'https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b600.mp4.m3u8',
+                    'manifest_url': 'https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b600.mp4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '360p SD',
+                    'protocol': 'm3u8',
+                    'tbr': 600,
+                    'width': 640,
+                    'height': 360,
+                }, {
+                    'url': 'https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b1200.mp4.m3u8',
+                    'manifest_url': 'https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b1200.mp4.m3u8',
+                    'ext': 'mp4',
+                    'format_id': '720p HD',
+                    'protocol': 'm3u8',
+                    'tbr': 1200,
+                    'width': 1280,
+                    'height': 720,
+                }]
+            )
+        ]
+
+        for m3u8_file, m3u8_url, expected_formats in _TEST_CASES:
+            with io.open('./test/testdata/m3u8/%s.m3u8' % m3u8_file,
+                         mode='r', encoding='utf-8') as f:
+                formats = self.ie._parse_m3u8_formats(
+                    f.read(), m3u8_url, ext='mp4')
+                self.ie._sort_formats(formats)
+                expect_value(self, formats, expected_formats, None)
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_YoutubeDL.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python
+# coding: utf-8
 
 from __future__ import unicode_literals
 
@@ -525,6 +526,7 @@ class TestYoutubeDL(unittest.TestCase):
             'id': '1234',
             'ext': 'mp4',
             'width': None,
+            'height': 1080,
         }
 
         def fname(templ):
@@ -534,16 +536,29 @@ class TestYoutubeDL(unittest.TestCase):
         self.assertEqual(fname('%(id)s-%(width)s.%(ext)s'), '1234-NA.mp4')
         # Replace missing fields with 'NA'
         self.assertEqual(fname('%(uploader_date)s-%(id)s.%(ext)s'), 'NA-1234.mp4')
+        self.assertEqual(fname('%(height)d.%(ext)s'), '1080.mp4')
+        self.assertEqual(fname('%(height)6d.%(ext)s'), '  1080.mp4')
+        self.assertEqual(fname('%(height)-6d.%(ext)s'), '1080  .mp4')
+        self.assertEqual(fname('%(height)06d.%(ext)s'), '001080.mp4')
+        self.assertEqual(fname('%(height) 06d.%(ext)s'), ' 01080.mp4')
+        self.assertEqual(fname('%(height)   06d.%(ext)s'), ' 01080.mp4')
+        self.assertEqual(fname('%(height)0 6d.%(ext)s'), ' 01080.mp4')
+        self.assertEqual(fname('%(height)0   6d.%(ext)s'), ' 01080.mp4')
+        self.assertEqual(fname('%(height)   0   6d.%(ext)s'), ' 01080.mp4')
+        self.assertEqual(fname('%%(height)06d.%(ext)s'), '%(height)06d.mp4')
+        self.assertEqual(fname('%(width)06d.%(ext)s'), 'NA.mp4')
+        self.assertEqual(fname('%(width)06d.%%(ext)s'), 'NA.%(ext)s')
+        self.assertEqual(fname('%%(width)06d.%(ext)s'), '%(width)06d.mp4')
 
     def test_format_note(self):
         ydl = YoutubeDL()
         self.assertEqual(ydl._format_note({}), '')
         assertRegexpMatches(self, ydl._format_note({
             'vbr': 10,
-        }), '^\s*10k$')
+        }), r'^\s*10k$')
         assertRegexpMatches(self, ydl._format_note({
             'fps': 30,
-        }), '^30fps$')
+        }), r'^30fps$')
 
     def test_postprocessors(self):
         filename = 'post-processor-testfile.mp4'
@@ -606,6 +621,8 @@ class TestYoutubeDL(unittest.TestCase):
             'duration': 30,
             'filesize': 10 * 1024,
             'playlist_id': '42',
+            'uploader': "變態妍字幕版 太妍 тест",
+            'creator': "тест ' 123 ' тест--",
         }
         second = {
             'id': '2',
@@ -616,6 +633,7 @@ class TestYoutubeDL(unittest.TestCase):
             'description': 'foo',
             'filesize': 5 * 1024,
             'playlist_id': '43',
+            'uploader': "тест 123",
         }
         videos = [first, second]
 
@@ -656,6 +674,26 @@ class TestYoutubeDL(unittest.TestCase):
         res = get_videos(f)
         self.assertEqual(res, ['1'])
 
+        f = match_filter_func('uploader = "變態妍字幕版 太妍 тест"')
+        res = get_videos(f)
+        self.assertEqual(res, ['1'])
+
+        f = match_filter_func('uploader != "變態妍字幕版 太妍 тест"')
+        res = get_videos(f)
+        self.assertEqual(res, ['2'])
+
+        f = match_filter_func('creator = "тест \' 123 \' тест--"')
+        res = get_videos(f)
+        self.assertEqual(res, ['1'])
+
+        f = match_filter_func("creator = 'тест \\' 123 \\' тест--'")
+        res = get_videos(f)
+        self.assertEqual(res, ['1'])
+
+        f = match_filter_func(r"creator = 'тест \' 123 \' тест--' & duration > 30")
+        res = get_videos(f)
+        self.assertEqual(res, [])
+
     def test_playlist_items_selection(self):
         entries = [{
             'id': compat_str(i),
@@ -717,6 +755,7 @@ class TestYoutubeDL(unittest.TestCase):
                     '_type': 'url_transparent',
                     'url': 'foo2:',
                     'ie_key': 'Foo2',
+                    'title': 'foo1 title'
                 }
 
         class Foo2IE(InfoExtractor):
@@ -733,7 +772,7 @@ class TestYoutubeDL(unittest.TestCase):
             _VALID_URL = r'foo3:'
 
             def _real_extract(self, url):
-                return _make_result([{'url': TEST_URL}])
+                return _make_result([{'url': TEST_URL}], title='foo3 title')
 
         ydl.add_info_extractor(Foo1IE(ydl))
         ydl.add_info_extractor(Foo2IE(ydl))
@@ -741,6 +780,7 @@ class TestYoutubeDL(unittest.TestCase):
         ydl.extract_info('foo1:')
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['url'], TEST_URL)
+        self.assertEqual(downloaded['title'], 'foo1 title')
 
 
 if __name__ == '__main__':

test/test_aes.py

@@ -8,7 +8,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from youtube_dl.aes import aes_decrypt, aes_encrypt, aes_cbc_decrypt, aes_decrypt_text
+from youtube_dl.aes import aes_decrypt, aes_encrypt, aes_cbc_decrypt, aes_cbc_encrypt, aes_decrypt_text
 from youtube_dl.utils import bytes_to_intlist, intlist_to_bytes
 import base64
 
@@ -34,6 +34,13 @@ class TestAES(unittest.TestCase):
         decrypted = intlist_to_bytes(aes_cbc_decrypt(data, self.key, self.iv))
         self.assertEqual(decrypted.rstrip(b'\x08'), self.secret_msg)
 
+    def test_cbc_encrypt(self):
+        data = bytes_to_intlist(self.secret_msg)
+        encrypted = intlist_to_bytes(aes_cbc_encrypt(data, self.key, self.iv))
+        self.assertEqual(
+            encrypted,
+            b"\x97\x92+\xe5\x0b\xc3\x18\x91ky9m&\xb3\xb5@\xe6'\xc2\x96.\xc8u\x88\xab9-[\x9e|\xf1\xcd")
+
     def test_decrypt_text(self):
         password = intlist_to_bytes(self.key).decode('utf-8')
         encrypted = base64.b64encode(

test/test_compat.py

@@ -27,11 +27,11 @@ from youtube_dl.compat import (
 class TestCompat(unittest.TestCase):
     def test_compat_getenv(self):
         test_str = 'тест'
-        compat_setenv('YOUTUBE-DL-TEST', test_str)
-        self.assertEqual(compat_getenv('YOUTUBE-DL-TEST'), test_str)
+        compat_setenv('YOUTUBE_DL_COMPAT_GETENV', test_str)
+        self.assertEqual(compat_getenv('YOUTUBE_DL_COMPAT_GETENV'), test_str)
 
     def test_compat_setenv(self):
-        test_var = 'YOUTUBE-DL-TEST'
+        test_var = 'YOUTUBE_DL_COMPAT_SETENV'
         test_str = 'тест'
         compat_setenv(test_var, test_str)
         compat_getenv(test_var)

test/test_download.py

@@ -65,15 +65,31 @@ defs = gettestcases()
 
 
 class TestDownload(unittest.TestCase):
+    # Parallel testing in nosetests. See
+    # http://nose.readthedocs.org/en/latest/doc_tests/test_multiprocess/multiprocess.html
+    _multiprocess_shared_ = True
+
     maxDiff = None
 
+    def __str__(self):
+        """Identify each test with the `add_ie` attribute, if available."""
+
+        def strclass(cls):
+            """From 2.7's unittest; 2.6 had _strclass so we can't import it."""
+            return '%s.%s' % (cls.__module__, cls.__name__)
+
+        add_ie = getattr(self, self._testMethodName).add_ie
+        return '%s (%s)%s:' % (self._testMethodName,
+                               strclass(self.__class__),
+                               ' [%s]' % add_ie if add_ie else '')
+
     def setUp(self):
         self.defs = defs
 
 # Dynamically generate tests
 
 
-def generator(test_case):
+def generator(test_case, tname):
 
     def test_template(self):
         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
@@ -102,6 +118,7 @@ def generator(test_case):
                 return
 
         params = get_params(test_case.get('params', {}))
+        params['outtmpl'] = tname + '_' + params['outtmpl']
         if is_playlist and 'playlist' not in test_case:
             params.setdefault('extract_flat', 'in_playlist')
             params.setdefault('skip_download', True)
@@ -134,7 +151,7 @@ def generator(test_case):
             try_num = 1
             while True:
                 try:
-                    # We're not using .download here sine that is just a shim
+                    # We're not using .download here since that is just a shim
                     # for outside error handling, and returns the exit code
                     # instead of the result dict.
                     res_dict = ydl.extract_info(
@@ -146,7 +163,7 @@ def generator(test_case):
                         raise
 
                     if try_num == RETRIES:
-                        report_warning('Failed due to network errors, skipping...')
+                        report_warning('%s failed due to network errors, skipping...' % tname)
                         return
 
                     print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
@@ -182,7 +199,16 @@ def generator(test_case):
                 self.assertEqual(
                     test_case['playlist_duration_sum'], got_duration)
 
-            for tc in test_cases:
+            # Generalize both playlists and single videos to unified format for
+            # simplicity
+            if 'entries' not in res_dict:
+                res_dict['entries'] = [res_dict]
+
+            for tc_num, tc in enumerate(test_cases):
+                tc_res_dict = res_dict['entries'][tc_num]
+                # First, check test cases' data against extracted data alone
+                expect_info_dict(self, tc_res_dict, tc.get('info_dict', {}))
+                # Now, check downloaded file consistency
                 tc_filename = get_tc_filename(tc)
                 if not test_case.get('params', {}).get('skip_download', False):
                     self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
@@ -200,13 +226,14 @@ def generator(test_case):
                     if 'md5' in tc:
                         md5_for_file = _file_md5(tc_filename)
                         self.assertEqual(md5_for_file, tc['md5'])
+                # Finally, check test cases' data again but this time against
+                # extracted data from info JSON file written during processing
                 info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
                 self.assertTrue(
                     os.path.exists(info_json_fn),
                     'Missing info file %s' % info_json_fn)
                 with io.open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)
-
                 expect_info_dict(self, info_dict, tc.get('info_dict', {}))
         finally:
             try_rm_tcs_files()
@@ -221,13 +248,15 @@ def generator(test_case):
 
 # And add them to TestDownload
 for n, test_case in enumerate(defs):
-    test_method = generator(test_case)
     tname = 'test_' + str(test_case['name'])
     i = 1
     while hasattr(TestDownload, tname):
         tname = 'test_%s_%d' % (test_case['name'], i)
         i += 1
+    test_method = generator(test_case, tname)
     test_method.__name__ = str(tname)
+    ie_list = test_case.get('add_ie')
+    test_method.add_ie = ie_list and ','.join(ie_list)
     setattr(TestDownload, test_method.__name__, test_method)
     del test_method
 

test/test_subtitles.py

@@ -21,7 +21,7 @@ from youtube_dl.extractor import (
     NPOIE,
     ComedyCentralIE,
     NRKTVIE,
-    RaiTVIE,
+    RaiPlayIE,
     VikiIE,
     ThePlatformIE,
     ThePlatformFeedIE,
@@ -258,9 +258,9 @@ class TestNRKSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['no']), '544fa917d3197fcbee64634559221cc2')
 
 
-class TestRaiSubtitles(BaseTestSubtitles):
-    url = 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html'
-    IE = RaiTVIE
+class TestRaiPlaySubtitles(BaseTestSubtitles):
+    url = 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html'
+    IE = RaiPlayIE
 
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True

test/test_utils.py

@@ -34,6 +34,9 @@ from youtube_dl.utils import (
     find_xpath_attr,
     fix_xml_ampersands,
     get_element_by_class,
+    get_element_by_attribute,
+    get_elements_by_class,
+    get_elements_by_attribute,
     InAdvancePagedList,
     intlist_to_bytes,
     is_html,
@@ -49,9 +52,11 @@ from youtube_dl.utils import (
     parse_filesize,
     parse_count,
     parse_iso8601,
+    pkcs1pad,
     read_batch_urls,
     sanitize_filename,
     sanitize_path,
+    expand_path,
     prepend_extension,
     replace_extension,
     remove_start,
@@ -91,6 +96,8 @@ from youtube_dl.utils import (
 from youtube_dl.compat import (
     compat_chr,
     compat_etree_fromstring,
+    compat_getenv,
+    compat_setenv,
     compat_urlparse,
     compat_parse_qs,
 )
@@ -210,6 +217,18 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(sanitize_path('./abc'), 'abc')
         self.assertEqual(sanitize_path('./../abc'), '..\\abc')
 
+    def test_expand_path(self):
+        def env(var):
+            return '%{0}%'.format(var) if sys.platform == 'win32' else '${0}'.format(var)
+
+        compat_setenv('YOUTUBE_DL_EXPATH_PATH', 'expanded')
+        self.assertEqual(expand_path(env('YOUTUBE_DL_EXPATH_PATH')), 'expanded')
+        self.assertEqual(expand_path(env('HOME')), compat_getenv('HOME'))
+        self.assertEqual(expand_path('~'), compat_getenv('HOME'))
+        self.assertEqual(
+            expand_path('~/%s' % env('YOUTUBE_DL_EXPATH_PATH')),
+            '%s/expanded' % compat_getenv('HOME'))
+
     def test_prepend_extension(self):
         self.assertEqual(prepend_extension('abc.ext', 'temp'), 'abc.temp.ext')
         self.assertEqual(prepend_extension('abc.ext', 'temp', 'ext'), 'abc.temp.ext')
@@ -451,6 +470,9 @@ class TestUtil(unittest.TestCase):
 
     def test_urljoin(self):
         self.assertEqual(urljoin('http://foo.de/', '/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin(b'http://foo.de/', '/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('http://foo.de/', b'/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin(b'http://foo.de/', b'/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
         self.assertEqual(urljoin('//foo.de/', '/a/b/c.txt'), '//foo.de/a/b/c.txt')
         self.assertEqual(urljoin('http://foo.de/', 'a/b/c.txt'), 'http://foo.de/a/b/c.txt')
         self.assertEqual(urljoin('http://foo.de', '/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
@@ -1047,6 +1069,47 @@ The first line
 '''
         self.assertEqual(dfxp2srt(dfxp_data_no_default_namespace), srt_data)
 
+        dfxp_data_with_style = '''<?xml version="1.0" encoding="utf-8"?>
+<tt xmlns="http://www.w3.org/2006/10/ttaf1" xmlns:ttp="http://www.w3.org/2006/10/ttaf1#parameter" ttp:timeBase="media" xmlns:tts="http://www.w3.org/2006/10/ttaf1#style" xml:lang="en" xmlns:ttm="http://www.w3.org/2006/10/ttaf1#metadata">
+  <head>
+    <styling>
+      <style id="s2" style="s0" tts:color="cyan" tts:fontWeight="bold" />
+      <style id="s1" style="s0" tts:color="yellow" tts:fontStyle="italic" />
+      <style id="s3" style="s0" tts:color="lime" tts:textDecoration="underline" />
+      <style id="s0" tts:backgroundColor="black" tts:fontStyle="normal" tts:fontSize="16" tts:fontFamily="sansSerif" tts:color="white" />
+    </styling>
+  </head>
+  <body tts:textAlign="center" style="s0">
+    <div>
+      <p begin="00:00:02.08" id="p0" end="00:00:05.84">default style<span tts:color="red">custom style</span></p>
+      <p style="s2" begin="00:00:02.08" id="p0" end="00:00:05.84"><span tts:color="lime">part 1<br /></span><span tts:color="cyan">part 2</span></p>
+      <p style="s3" begin="00:00:05.84" id="p1" end="00:00:09.56">line 3<br />part 3</p>
+      <p style="s1" tts:textDecoration="underline" begin="00:00:09.56" id="p2" end="00:00:12.36"><span style="s2" tts:color="lime">inner<br /> </span>style</p>
+    </div>
+  </body>
+</tt>'''
+        srt_data = '''1
+00:00:02,080 --> 00:00:05,839
+<font color="white" face="sansSerif" size="16">default style<font color="red">custom style</font></font>
+
+2
+00:00:02,080 --> 00:00:05,839
+<b><font color="cyan" face="sansSerif" size="16"><font color="lime">part 1
+</font>part 2</font></b>
+
+3
+00:00:05,839 --> 00:00:09,560
+<u><font color="lime">line 3
+part 3</font></u>
+
+4
+00:00:09,560 --> 00:00:12,359
+<i><u><font color="yellow"><font color="lime">inner
+ </font>style</font></u></i>
+
+'''
+        self.assertEqual(dfxp2srt(dfxp_data_with_style), srt_data)
+
     def test_cli_option(self):
         self.assertEqual(cli_option({'proxy': '127.0.0.1:3128'}, '--proxy', 'proxy'), ['--proxy', '127.0.0.1:3128'])
         self.assertEqual(cli_option({'proxy': None}, '--proxy', 'proxy'), [])
@@ -1101,6 +1164,14 @@ The first line
             ohdave_rsa_encrypt(b'aa111222', e, N),
             '726664bd9a23fd0c70f9f1b84aab5e3905ce1e45a584e9cbcf9bcc7510338fc1986d6c599ff990d923aa43c51c0d9013cd572e13bc58f4ae48f2ed8c0b0ba881')
 
+    def test_pkcs1pad(self):
+        data = [1, 2, 3]
+        padded_data = pkcs1pad(data, 32)
+        self.assertEqual(padded_data[:2], [0, 2])
+        self.assertEqual(padded_data[28:], [0, 1, 2, 3])
+
+        self.assertRaises(ValueError, pkcs1pad, data, 8)
+
     def test_encode_base_n(self):
         self.assertEqual(encode_base_n(0, 30), '0')
         self.assertEqual(encode_base_n(80, 30), '2k')
@@ -1124,6 +1195,32 @@ The first line
         self.assertEqual(get_element_by_class('foo', html), 'nice')
         self.assertEqual(get_element_by_class('no-such-class', html), None)
 
+    def test_get_element_by_attribute(self):
+        html = '''
+            <span class="foo bar">nice</span>
+        '''
+
+        self.assertEqual(get_element_by_attribute('class', 'foo bar', html), 'nice')
+        self.assertEqual(get_element_by_attribute('class', 'foo', html), None)
+        self.assertEqual(get_element_by_attribute('class', 'no-such-foo', html), None)
+
+    def test_get_elements_by_class(self):
+        html = '''
+            <span class="foo bar">nice</span><span class="foo bar">also nice</span>
+        '''
+
+        self.assertEqual(get_elements_by_class('foo', html), ['nice', 'also nice'])
+        self.assertEqual(get_elements_by_class('no-such-class', html), [])
+
+    def test_get_elements_by_attribute(self):
+        html = '''
+            <span class="foo bar">nice</span><span class="foo bar">also nice</span>
+        '''
+
+        self.assertEqual(get_elements_by_attribute('class', 'foo bar', html), ['nice', 'also nice'])
+        self.assertEqual(get_elements_by_attribute('class', 'foo', html), [])
+        self.assertEqual(get_elements_by_attribute('class', 'no-such-foo', html), [])
+
 
 if __name__ == '__main__':
     unittest.main()

test/testdata/m3u8/pluzz_francetv_11507.m3u8

@@ -0,0 +1,14 @@
+#EXTM3U
+    
+#EXT-X-VERSION:5
+    
+#EXT-X-MEDIA:TYPE=SUBTITLES,GROUP-ID="subs",NAME="Francais",DEFAULT=NO,FORCED=NO,URI="http://replayftv-pmd.francetv.fr/subtitles/2017/16/156589847-1492488987.m3u8",LANGUAGE="fra"
+    
+#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID="aac",LANGUAGE="fra",NAME="Francais",DEFAULT=YES, AUTOSELECT=YES
+#EXT-X-STREAM-INF:SUBTITLES="subs",AUDIO="aac",PROGRAM-ID=1,BANDWIDTH=180000,RESOLUTION=256x144,CODECS="avc1.66.30, mp4a.40.2"
+http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_0_av.m3u8?null=0
+#EXT-X-STREAM-INF:SUBTITLES="subs",AUDIO="aac",PROGRAM-ID=1,BANDWIDTH=303000,RESOLUTION=320x180,CODECS="avc1.66.30, mp4a.40.2"
+http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_1_av.m3u8?null=0
+#EXT-X-STREAM-INF:SUBTITLES="subs",AUDIO="aac",PROGRAM-ID=1,BANDWIDTH=575000,RESOLUTION=512x288,CODECS="avc1.66.30, mp4a.40.2"
+http://replayftv-vh.akamaihd.net/i/streaming-adaptatif_france-dom-tom/2017/S16/J2/156589847-58f59130c1f52-,standard1,standard2,standard3,standard4,standard5,.mp4.csmil/index_2_av.m3u8?null=0
+#EXT-X-STREAM-INF:SUBTITLES="subs",AUDIO="aac",PROGRAM-ID=1,BANDWIDTH=831000,RESOLUTION=704x396,CODECS="avc1.77.30, mp4a.40.2"

test/testdata/m3u8/teamcoco_11995.m3u8

@@ -0,0 +1,16 @@
+#EXTM3U
+#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID="audio-0",NAME="Default",AUTOSELECT=YES,DEFAULT=YES,URI="hls/CONAN_020217_Highlight_show-audio-160k_v4.m3u8"
+#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID="audio-1",NAME="Default",AUTOSELECT=YES,DEFAULT=YES,URI="hls/CONAN_020217_Highlight_show-audio-64k_v4.m3u8"
+#EXT-X-I-FRAME-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=37862000,CODECS="avc1.4d001f",URI="hls/CONAN_020217_Highlight_show-2m_iframe.m3u8"
+#EXT-X-I-FRAME-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=18750000,CODECS="avc1.4d001e",URI="hls/CONAN_020217_Highlight_show-1m_iframe.m3u8"
+#EXT-X-I-FRAME-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=6535000,CODECS="avc1.42001e",URI="hls/CONAN_020217_Highlight_show-400k_iframe.m3u8"
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=2374000,RESOLUTION=1024x576,CODECS="avc1.4d001f,mp4a.40.2",AUDIO="audio-0"
+hls/CONAN_020217_Highlight_show-2m_v4.m3u8
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=1205000,RESOLUTION=640x360,CODECS="avc1.4d001e,mp4a.40.2",AUDIO="audio-0"
+hls/CONAN_020217_Highlight_show-1m_v4.m3u8
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=522000,RESOLUTION=400x224,CODECS="avc1.42001e,mp4a.40.2",AUDIO="audio-0"
+hls/CONAN_020217_Highlight_show-400k_v4.m3u8
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=413000,RESOLUTION=400x224,CODECS="avc1.42001e,mp4a.40.5",AUDIO="audio-1"
+hls/CONAN_020217_Highlight_show-400k_v4.m3u8
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=71000,CODECS="mp4a.40.5",AUDIO="audio-1"
+hls/CONAN_020217_Highlight_show-audio-64k_v4.m3u8

test/testdata/m3u8/toggle_mobile_12211.m3u8

@@ -0,0 +1,13 @@
+#EXTM3U
+#EXT-X-VERSION:4
+#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID="audio",LANGUAGE="eng",NAME="English",URI="http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_sa2ntrdg/name/a.mp4/index.m3u8"
+#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID="audio",LANGUAGE="und",NAME="Undefined",URI="http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_r7y0nitg/name/a.mp4/index.m3u8"
+
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=155648,RESOLUTION=320x180,AUDIO="audio"
+http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_qlk9hlzr/name/a.mp4/index.m3u8
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=502784,RESOLUTION=480x270,AUDIO="audio"
+http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/2/pv/1/flavorId/0_oefackmi/name/a.mp4/index.m3u8
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=827392,RESOLUTION=640x360,AUDIO="audio"
+http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/12/pv/1/flavorId/0_vyg9pj7k/name/a.mp4/index.m3u8
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=1396736,RESOLUTION=854x480,AUDIO="audio"
+http://k.toggle.sg/fhls/p/2082311/sp/208231100/serveFlavor/entryId/0_89q6e8ku/v/12/pv/1/flavorId/0_50n4psvx/name/a.mp4/index.m3u8

test/testdata/m3u8/twitch_vod.m3u8

@@ -0,0 +1,20 @@
+#EXTM3U
+#EXT-X-TWITCH-INFO:ORIGIN="s3",CLUSTER="edgecast_vod",REGION="EU",MANIFEST-CLUSTER="edgecast_vod",USER-IP="109.171.17.81"
+#EXT-X-MEDIA:TYPE=VIDEO,GROUP-ID="chunked",NAME="Source",AUTOSELECT=YES,DEFAULT=YES
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=3214134,CODECS="avc1.100.31,mp4a.40.2",RESOLUTION="1280x720",VIDEO="chunked"
+https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/chunked/index-muted-HM49I092CC.m3u8
+#EXT-X-MEDIA:TYPE=VIDEO,GROUP-ID="high",NAME="High",AUTOSELECT=YES,DEFAULT=YES
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=1603789,CODECS="avc1.42C01F,mp4a.40.2",RESOLUTION="1280x720",VIDEO="high"
+https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/high/index-muted-HM49I092CC.m3u8
+#EXT-X-MEDIA:TYPE=VIDEO,GROUP-ID="medium",NAME="Medium",AUTOSELECT=YES,DEFAULT=YES
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=893387,CODECS="avc1.42C01E,mp4a.40.2",RESOLUTION="852x480",VIDEO="medium"
+https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/medium/index-muted-HM49I092CC.m3u8
+#EXT-X-MEDIA:TYPE=VIDEO,GROUP-ID="low",NAME="Low",AUTOSELECT=YES,DEFAULT=YES
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=628347,CODECS="avc1.42C01E,mp4a.40.2",RESOLUTION="640x360",VIDEO="low"
+https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/low/index-muted-HM49I092CC.m3u8
+#EXT-X-MEDIA:TYPE=VIDEO,GROUP-ID="mobile",NAME="Mobile",AUTOSELECT=YES,DEFAULT=YES
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=280474,CODECS="avc1.42C00D,mp4a.40.2",RESOLUTION="400x226",VIDEO="mobile"
+https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/mobile/index-muted-HM49I092CC.m3u8
+#EXT-X-MEDIA:TYPE=VIDEO,GROUP-ID="audio_only",NAME="Audio Only",AUTOSELECT=NO,DEFAULT=NO
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=182725,CODECS="mp4a.40.2",VIDEO="audio_only"
+https://vod.edgecast.hls.ttvnw.net/e5da31ab49_riotgames_15001215120_261543898/audio_only/index-muted-HM49I092CC.m3u8

test/testdata/m3u8/vidio.m3u8

@@ -0,0 +1,10 @@
+#EXTM3U
+
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=300000,RESOLUTION=480x270,NAME="270p 3G"
+https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b300.mp4.m3u8
+
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=600000,RESOLUTION=640x360,NAME="360p SD"
+https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b600.mp4.m3u8
+
+#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=1200000,RESOLUTION=1280x720,NAME="720p HD"
+https://cdn1-a.production.vidio.static6.com/uploads/165683/dj_ambred-4383-b1200.mp4.m3u8

youtube_dl/YoutubeDL.py

@@ -29,10 +29,10 @@ import random
 from .compat import (
     compat_basestring,
     compat_cookiejar,
-    compat_expanduser,
     compat_get_terminal_size,
     compat_http_client,
     compat_kwargs,
+    compat_numeric_types,
     compat_os_name,
     compat_str,
     compat_tokenize_tokenize,
@@ -53,9 +53,12 @@ from .utils import (
     encode_compat_str,
     encodeFilename,
     error_to_compat_str,
+    expand_path,
     ExtractorError,
     format_bytes,
     formatSeconds,
+    GeoRestrictedError,
+    ISO3166Utils,
     locked_file,
     make_HTTPS_handler,
     MaxDownloadsReached,
@@ -272,6 +275,12 @@ class YoutubeDL(object):
                        If it returns None, the video is downloaded.
                        match_filter_func in utils.py is one example for this.
     no_color:          Do not emit color codes in output.
+    geo_bypass:        Bypass geographic restriction via faking X-Forwarded-For
+                       HTTP header (experimental)
+    geo_bypass_country:
+                       Two-letter ISO 3166-2 country code that will be used for
+                       explicit geographic restriction bypassing via faking
+                       X-Forwarded-For HTTP header (experimental)
 
     The following options determine which downloader is picked:
     external_downloader: Executable of the external downloader to call.
@@ -319,11 +328,21 @@ class YoutubeDL(object):
         self.params.update(params)
         self.cache = Cache(self)
 
-        if self.params.get('cn_verification_proxy') is not None:
-            self.report_warning('--cn-verification-proxy is deprecated. Use --geo-verification-proxy instead.')
+        def check_deprecated(param, option, suggestion):
+            if self.params.get(param) is not None:
+                self.report_warning(
+                    '%s is deprecated. Use %s instead.' % (option, suggestion))
+                return True
+            return False
+
+        if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
             if self.params.get('geo_verification_proxy') is None:
                 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
 
+        check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
+        check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
+        check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
+
         if params.get('bidi_workaround', False):
             try:
                 import pty
@@ -585,10 +604,7 @@ class YoutubeDL(object):
             autonumber_size = self.params.get('autonumber_size')
             if autonumber_size is None:
                 autonumber_size = 5
-            autonumber_templ = '%0' + str(autonumber_size) + 'd'
-            template_dict['autonumber'] = autonumber_templ % (self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
-            if template_dict.get('playlist_index') is not None:
-                template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
+            template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
             if template_dict.get('resolution') is None:
                 if template_dict.get('width') and template_dict.get('height'):
                     template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
@@ -600,15 +616,63 @@ class YoutubeDL(object):
             sanitize = lambda k, v: sanitize_filename(
                 compat_str(v),
                 restricted=self.params.get('restrictfilenames'),
-                is_id=(k == 'id'))
-            template_dict = dict((k, sanitize(k, v))
+                is_id=(k == 'id' or k.endswith('_id')))
+            template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v))
                                  for k, v in template_dict.items()
                                  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)
-            tmpl = compat_expanduser(outtmpl)
-            filename = tmpl % template_dict
+
+            # For fields playlist_index and autonumber convert all occurrences
+            # of %(field)s to %(field)0Nd for backward compatibility
+            field_size_compat_map = {
+                'playlist_index': len(str(template_dict['n_entries'])),
+                'autonumber': autonumber_size,
+            }
+            FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
+            mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl)
+            if mobj:
+                outtmpl = re.sub(
+                    FIELD_SIZE_COMPAT_RE,
+                    r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')],
+                    outtmpl)
+
+            NUMERIC_FIELDS = set((
+                'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
+                'timestamp', 'upload_year', 'upload_month', 'upload_day',
+                'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
+                'average_rating', 'comment_count', 'age_limit',
+                'start_time', 'end_time',
+                'chapter_number', 'season_number', 'episode_number',
+                'track_number', 'disc_number', 'release_year',
+                'playlist_index',
+            ))
+
+            # Missing numeric fields used together with integer presentation types
+            # in format specification will break the argument substitution since
+            # string 'NA' is returned for missing fields. We will patch output
+            # template for missing fields to meet string presentation type.
+            for numeric_field in NUMERIC_FIELDS:
+                if numeric_field not in template_dict:
+                    # As of [1] format syntax is:
+                    #  %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
+                    # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
+                    FORMAT_RE = r'''(?x)
+                        (?<!%)
+                        %
+                        \({0}\)  # mapping key
+                        (?:[#0\-+ ]+)?  # conversion flags (optional)
+                        (?:\d+)?  # minimum field width (optional)
+                        (?:\.\d+)?  # precision (optional)
+                        [hlL]?  # length modifier (optional)
+                        [diouxXeEfFgGcrs%]  # conversion type
+                    '''
+                    outtmpl = re.sub(
+                        FORMAT_RE.format(numeric_field),
+                        r'%({0})s'.format(numeric_field), outtmpl)
+
+            filename = expand_path(outtmpl % template_dict)
             # Temporary fix for #4787
             # 'Treat' all problem characters by passing filename through preferredencoding
             # to workaround encoding issues with subprocess on python2 @ Windows
@@ -707,6 +771,14 @@ class YoutubeDL(object):
                     return self.process_ie_result(ie_result, download, extra_info)
                 else:
                     return ie_result
+            except GeoRestrictedError as e:
+                msg = e.msg
+                if e.countries:
+                    msg += '\nThis video is available in %s.' % ', '.join(
+                        map(ISO3166Utils.short2full, e.countries))
+                msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
+                self.report_error(msg)
+                break
             except ExtractorError as e:  # An error we somewhat expected
                 self.report_error(compat_str(e), e.format_traceback())
                 break
@@ -764,6 +836,12 @@ class YoutubeDL(object):
                 ie_result['url'], ie_key=ie_result.get('ie_key'),
                 extra_info=extra_info, download=False, process=False)
 
+            # extract_info may return None when ignoreerrors is enabled and
+            # extraction failed with an error, don't crash and return early
+            # in this case
+            if not info:
+                return info
+
             force_properties = dict(
                 (k, v) for k, v in ie_result.items() if v is not None)
             for f in ('_type', 'url', 'ie_key'):
@@ -772,11 +850,18 @@ class YoutubeDL(object):
             new_result = info.copy()
             new_result.update(force_properties)
 
-            assert new_result.get('_type') != 'url_transparent'
+            # Extracted info may not be a video result (i.e.
+            # info.get('_type', 'video') != video) but rather an url or
+            # url_transparent. In such cases outer metadata (from ie_result)
+            # should be propagated to inner one (info). For this to happen
+            # _type of info should be overridden with url_transparent. This
+            # fixes issue from https://github.com/rg3/youtube-dl/pull/11163.
+            if new_result.get('_type') == 'url':
+                new_result['_type'] = 'url_transparent'
 
             return self.process_ie_result(
                 new_result, download=download, extra_info=extra_info)
-        elif result_type == 'playlist' or result_type == 'multi_video':
+        elif result_type in ('playlist', 'multi_video'):
             # We process each entry in the playlist
             playlist = ie_result.get('title') or ie_result.get('id')
             self.to_screen('[download] Downloading playlist: %s' % playlist)
@@ -847,8 +932,14 @@ class YoutubeDL(object):
             if self.params.get('playlistrandom', False):
                 random.shuffle(entries)
 
+            x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
+
             for i, entry in enumerate(entries, 1):
                 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
+                # This __x_forwarded_for_ip thing is a bit ugly but requires
+                # minimal changes
+                if x_forwarded_for:
+                    entry['__x_forwarded_for_ip'] = x_forwarded_for
                 extra = {
                     'n_entries': n_entries,
                     'playlist': playlist,
@@ -1233,6 +1324,11 @@ class YoutubeDL(object):
         if cookies:
             res['Cookie'] = cookies
 
+        if 'X-Forwarded-For' not in res:
+            x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
+            if x_forwarded_for_ip:
+                res['X-Forwarded-For'] = x_forwarded_for_ip
+
         return res
 
     def _calc_cookies(self, info_dict):
@@ -1375,6 +1471,9 @@ class YoutubeDL(object):
             full_format_info = info_dict.copy()
             full_format_info.update(format)
             format['http_headers'] = self._calc_headers(full_format_info)
+        # Remove private housekeeping stuff
+        if '__x_forwarded_for_ip' in info_dict:
+            del info_dict['__x_forwarded_for_ip']
 
         # TODO Central sorting goes here
 
@@ -1785,6 +1884,7 @@ class YoutubeDL(object):
         """Download a given list of URLs."""
         outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
         if (len(url_list) > 1 and
+                outtmpl != '-' and
                 '%' not in outtmpl and
                 self.params.get('max_downloads') != 1):
             raise SameFileError(outtmpl)
@@ -2082,7 +2182,7 @@ class YoutubeDL(object):
         if opts_cookiefile is None:
             self.cookiejar = compat_cookiejar.CookieJar()
         else:
-            opts_cookiefile = compat_expanduser(opts_cookiefile)
+            opts_cookiefile = expand_path(opts_cookiefile)
             self.cookiejar = compat_cookiejar.MozillaCookieJar(
                 opts_cookiefile)
             if os.access(opts_cookiefile, os.R_OK):

youtube_dl/__init__.py

@@ -16,7 +16,6 @@ from .options import (
     parseOpts,
 )
 from .compat import (
-    compat_expanduser,
     compat_getpass,
     compat_shlex_split,
     workaround_optparse_bug9161,
@@ -26,6 +25,7 @@ from .utils import (
     decodeOption,
     DEFAULT_OUTTMPL,
     DownloadError,
+    expand_path,
     match_filter_func,
     MaxDownloadsReached,
     preferredencoding,
@@ -88,7 +88,7 @@ def _real_main(argv=None):
                 batchfd = sys.stdin
             else:
                 batchfd = io.open(
-                    compat_expanduser(opts.batchfile),
+                    expand_path(opts.batchfile),
                     'r', encoding='utf-8', errors='ignore')
             batch_urls = read_batch_urls(batchfd)
             if opts.verbose:
@@ -196,7 +196,7 @@ def _real_main(argv=None):
     if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
         raise ValueError('Playlist end must be greater than playlist start')
     if opts.extractaudio:
-        if opts.audioformat not in ['best', 'aac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']:
+        if opts.audioformat not in ['best', 'aac', 'flac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']:
             parser.error('invalid audio format specified')
     if opts.audioquality:
         opts.audioquality = opts.audioquality.strip('k').strip('K')
@@ -238,18 +238,15 @@ def _real_main(argv=None):
 
     any_getting = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
     any_printing = opts.print_json
-    download_archive_fn = compat_expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive
+    download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive
 
     # PostProcessors
     postprocessors = []
-    # Add the metadata pp first, the other pps will copy it
     if opts.metafromtitle:
         postprocessors.append({
             'key': 'MetadataFromTitle',
             'titleformat': opts.metafromtitle
         })
-    if opts.addmetadata:
-        postprocessors.append({'key': 'FFmpegMetadata'})
     if opts.extractaudio:
         postprocessors.append({
             'key': 'FFmpegExtractAudio',
@@ -262,6 +259,16 @@ def _real_main(argv=None):
             'key': 'FFmpegVideoConvertor',
             'preferedformat': opts.recodevideo,
         })
+    # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
+    # FFmpegExtractAudioPP as containers before conversion may not support
+    # metadata (3gp, webm, etc.)
+    # And this post-processor should be placed before other metadata
+    # manipulating post-processors (FFmpegEmbedSubtitle) to prevent loss of
+    # extra metadata. By default ffmpeg preserves metadata applicable for both
+    # source and target containers. From this point the container won't change,
+    # so metadata can be added here.
+    if opts.addmetadata:
+        postprocessors.append({'key': 'FFmpegMetadata'})
     if opts.convertsubtitles:
         postprocessors.append({
             'key': 'FFmpegSubtitlesConvertor',
@@ -336,6 +343,7 @@ def _real_main(argv=None):
         'retries': opts.retries,
         'fragment_retries': opts.fragment_retries,
         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
+        'keep_fragments': opts.keep_fragments,
         'buffersize': opts.buffersize,
         'noresizebuffer': opts.noresizebuffer,
         'continuedl': opts.continue_dl,
@@ -414,6 +422,11 @@ def _real_main(argv=None):
         'cn_verification_proxy': opts.cn_verification_proxy,
         'geo_verification_proxy': opts.geo_verification_proxy,
         'config_location': opts.config_location,
+        'geo_bypass': opts.geo_bypass,
+        'geo_bypass_country': opts.geo_bypass_country,
+        # just for deprecation check
+        'autonumber': opts.autonumber if opts.autonumber is True else None,
+        'usetitle': opts.usetitle if opts.usetitle is True else None,
     }
 
     with YoutubeDL(ydl_opts) as ydl:
@@ -437,7 +450,7 @@ def _real_main(argv=None):
 
         try:
             if opts.load_info_filename is not None:
-                retcode = ydl.download_with_info_file(compat_expanduser(opts.load_info_filename))
+                retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
             else:
                 retcode = ydl.download(all_urls)
         except MaxDownloadsReached:

youtube_dl/aes.py

@@ -60,6 +60,34 @@ def aes_cbc_decrypt(data, key, iv):
     return decrypted_data
 
 
+def aes_cbc_encrypt(data, key, iv):
+    """
+    Encrypt with aes in CBC mode. Using PKCS#7 padding
+
+    @param {int[]} data        cleartext
+    @param {int[]} key         16/24/32-Byte cipher key
+    @param {int[]} iv          16-Byte IV
+    @returns {int[]}           encrypted data
+    """
+    expanded_key = key_expansion(key)
+    block_count = int(ceil(float(len(data)) / BLOCK_SIZE_BYTES))
+
+    encrypted_data = []
+    previous_cipher_block = iv
+    for i in range(block_count):
+        block = data[i * BLOCK_SIZE_BYTES: (i + 1) * BLOCK_SIZE_BYTES]
+        remaining_length = BLOCK_SIZE_BYTES - len(block)
+        block += [remaining_length] * remaining_length
+        mixed_block = xor(block, previous_cipher_block)
+
+        encrypted_block = aes_encrypt(mixed_block, expanded_key)
+        encrypted_data += encrypted_block
+
+        previous_cipher_block = encrypted_block
+
+    return encrypted_data
+
+
 def key_expansion(data):
     """
     Generate key schedule

youtube_dl/cache.py

@@ -8,8 +8,11 @@ import re
 import shutil
 import traceback
 
-from .compat import compat_expanduser, compat_getenv
-from .utils import write_json_file
+from .compat import compat_getenv
+from .utils import (
+    expand_path,
+    write_json_file,
+)
 
 
 class Cache(object):
@@ -21,7 +24,7 @@ class Cache(object):
         if res is None:
             cache_root = compat_getenv('XDG_CACHE_HOME', '~/.cache')
             res = os.path.join(cache_root, 'youtube-dl')
-        return compat_expanduser(res)
+        return expand_path(res)
 
     def _get_cache_fn(self, section, key, dtype):
         assert re.match(r'^[a-zA-Z0-9_.-]+$', section), \

youtube_dl/compat.py

@@ -2692,7 +2692,7 @@ else:
                 userhome = pwent.pw_dir
             userhome = userhome.rstrip('/')
             return (userhome + path[i:]) or '/'
-    elif compat_os_name == 'nt' or compat_os_name == 'ce':
+    elif compat_os_name in ('nt', 'ce'):
         def compat_expanduser(path):
             """Expand ~ and ~user constructs.
 
@@ -2760,6 +2760,12 @@ else:
     compat_kwargs = lambda kwargs: kwargs
 
 
+try:
+    compat_numeric_types = (int, float, long, complex)
+except NameError:  # Python 3
+    compat_numeric_types = (int, float, complex)
+
+
 if sys.version_info < (2, 7):
     def compat_socket_create_connection(address, timeout, source_address=None):
         host, port = address
@@ -2895,6 +2901,7 @@ __all__ = [
     'compat_input',
     'compat_itertools_count',
     'compat_kwargs',
+    'compat_numeric_types',
     'compat_ord',
     'compat_os_name',
     'compat_parse_qs',

youtube_dl/downloader/__init__.py

@@ -43,6 +43,9 @@ def get_suitable_downloader(info_dict, params={}):
         if ed.can_download(info_dict):
             return ed
 
+    if protocol.startswith('m3u8') and info_dict.get('is_live'):
+        return FFmpegFD
+
     if protocol == 'm3u8' and params.get('hls_prefer_native') is True:
         return HlsFD
 

youtube_dl/downloader/common.py

@@ -187,6 +187,9 @@ class FileDownloader(object):
             return filename[:-len('.part')]
         return filename
 
+    def ytdl_filename(self, filename):
+        return filename + '.ytdl'
+
     def try_rename(self, old_filename, new_filename):
         try:
             if old_filename == new_filename:
@@ -327,27 +330,31 @@ class FileDownloader(object):
             os.path.exists(encodeFilename(filename))
         )
 
-        continuedl_and_exists = (
-            self.params.get('continuedl', True) and
-            os.path.isfile(encodeFilename(filename)) and
-            not self.params.get('nopart', False)
-        )
+        if not hasattr(filename, 'write'):
+            continuedl_and_exists = (
+                self.params.get('continuedl', True) and
+                os.path.isfile(encodeFilename(filename)) and
+                not self.params.get('nopart', False)
+            )
 
-        # Check file already present
-        if filename != '-' and (nooverwrites_and_exists or continuedl_and_exists):
-            self.report_file_already_downloaded(filename)
-            self._hook_progress({
-                'filename': filename,
-                'status': 'finished',
-                'total_bytes': os.path.getsize(encodeFilename(filename)),
-            })
-            return True
+            # Check file already present
+            if filename != '-' and (nooverwrites_and_exists or continuedl_and_exists):
+                self.report_file_already_downloaded(filename)
+                self._hook_progress({
+                    'filename': filename,
+                    'status': 'finished',
+                    'total_bytes': os.path.getsize(encodeFilename(filename)),
+                })
+                return True
 
         min_sleep_interval = self.params.get('sleep_interval')
         if min_sleep_interval:
             max_sleep_interval = self.params.get('max_sleep_interval', min_sleep_interval)
             sleep_interval = random.uniform(min_sleep_interval, max_sleep_interval)
-            self.to_screen('[download] Sleeping %s seconds...' % sleep_interval)
+            self.to_screen(
+                '[download] Sleeping %s seconds...' % (
+                    int(sleep_interval) if sleep_interval.is_integer()
+                    else '%.2f' % sleep_interval))
             time.sleep(sleep_interval)
 
         return self.real_download(filename, info_dict)

youtube_dl/downloader/dash.py

@@ -1,13 +1,7 @@
 from __future__ import unicode_literals
 
-import os
-
 from .fragment import FragmentFD
 from ..compat import compat_urllib_error
-from ..utils import (
-    sanitize_open,
-    encodeFilename,
-)
 
 
 class DashSegmentsFD(FragmentFD):
@@ -28,28 +22,24 @@ class DashSegmentsFD(FragmentFD):
 
         self._prepare_and_start_frag_download(ctx)
 
-        segments_filenames = []
-
         fragment_retries = self.params.get('fragment_retries', 0)
         skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
 
-        def process_segment(segment, tmp_filename, num):
-            segment_url = segment['url']
-            segment_name = 'Frag%d' % num
-            target_filename = '%s-%s' % (tmp_filename, segment_name)
+        frag_index = 0
+        for i, segment in enumerate(segments):
+            frag_index += 1
+            if frag_index <= ctx['fragment_index']:
+                continue
             # In DASH, the first segment contains necessary headers to
             # generate a valid MP4 file, so always abort for the first segment
-            fatal = num == 0 or not skip_unavailable_fragments
+            fatal = i == 0 or not skip_unavailable_fragments
             count = 0
             while count <= fragment_retries:
                 try:
-                    success = ctx['dl'].download(target_filename, {'url': segment_url})
+                    success, frag_content = self._download_fragment(ctx, segment['url'], info_dict)
                     if not success:
                         return False
-                    down, target_sanitized = sanitize_open(target_filename, 'rb')
-                    ctx['dest_stream'].write(down.read())
-                    down.close()
-                    segments_filenames.append(target_sanitized)
+                    self._append_fragment(ctx, frag_content)
                     break
                 except compat_urllib_error.HTTPError as err:
                     # YouTube may often return 404 HTTP error for a fragment causing the
@@ -60,22 +50,14 @@ class DashSegmentsFD(FragmentFD):
                     # HTTP error.
                     count += 1
                     if count <= fragment_retries:
-                        self.report_retry_fragment(err, segment_name, count, fragment_retries)
+                        self.report_retry_fragment(err, frag_index, count, fragment_retries)
             if count > fragment_retries:
                 if not fatal:
-                    self.report_skip_fragment(segment_name)
-                    return True
+                    self.report_skip_fragment(frag_index)
+                    continue
                 self.report_error('giving up after %s fragment retries' % fragment_retries)
                 return False
-            return True
-
-        for i, segment in enumerate(segments):
-            if not process_segment(segment, ctx['tmpfilename'], i):
-                return False
 
         self._finish_frag_download(ctx)
 
-        for segment_file in segments_filenames:
-            os.remove(encodeFilename(segment_file))
-
         return True

youtube_dl/downloader/external.py

@@ -6,7 +6,10 @@ import sys
 import re
 
 from .common import FileDownloader
-from ..compat import compat_setenv
+from ..compat import (
+    compat_setenv,
+    compat_str,
+)
 from ..postprocessor.ffmpeg import FFmpegPostProcessor, EXT_TO_OUT_FORMATS
 from ..utils import (
     cli_option,
@@ -270,12 +273,16 @@ class FFmpegFD(ExternalFD):
                 args += ['-rtmp_live', 'live']
 
         args += ['-i', url, '-c', 'copy']
+
+        if self.params.get('test', False):
+            args += ['-fs', compat_str(self._TEST_FILE_SIZE)]
+
         if protocol in ('m3u8', 'm3u8_native'):
             if self.params.get('hls_use_mpegts', False) or tmpfilename == '-':
                 args += ['-f', 'mpegts']
             else:
                 args += ['-f', 'mp4']
-                if (ffpp.basename == 'ffmpeg' and is_outdated_version(ffpp._versions['ffmpeg'], '3.2')) and (not info_dict.get('acodec') or info_dict['acodec'].split('.')[0] in ('aac', 'mp4a')):
+                if (ffpp.basename == 'ffmpeg' and is_outdated_version(ffpp._versions['ffmpeg'], '3.2', False)) and (not info_dict.get('acodec') or info_dict['acodec'].split('.')[0] in ('aac', 'mp4a')):
                     args += ['-bsf:a', 'aac_adtstoasc']
         elif protocol == 'rtmp':
             args += ['-f', 'flv']

youtube_dl/downloader/f4m.py

@@ -3,7 +3,6 @@ from __future__ import division, unicode_literals
 import base64
 import io
 import itertools
-import os
 import time
 
 from .fragment import FragmentFD
@@ -16,9 +15,7 @@ from ..compat import (
     compat_struct_unpack,
 )
 from ..utils import (
-    encodeFilename,
     fix_xml_ampersands,
-    sanitize_open,
     xpath_text,
 )
 
@@ -366,17 +363,21 @@ class F4mFD(FragmentFD):
 
         dest_stream = ctx['dest_stream']
 
-        write_flv_header(dest_stream)
-        if not live:
-            write_metadata_tag(dest_stream, metadata)
+        if ctx['complete_frags_downloaded_bytes'] == 0:
+            write_flv_header(dest_stream)
+            if not live:
+                write_metadata_tag(dest_stream, metadata)
 
         base_url_parsed = compat_urllib_parse_urlparse(base_url)
 
         self._start_frag_download(ctx)
 
-        frags_filenames = []
+        frag_index = 0
         while fragments_list:
             seg_i, frag_i = fragments_list.pop(0)
+            frag_index += 1
+            if frag_index <= ctx['fragment_index']:
+                continue
             name = 'Seg%d-Frag%d' % (seg_i, frag_i)
             query = []
             if base_url_parsed.query:
@@ -386,17 +387,10 @@ class F4mFD(FragmentFD):
             if info_dict.get('extra_param_to_segment_url'):
                 query.append(info_dict['extra_param_to_segment_url'])
             url_parsed = base_url_parsed._replace(path=base_url_parsed.path + name, query='&'.join(query))
-            frag_filename = '%s-%s' % (ctx['tmpfilename'], name)
             try:
-                success = ctx['dl'].download(frag_filename, {
-                    'url': url_parsed.geturl(),
-                    'http_headers': info_dict.get('http_headers'),
-                })
+                success, down_data = self._download_fragment(ctx, url_parsed.geturl(), info_dict)
                 if not success:
                     return False
-                (down, frag_sanitized) = sanitize_open(frag_filename, 'rb')
-                down_data = down.read()
-                down.close()
                 reader = FlvReader(down_data)
                 while True:
                     try:
@@ -411,12 +405,8 @@ class F4mFD(FragmentFD):
                             break
                         raise
                     if box_type == b'mdat':
-                        dest_stream.write(box_data)
+                        self._append_fragment(ctx, box_data)
                         break
-                if live:
-                    os.remove(encodeFilename(frag_sanitized))
-                else:
-                    frags_filenames.append(frag_sanitized)
             except (compat_urllib_error.HTTPError, ) as err:
                 if live and (err.code == 404 or err.code == 410):
                     # We didn't keep up with the live window. Continue
@@ -436,7 +426,4 @@ class F4mFD(FragmentFD):
 
         self._finish_frag_download(ctx)
 
-        for frag_file in frags_filenames:
-            os.remove(encodeFilename(frag_file))
-
         return True

youtube_dl/downloader/fragment.py

@@ -2,6 +2,7 @@ from __future__ import division, unicode_literals
 
 import os
 import time
+import json
 
 from .common import FileDownloader
 from .http import HttpFD
@@ -28,15 +29,37 @@ class FragmentFD(FileDownloader):
                         and hlsnative only)
     skip_unavailable_fragments:
                         Skip unavailable fragments (DASH and hlsnative only)
+    keep_fragments:     Keep downloaded fragments on disk after downloading is
+                        finished
+
+    For each incomplete fragment download youtube-dl keeps on disk a special
+    bookkeeping file with download state and metadata (in future such files will
+    be used for any incomplete download handled by youtube-dl). This file is
+    used to properly handle resuming, check download file consistency and detect
+    potential errors. The file has a .ytdl extension and represents a standard
+    JSON file of the following format:
+
+    extractor:
+        Dictionary of extractor related data. TBD.
+
+    downloader:
+        Dictionary of downloader related data. May contain following data:
+            current_fragment:
+                Dictionary with current (being downloaded) fragment data:
+                index:  0-based index of current fragment among all fragments
+            fragment_count:
+                Total count of fragments
+                
+    This feature is experimental and file format may change in future.
     """
 
-    def report_retry_fragment(self, err, fragment_name, count, retries):
+    def report_retry_fragment(self, err, frag_index, count, retries):
         self.to_screen(
-            '[download] Got server HTTP error: %s. Retrying fragment %s (attempt %d of %s)...'
-            % (error_to_compat_str(err), fragment_name, count, self.format_retries(retries)))
+            '[download] Got server HTTP error: %s. Retrying fragment %d (attempt %d of %s)...'
+            % (error_to_compat_str(err), frag_index, count, self.format_retries(retries)))
 
-    def report_skip_fragment(self, fragment_name):
-        self.to_screen('[download] Skipping fragment %s...' % fragment_name)
+    def report_skip_fragment(self, frag_index):
+        self.to_screen('[download] Skipping fragment %d...' % frag_index)
 
     def _prepare_url(self, info_dict, url):
         headers = info_dict.get('http_headers')
@@ -46,6 +69,51 @@ class FragmentFD(FileDownloader):
         self._prepare_frag_download(ctx)
         self._start_frag_download(ctx)
 
+    @staticmethod
+    def __do_ytdl_file(ctx):
+        return not ctx['live'] and not ctx['tmpfilename'] == '-'
+
+    def _read_ytdl_file(self, ctx):
+        stream, _ = sanitize_open(self.ytdl_filename(ctx['filename']), 'r')
+        ctx['fragment_index'] = json.loads(stream.read())['downloader']['current_fragment']['index']
+        stream.close()
+
+    def _write_ytdl_file(self, ctx):
+        frag_index_stream, _ = sanitize_open(self.ytdl_filename(ctx['filename']), 'w')
+        downloader = {
+            'current_fragment': {
+                'index': ctx['fragment_index'],
+            },
+        }
+        if ctx.get('fragment_count') is not None:
+            downloader['fragment_count'] = ctx['fragment_count']
+        frag_index_stream.write(json.dumps({'downloader': downloader}))
+        frag_index_stream.close()
+
+    def _download_fragment(self, ctx, frag_url, info_dict, headers=None):
+        fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], ctx['fragment_index'])
+        success = ctx['dl'].download(fragment_filename, {
+            'url': frag_url,
+            'http_headers': headers or info_dict.get('http_headers'),
+        })
+        if not success:
+            return False, None
+        down, frag_sanitized = sanitize_open(fragment_filename, 'rb')
+        ctx['fragment_filename_sanitized'] = frag_sanitized
+        frag_content = down.read()
+        down.close()
+        return True, frag_content
+
+    def _append_fragment(self, ctx, frag_content):
+        try:
+            ctx['dest_stream'].write(frag_content)
+        finally:
+            if self.__do_ytdl_file(ctx):
+                self._write_ytdl_file(ctx)
+            if not self.params.get('keep_fragments', False):
+                os.remove(ctx['fragment_filename_sanitized'])
+            del ctx['fragment_filename_sanitized']
+
     def _prepare_frag_download(self, ctx):
         if 'live' not in ctx:
             ctx['live'] = False
@@ -61,15 +129,43 @@ class FragmentFD(FileDownloader):
                 'noprogress': True,
                 'ratelimit': self.params.get('ratelimit'),
                 'retries': self.params.get('retries', 0),
+                'nopart': self.params.get('nopart', False),
                 'test': self.params.get('test', False),
             }
         )
         tmpfilename = self.temp_name(ctx['filename'])
-        dest_stream, tmpfilename = sanitize_open(tmpfilename, 'wb')
+        open_mode = 'wb'
+        resume_len = 0
+
+        # Establish possible resume length
+        if os.path.isfile(encodeFilename(tmpfilename)):
+            open_mode = 'ab'
+            resume_len = os.path.getsize(encodeFilename(tmpfilename))
+
+        # Should be initialized before ytdl file check
+        ctx.update({
+            'tmpfilename': tmpfilename,
+            'fragment_index': 0,
+        })
+
+        if self.__do_ytdl_file(ctx):
+            if os.path.isfile(encodeFilename(self.ytdl_filename(ctx['filename']))):
+                self._read_ytdl_file(ctx)
+            else:
+                self._write_ytdl_file(ctx)
+            if ctx['fragment_index'] > 0:
+                assert resume_len > 0
+            else:
+                assert resume_len == 0
+
+        dest_stream, tmpfilename = sanitize_open(tmpfilename, open_mode)
+
         ctx.update({
             'dl': dl,
             'dest_stream': dest_stream,
             'tmpfilename': tmpfilename,
+            # Total complete fragments downloaded so far in bytes
+            'complete_frags_downloaded_bytes': resume_len,
         })
 
     def _start_frag_download(self, ctx):
@@ -78,9 +174,9 @@ class FragmentFD(FileDownloader):
         # hook
         state = {
             'status': 'downloading',
-            'downloaded_bytes': 0,
-            'frag_index': 0,
-            'frag_count': total_frags,
+            'downloaded_bytes': ctx['complete_frags_downloaded_bytes'],
+            'fragment_index': ctx['fragment_index'],
+            'fragment_count': total_frags,
             'filename': ctx['filename'],
             'tmpfilename': ctx['tmpfilename'],
         }
@@ -88,8 +184,6 @@ class FragmentFD(FileDownloader):
         start = time.time()
         ctx.update({
             'started': start,
-            # Total complete fragments downloaded so far in bytes
-            'complete_frags_downloaded_bytes': 0,
             # Amount of fragment's bytes downloaded by the time of the previous
             # frag progress hook invocation
             'prev_frag_downloaded_bytes': 0,
@@ -105,11 +199,12 @@ class FragmentFD(FileDownloader):
             if not ctx['live']:
                 estimated_size = (
                     (ctx['complete_frags_downloaded_bytes'] + frag_total_bytes) /
-                    (state['frag_index'] + 1) * total_frags)
+                    (state['fragment_index'] + 1) * total_frags)
                 state['total_bytes_estimate'] = estimated_size
 
             if s['status'] == 'finished':
-                state['frag_index'] += 1
+                state['fragment_index'] += 1
+                ctx['fragment_index'] = state['fragment_index']
                 state['downloaded_bytes'] += frag_total_bytes - ctx['prev_frag_downloaded_bytes']
                 ctx['complete_frags_downloaded_bytes'] = state['downloaded_bytes']
                 ctx['prev_frag_downloaded_bytes'] = 0
@@ -131,6 +226,10 @@ class FragmentFD(FileDownloader):
 
     def _finish_frag_download(self, ctx):
         ctx['dest_stream'].close()
+        if self.__do_ytdl_file(ctx):
+            ytdl_filename = encodeFilename(self.ytdl_filename(ctx['filename']))
+            if os.path.isfile(ytdl_filename):
+                os.remove(ytdl_filename)
         elapsed = time.time() - ctx['started']
         self.try_rename(ctx['tmpfilename'], ctx['filename'])
         fsize = os.path.getsize(encodeFilename(ctx['filename']))

youtube_dl/downloader/hls.py

@@ -1,6 +1,5 @@
 from __future__ import unicode_literals
 
-import os.path
 import re
 import binascii
 try:
@@ -18,8 +17,6 @@ from ..compat import (
     compat_struct_pack,
 )
 from ..utils import (
-    encodeFilename,
-    sanitize_open,
     parse_m3u8_attributes,
     update_url_query,
 )
@@ -34,7 +31,7 @@ class HlsFD(FragmentFD):
     def can_download(manifest, info_dict):
         UNSUPPORTED_FEATURES = (
             r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)',  # encrypted streams [1]
-            r'#EXT-X-BYTERANGE',  # playlists composed of byte ranges of media files [2]
+            # 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)
@@ -52,7 +49,9 @@ class HlsFD(FragmentFD):
             # 4. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.5
         )
         check_results = [not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES]
-        check_results.append(can_decrypt_frag or '#EXT-X-KEY:METHOD=AES-128' not in manifest)
+        is_aes128_enc = '#EXT-X-KEY:METHOD=AES-128' in manifest
+        check_results.append(can_decrypt_frag or not is_aes128_enc)
+        check_results.append(not (is_aes128_enc and r'#EXT-X-BYTERANGE' in manifest))
         check_results.append(not info_dict.get('is_live'))
         return all(check_results)
 
@@ -100,31 +99,31 @@ class HlsFD(FragmentFD):
         i = 0
         media_sequence = 0
         decrypt_info = {'METHOD': 'NONE'}
-        frags_filenames = []
+        byte_range = {}
+        frag_index = 0
         for line in s.splitlines():
             line = line.strip()
             if line:
                 if not line.startswith('#'):
+                    frag_index += 1
+                    if frag_index <= ctx['fragment_index']:
+                        continue
                     frag_url = (
                         line
                         if re.match(r'^https?://', line)
                         else compat_urlparse.urljoin(man_url, line))
-                    frag_name = 'Frag%d' % i
-                    frag_filename = '%s-%s' % (ctx['tmpfilename'], frag_name)
                     if extra_query:
                         frag_url = update_url_query(frag_url, extra_query)
                     count = 0
+                    headers = info_dict.get('http_headers', {})
+                    if byte_range:
+                        headers['Range'] = 'bytes=%d-%d' % (byte_range['start'], byte_range['end'])
                     while count <= fragment_retries:
                         try:
-                            success = ctx['dl'].download(frag_filename, {
-                                'url': frag_url,
-                                'http_headers': info_dict.get('http_headers'),
-                            })
+                            success, frag_content = self._download_fragment(
+                                ctx, frag_url, info_dict, headers)
                             if not success:
                                 return False
-                            down, frag_sanitized = sanitize_open(frag_filename, 'rb')
-                            frag_content = down.read()
-                            down.close()
                             break
                         except compat_urllib_error.HTTPError as err:
                             # Unavailable (possibly temporary) fragments may be served.
@@ -133,28 +132,29 @@ class HlsFD(FragmentFD):
                             # https://github.com/rg3/youtube-dl/issues/10448).
                             count += 1
                             if count <= fragment_retries:
-                                self.report_retry_fragment(err, frag_name, count, fragment_retries)
+                                self.report_retry_fragment(err, frag_index, count, fragment_retries)
                     if count > fragment_retries:
                         if skip_unavailable_fragments:
                             i += 1
                             media_sequence += 1
-                            self.report_skip_fragment(frag_name)
+                            self.report_skip_fragment(frag_index)
                             continue
                         self.report_error(
                             'giving up after %s fragment retries' % fragment_retries)
                         return False
                     if decrypt_info['METHOD'] == 'AES-128':
                         iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence)
+                        decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen(decrypt_info['URI']).read()
                         frag_content = AES.new(
                             decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
-                    ctx['dest_stream'].write(frag_content)
-                    frags_filenames.append(frag_sanitized)
+                    self._append_fragment(ctx, frag_content)
                     # We only download the first fragment during the test
                     if test:
                         break
                     i += 1
                     media_sequence += 1
                 elif line.startswith('#EXT-X-KEY'):
+                    decrypt_url = decrypt_info.get('URI')
                     decrypt_info = parse_m3u8_attributes(line[11:])
                     if decrypt_info['METHOD'] == 'AES-128':
                         if 'IV' in decrypt_info:
@@ -164,13 +164,18 @@ class HlsFD(FragmentFD):
                                 man_url, decrypt_info['URI'])
                         if extra_query:
                             decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_query)
-                        decrypt_info['KEY'] = self.ydl.urlopen(decrypt_info['URI']).read()
+                        if decrypt_url != decrypt_info['URI']:
+                            decrypt_info['KEY'] = None
                 elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):
                     media_sequence = int(line[22:])
+                elif line.startswith('#EXT-X-BYTERANGE'):
+                    splitted_byte_range = line[17:].split('@')
+                    sub_range_start = int(splitted_byte_range[1]) if len(splitted_byte_range) == 2 else byte_range['end']
+                    byte_range = {
+                        'start': sub_range_start,
+                        'end': sub_range_start + int(splitted_byte_range[0]),
+                    }
 
         self._finish_frag_download(ctx)
 
-        for frag_file in frags_filenames:
-            os.remove(encodeFilename(frag_file))
-
         return True

youtube_dl/downloader/ism.py

@@ -1,6 +1,5 @@
 from __future__ import unicode_literals
 
-import os
 import time
 import struct
 import binascii
@@ -8,10 +7,6 @@ import io
 
 from .fragment import FragmentFD
 from ..compat import compat_urllib_error
-from ..utils import (
-    sanitize_open,
-    encodeFilename,
-)
 
 
 u8 = struct.Struct(b'>B')
@@ -225,47 +220,39 @@ class IsmFD(FragmentFD):
 
         self._prepare_and_start_frag_download(ctx)
 
-        segments_filenames = []
-
         fragment_retries = self.params.get('fragment_retries', 0)
         skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
 
         track_written = False
+        frag_index = 0
         for i, segment in enumerate(segments):
-            segment_url = segment['url']
-            segment_name = 'Frag%d' % i
-            target_filename = '%s-%s' % (ctx['tmpfilename'], segment_name)
+            frag_index += 1
+            if frag_index <= ctx['fragment_index']:
+                continue
             count = 0
             while count <= fragment_retries:
                 try:
-                    success = ctx['dl'].download(target_filename, {'url': segment_url})
+                    success, frag_content = self._download_fragment(ctx, segment['url'], info_dict)
                     if not success:
                         return False
-                    down, target_sanitized = sanitize_open(target_filename, 'rb')
-                    down_data = down.read()
                     if not track_written:
-                        tfhd_data = extract_box_data(down_data, [b'moof', b'traf', b'tfhd'])
+                        tfhd_data = extract_box_data(frag_content, [b'moof', b'traf', b'tfhd'])
                         info_dict['_download_params']['track_id'] = u32.unpack(tfhd_data[4:8])[0]
                         write_piff_header(ctx['dest_stream'], info_dict['_download_params'])
                         track_written = True
-                    ctx['dest_stream'].write(down_data)
-                    down.close()
-                    segments_filenames.append(target_sanitized)
+                    self._append_fragment(ctx, frag_content)
                     break
                 except compat_urllib_error.HTTPError as err:
                     count += 1
                     if count <= fragment_retries:
-                        self.report_retry_fragment(err, segment_name, count, fragment_retries)
+                        self.report_retry_fragment(err, frag_index, count, fragment_retries)
             if count > fragment_retries:
                 if skip_unavailable_fragments:
-                    self.report_skip_fragment(segment_name)
+                    self.report_skip_fragment(frag_index)
                     continue
                 self.report_error('giving up after %s fragment retries' % fragment_retries)
                 return False
 
         self._finish_frag_download(ctx)
 
-        for segment_file in segments_filenames:
-            os.remove(encodeFilename(segment_file))
-
         return True

youtube_dl/downloader/rtmp.py

@@ -169,7 +169,7 @@ class RtmpFD(FileDownloader):
             self.report_error('[rtmpdump] Could not connect to RTMP server.')
             return False
 
-        while (retval == RD_INCOMPLETE or retval == RD_FAILED) and not test and not live:
+        while retval in (RD_INCOMPLETE, RD_FAILED) and not test and not live:
             prevsize = os.path.getsize(encodeFilename(tmpfilename))
             self.to_screen('[rtmpdump] %s bytes' % prevsize)
             time.sleep(5.0)  # This seems to be needed

youtube_dl/extractor/addanime.py

@@ -25,7 +25,8 @@ class AddAnimeIE(InfoExtractor):
             'ext': 'mp4',
             'description': 'One Piece 606',
             'title': 'One Piece 606',
-        }
+        },
+        'skip': 'Video is gone',
     }, {
         'url': 'http://add-anime.net/video/MDUGWYKNGBD8/One-Piece-687',
         'only_matching': True,

youtube_dl/extractor/adn.py

@@ -0,0 +1,136 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import base64
+import json
+import os
+
+from .common import InfoExtractor
+from ..aes import aes_cbc_decrypt
+from ..compat import compat_ord
+from ..utils import (
+    bytes_to_intlist,
+    ExtractorError,
+    float_or_none,
+    intlist_to_bytes,
+    srt_subtitles_timecode,
+    strip_or_none,
+)
+
+
+class ADNIE(InfoExtractor):
+    IE_DESC = 'Anime Digital Network'
+    _VALID_URL = r'https?://(?:www\.)?animedigitalnetwork\.fr/video/[^/]+/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites',
+        'md5': 'e497370d847fd79d9d4c74be55575c7a',
+        'info_dict': {
+            'id': '7778',
+            'ext': 'mp4',
+            'title': 'Blue Exorcist - Kyôto Saga - Épisode 1',
+            'description': 'md5:2f7b5aa76edbc1a7a92cedcda8a528d5',
+        }
+    }
+
+    def _get_subtitles(self, sub_path, video_id):
+        if not sub_path:
+            return None
+
+        enc_subtitles = self._download_webpage(
+            'http://animedigitalnetwork.fr/' + sub_path,
+            video_id, fatal=False)
+        if not enc_subtitles:
+            return None
+
+        # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
+        dec_subtitles = intlist_to_bytes(aes_cbc_decrypt(
+            bytes_to_intlist(base64.b64decode(enc_subtitles[24:])),
+            bytes_to_intlist(b'\nd\xaf\xd2J\xd0\xfc\xe1\xfc\xdf\xb61\xe8\xe1\xf0\xcc'),
+            bytes_to_intlist(base64.b64decode(enc_subtitles[:24]))
+        ))
+        subtitles_json = self._parse_json(
+            dec_subtitles[:-compat_ord(dec_subtitles[-1])],
+            None, fatal=False)
+        if not subtitles_json:
+            return None
+
+        subtitles = {}
+        for sub_lang, sub in subtitles_json.items():
+            srt = ''
+            for num, current in enumerate(sub):
+                start, end, text = (
+                    float_or_none(current.get('startTime')),
+                    float_or_none(current.get('endTime')),
+                    current.get('text'))
+                if start is None or end is None or text is None:
+                    continue
+                srt += os.linesep.join(
+                    (
+                        '%d' % num,
+                        '%s --> %s' % (
+                            srt_subtitles_timecode(start),
+                            srt_subtitles_timecode(end)),
+                        text,
+                        os.linesep,
+                    ))
+
+            if sub_lang == 'vostf':
+                sub_lang = 'fr'
+            subtitles.setdefault(sub_lang, []).extend([{
+                'ext': 'json',
+                'data': json.dumps(sub),
+            }, {
+                'ext': 'srt',
+                'data': srt,
+            }])
+        return subtitles
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        player_config = self._parse_json(self._search_regex(
+            r'playerConfig\s*=\s*({.+});', webpage, 'player config'), video_id)
+
+        video_info = {}
+        video_info_str = self._search_regex(
+            r'videoInfo\s*=\s*({.+});', webpage,
+            'video info', fatal=False)
+        if video_info_str:
+            video_info = self._parse_json(
+                video_info_str, video_id, fatal=False) or {}
+
+        options = player_config.get('options') or {}
+        metas = options.get('metas') or {}
+        title = metas.get('title') or video_info['title']
+        links = player_config.get('links') or {}
+
+        formats = []
+        for format_id, qualities in links.items():
+            for load_balancer_url in qualities.values():
+                load_balancer_data = self._download_json(
+                    load_balancer_url, video_id, fatal=False) or {}
+                m3u8_url = load_balancer_data.get('location')
+                if not m3u8_url:
+                    continue
+                m3u8_formats = self._extract_m3u8_formats(
+                    m3u8_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id=format_id, fatal=False)
+                if format_id == 'vf':
+                    for f in m3u8_formats:
+                        f['language'] = 'fr'
+                formats.extend(m3u8_formats)
+        error = options.get('error')
+        if not formats and error:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': strip_or_none(metas.get('summary') or video_info.get('resume')),
+            'thumbnail': video_info.get('image'),
+            'formats': formats,
+            'subtitles': self.extract_subtitles(player_config.get('subtitles'), video_id),
+            'episode': metas.get('subtitle') or video_info.get('videoTitle'),
+            'series': video_info.get('playlistTitle'),
+        }

youtube_dl/extractor/adobepass.py

@@ -31,6 +31,21 @@ MSO_INFO = {
         'username_field': 'user',
         'password_field': 'passwd',
     },
+    'TWC': {
+        'name': 'Time Warner Cable | Spectrum',
+        'username_field': 'Ecom_User_ID',
+        'password_field': 'Ecom_Password',
+    },
+    'Charter_Direct': {
+        'name': 'Charter Spectrum',
+        'username_field': 'IDToken1',
+        'password_field': 'IDToken2',
+    },
+    'Verizon': {
+        'name': 'Verizon FiOS',
+        'username_field': 'IDToken1',
+        'password_field': 'IDToken2',
+    },
     'thr030': {
         'name': '3 Rivers Communications'
     },
@@ -1374,40 +1389,72 @@ class AdobePassIE(InfoExtractor):
                     # Comcast page flow varies by video site and whether you
                     # are on Comcast's network.
                     provider_redirect_page, urlh = provider_redirect_page_res
-                    # Check for Comcast auto login
                     if 'automatically signing you in' in provider_redirect_page:
                         oauth_redirect_url = self._html_search_regex(
                             r'window\.location\s*=\s*[\'"]([^\'"]+)',
                             provider_redirect_page, 'oauth redirect')
-                        # Just need to process the request. No useful data comes back
                         self._download_webpage(
                             oauth_redirect_url, video_id, 'Confirming auto login')
                     else:
                         if '<form name="signin"' in provider_redirect_page:
-                            # already have the form, just fill it
                             provider_login_page_res = provider_redirect_page_res
                         elif 'http-equiv="refresh"' in provider_redirect_page:
-                            # redirects to the login page
                             oauth_redirect_url = self._html_search_regex(
                                 r'content="0;\s*url=([^\'"]+)',
                                 provider_redirect_page, 'meta refresh redirect')
                             provider_login_page_res = self._download_webpage_handle(
-                                oauth_redirect_url,
-                                video_id, 'Downloading Provider Login Page')
+                                oauth_redirect_url, video_id,
+                                'Downloading Provider Login Page')
                         else:
                             provider_login_page_res = post_form(
-                                provider_redirect_page_res, 'Downloading Provider Login Page')
+                                provider_redirect_page_res,
+                                'Downloading Provider Login Page')
 
-                        mvpd_confirm_page_res = post_form(provider_login_page_res, 'Logging in', {
-                            mso_info.get('username_field', 'username'): username,
-                            mso_info.get('password_field', 'password'): password,
-                        })
+                        mvpd_confirm_page_res = post_form(
+                            provider_login_page_res, 'Logging in', {
+                                mso_info['username_field']: username,
+                                mso_info['password_field']: password,
+                            })
                         mvpd_confirm_page, urlh = mvpd_confirm_page_res
                         if '<button class="submit" value="Resume">Resume</button>' in mvpd_confirm_page:
                             post_form(mvpd_confirm_page_res, 'Confirming Login')
-
+                elif mso_id == 'Verizon':
+                    # In general, if you're connecting from a Verizon-assigned IP,
+                    # you will not actually pass your credentials.
+                    provider_redirect_page, urlh = provider_redirect_page_res
+                    if 'Please wait ...' in provider_redirect_page:
+                        saml_redirect_url = self._html_search_regex(
+                            r'self\.parent\.location=(["\'])(?P<url>.+?)\1',
+                            provider_redirect_page,
+                            'SAML Redirect URL', group='url')
+                        saml_login_page = self._download_webpage(
+                            saml_redirect_url, video_id,
+                            'Downloading SAML Login Page')
+                    else:
+                        saml_login_page_res = post_form(
+                            provider_redirect_page_res, 'Logging in', {
+                                mso_info['username_field']: username,
+                                mso_info['password_field']: password,
+                            })
+                        saml_login_page, urlh = saml_login_page_res
+                        if 'Please try again.' in saml_login_page:
+                            raise ExtractorError(
+                                'We\'re sorry, but either the User ID or Password entered is not correct.')
+                    saml_login_url = self._search_regex(
+                        r'xmlHttp\.open\("POST"\s*,\s*(["\'])(?P<url>.+?)\1',
+                        saml_login_page, 'SAML Login URL', group='url')
+                    saml_response_json = self._download_json(
+                        saml_login_url, video_id, 'Downloading SAML Response',
+                        headers={'Content-Type': 'text/xml'})
+                    self._download_webpage(
+                        saml_response_json['targetValue'], video_id,
+                        'Confirming Login', data=urlencode_postdata({
+                            'SAMLResponse': saml_response_json['SAMLResponse'],
+                            'RelayState': saml_response_json['RelayState']
+                        }), headers={
+                            'Content-Type': 'application/x-www-form-urlencoded'
+                        })
                 else:
-                    # Normal, non-Comcast flow
                     provider_login_page_res = post_form(
                         provider_redirect_page_res, 'Downloading Provider Login Page')
                     mvpd_confirm_page_res = post_form(provider_login_page_res, 'Logging in', {
@@ -1448,6 +1495,8 @@ class AdobePassIE(InfoExtractor):
                     self._downloader.cache.store(self._MVPD_CACHE, requestor_id, {})
                     count += 1
                     continue
+                if '<error' in authorize:
+                    raise ExtractorError(xml_text(authorize, 'details'), expected=True)
                 authz_token = unescapeHTML(xml_text(authorize, 'authzToken'))
                 requestor_info[guid] = authz_token
                 self._downloader.cache.store(self._MVPD_CACHE, requestor_id, requestor_info)

youtube_dl/extractor/aenetworks.py

@@ -23,7 +23,19 @@ class AENetworksBaseIE(ThePlatformIE):
 class AENetworksIE(AENetworksBaseIE):
     IE_NAME = 'aenetworks'
     IE_DESC = 'A+E Networks: A&E, Lifetime, History.com, FYI Network'
-    _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:history|aetv|mylifetime)\.com|fyi\.tv)/(?:shows/(?P<show_path>[^/]+(?:/[^/]+){0,2})|movies/(?P<movie_display_id>[^/]+)/full-movie)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?
+                        (?P<domain>
+                            (?:history|aetv|mylifetime|lifetimemovieclub)\.com|
+                            fyi\.tv
+                        )/
+                        (?:
+                            shows/(?P<show_path>[^/]+(?:/[^/]+){0,2})|
+                            movies/(?P<movie_display_id>[^/]+)(?:/full-movie)?|
+                            specials/(?P<special_display_id>[^/]+)/full-special
+                        )
+                    '''
     _TESTS = [{
         'url': 'http://www.history.com/shows/mountain-men/season-1/episode-1',
         'md5': 'a97a65f7e823ae10e9244bc5433d5fe6',
@@ -62,17 +74,24 @@ class AENetworksIE(AENetworksBaseIE):
     }, {
         'url': 'http://www.mylifetime.com/movies/center-stage-on-pointe/full-movie',
         'only_matching': True
+    }, {
+        'url': 'https://www.lifetimemovieclub.com/movies/a-killer-among-us',
+        'only_matching': True
+    }, {
+        'url': 'http://www.history.com/specials/sniper-into-the-kill-zone/full-special',
+        'only_matching': True
     }]
     _DOMAIN_TO_REQUESTOR_ID = {
         'history.com': 'HISTORY',
         'aetv.com': 'AETV',
         'mylifetime.com': 'LIFETIME',
+        'lifetimemovieclub.com': 'LIFETIMEMOVIECLUB',
         'fyi.tv': 'FYI',
     }
 
     def _real_extract(self, url):
-        domain, show_path, movie_display_id = re.match(self._VALID_URL, url).groups()
-        display_id = show_path or movie_display_id
+        domain, show_path, movie_display_id, special_display_id = re.match(self._VALID_URL, url).groups()
+        display_id = show_path or movie_display_id or special_display_id
         webpage = self._download_webpage(url, display_id)
         if show_path:
             url_parts = show_path.split('/')
@@ -103,7 +122,10 @@ class AENetworksIE(AENetworksBaseIE):
         }
         video_id = self._html_search_meta('aetn:VideoID', webpage)
         media_url = self._search_regex(
-            r"media_url\s*=\s*'([^']+)'", webpage, 'video url')
+            [r"media_url\s*=\s*'(?P<url>[^']+)'",
+             r'data-media-url=(?P<url>(?:https?:)?//[^\s>]+)',
+             r'data-media-url=(["\'])(?P<url>(?:(?!\1).)+?)\1'],
+            webpage, 'video url', group='url')
         theplatform_metadata = self._download_theplatform_metadata(self._search_regex(
             r'https?://link.theplatform.com/s/([^?]+)', media_url, 'theplatform_path'), video_id)
         info = self._parse_theplatform_metadata(theplatform_metadata)

youtube_dl/extractor/afreecatv.py

@@ -4,15 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse_urlparse,
-    compat_urlparse,
-)
+from ..compat import compat_xpath
 from ..utils import (
+    determine_ext,
     ExtractorError,
     int_or_none,
-    update_url_query,
-    xpath_element,
     xpath_text,
 )
 
@@ -43,7 +39,8 @@ class AfreecaTVIE(InfoExtractor):
             'uploader': 'dailyapril',
             'uploader_id': 'dailyapril',
             'upload_date': '20160503',
-        }
+        },
+        'skip': 'Video is gone',
     }, {
         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
         'info_dict': {
@@ -71,6 +68,76 @@ class AfreecaTVIE(InfoExtractor):
                 'upload_date': '20160502',
             },
         }],
+        'skip': 'Video is gone',
+    }, {
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/18650793',
+        'info_dict': {
+            'id': '18650793',
+            'ext': 'mp4',
+            'title': '오늘은 다르다! 쏘님의 우월한 위아래~ 댄스리액션!',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': '윈아디',
+            'uploader_id': 'badkids',
+            'duration': 107,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/10481652',
+        'info_dict': {
+            'id': '10481652',
+            'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': 'dailyapril',
+            'uploader_id': 'dailyapril',
+            'duration': 6492,
+        },
+        'playlist_count': 2,
+        'playlist': [{
+            'md5': 'd8b7c174568da61d774ef0203159bf97',
+            'info_dict': {
+                'id': '20160502_c4c62b9d_174361386_1',
+                'ext': 'mp4',
+                'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 1)",
+                'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+                'uploader': 'dailyapril',
+                'uploader_id': 'dailyapril',
+                'upload_date': '20160502',
+                'duration': 3601,
+            },
+        }, {
+            'md5': '58f2ce7f6044e34439ab2d50612ab02b',
+            'info_dict': {
+                'id': '20160502_39e739bb_174361386_2',
+                'ext': 'mp4',
+                'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 2)",
+                'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+                'uploader': 'dailyapril',
+                'uploader_id': 'dailyapril',
+                'upload_date': '20160502',
+                'duration': 2891,
+            },
+        }],
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # non standard key
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/20515605',
+        'info_dict': {
+            'id': '20170411_BE689A0E_190960999_1_2_h',
+            'ext': 'mp4',
+            'title': '혼자사는여자집',
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': '♥이슬이',
+            'uploader_id': 'dasl8121',
+            'upload_date': '20170411',
+            'duration': 213,
+        },
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
         'only_matching': True,
@@ -85,42 +152,77 @@ class AfreecaTVIE(InfoExtractor):
         m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
         if m:
             video_key['upload_date'] = m.group('upload_date')
-            video_key['part'] = m.group('part')
+            video_key['part'] = int(m.group('part'))
         return video_key
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        parsed_url = compat_urllib_parse_urlparse(url)
-        info_url = compat_urlparse.urlunparse(parsed_url._replace(
-            netloc='afbbs.afreecatv.com:8080',
-            path='/api/video/get_video_info.php'))
 
         video_xml = self._download_xml(
-            update_url_query(info_url, {'nTitleNo': video_id}), video_id)
+            'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
+            video_id, query={'nTitleNo': video_id})
 
-        if xpath_element(video_xml, './track/video/file') is None:
+        video_element = video_xml.findall(compat_xpath('./track/video'))[1]
+        if video_element is None or video_element.text is None:
             raise ExtractorError('Specified AfreecaTV video does not exist',
                                  expected=True)
 
-        title = xpath_text(video_xml, './track/title', 'title')
+        video_url = video_element.text.strip()
+
+        title = xpath_text(video_xml, './track/title', 'title', fatal=True)
+
         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
-        duration = int_or_none(xpath_text(video_xml, './track/duration',
-                                          'duration'))
+        duration = int_or_none(xpath_text(
+            video_xml, './track/duration', 'duration'))
         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 
-        entries = []
-        for i, video_file in enumerate(video_xml.findall('./track/video/file')):
-            video_key = self.parse_video_key(video_file.get('key', ''))
-            if not video_key:
-                continue
-            entries.append({
-                'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
-                'title': title,
-                'upload_date': video_key.get('upload_date'),
-                'duration': int_or_none(video_file.get('duration')),
-                'url': video_file.text,
+        common_entry = {
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'thumbnail': thumbnail,
+        }
+
+        info = common_entry.copy()
+        info.update({
+            'id': video_id,
+            'title': title,
+            'duration': duration,
+        })
+
+        if not video_url:
+            entries = []
+            file_elements = video_element.findall(compat_xpath('./file'))
+            one = len(file_elements) == 1
+            for file_num, file_element in enumerate(file_elements, start=1):
+                file_url = file_element.text
+                if not file_url:
+                    continue
+                key = file_element.get('key', '')
+                upload_date = self._search_regex(
+                    r'^(\d{8})_', key, 'upload date', default=None)
+                file_duration = int_or_none(file_element.get('duration'))
+                format_id = key if key else '%s_%s' % (video_id, file_num)
+                formats = self._extract_m3u8_formats(
+                    file_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls',
+                    note='Downloading part %d m3u8 information' % file_num)
+                title = title if one else '%s (part %d)' % (title, file_num)
+                file_info = common_entry.copy()
+                file_info.update({
+                    'id': format_id,
+                    'title': title,
+                    'upload_date': upload_date,
+                    'duration': file_duration,
+                    'formats': formats,
+                })
+                entries.append(file_info)
+            entries_info = info.copy()
+            entries_info.update({
+                '_type': 'multi_video',
+                'entries': entries,
             })
+            return entries_info
 
         info = {
             'id': video_id,
@@ -131,17 +233,18 @@ class AfreecaTVIE(InfoExtractor):
             'thumbnail': thumbnail,
         }
 
-        if len(entries) > 1:
-            info['_type'] = 'multi_video'
-            info['entries'] = entries
-        elif len(entries) == 1:
-            info['url'] = entries[0]['url']
-            info['upload_date'] = entries[0].get('upload_date')
+        if determine_ext(video_url) == 'm3u8':
+            info['formats'] = self._extract_m3u8_formats(
+                video_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                m3u8_id='hls')
         else:
-            raise ExtractorError(
-                'No files found for the specified AfreecaTV video, either'
-                ' the URL is incorrect or the video has been made private.',
-                expected=True)
+            app, playpath = video_url.split('mp4:')
+            info.update({
+                'url': app,
+                'ext': 'flv',
+                'play_path': 'mp4:' + playpath,
+                'rtmp_live': True,  # downloading won't end without this
+            })
 
         return info
 
@@ -221,10 +324,23 @@ class AfreecaTVGlobalIE(AfreecaTVIE):
                 s_url = s.get('purl')
                 if not s_url:
                     continue
-                # TODO: extract rtmp formats
-                if s.get('stype') == 'HLS':
+                stype = s.get('stype')
+                if stype == 'HLS':
                     formats.extend(self._extract_m3u8_formats(
-                        s_url, channel_id, 'mp4', fatal=False))
+                        s_url, channel_id, 'mp4', m3u8_id=stype, fatal=False))
+                elif stype == 'RTMP':
+                    format_id = [stype]
+                    label = s.get('label')
+                    if label:
+                        format_id.append(label)
+                    formats.append({
+                        'format_id': '-'.join(format_id),
+                        'url': s_url,
+                        'tbr': int_or_none(s.get('bps')),
+                        'height': int_or_none(s.get('brt')),
+                        'ext': 'flv',
+                        'rtmp_live': True,
+                    })
             self._sort_formats(formats)
 
             info.update({

youtube_dl/extractor/airmozilla.py

@@ -15,12 +15,12 @@ class AirMozillaIE(InfoExtractor):
     _VALID_URL = r'https?://air\.mozilla\.org/(?P<id>[0-9a-z-]+)/?'
     _TEST = {
         'url': 'https://air.mozilla.org/privacy-lab-a-meetup-for-privacy-minded-people-in-san-francisco/',
-        'md5': '2e3e7486ba5d180e829d453875b9b8bf',
+        'md5': '8d02f53ee39cf006009180e21df1f3ba',
         'info_dict': {
             'id': '6x4q2w',
             'ext': 'mp4',
             'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco',
-            'thumbnail': r're:https?://vid\.ly/(?P<id>[0-9a-z-]+)/poster',
+            'thumbnail': r're:https?://.*/poster\.jpg',
             'description': 'Brings together privacy professionals and others interested in privacy at for-profits, non-profits, and NGOs in an effort to contribute to the state of the ecosystem...',
             'timestamp': 1422487800,
             'upload_date': '20150128',
@@ -34,21 +34,13 @@ class AirMozillaIE(InfoExtractor):
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
-        video_id = self._html_search_regex(r'//vid.ly/(.*?)/embed', webpage, 'id')
+        video_id = self._html_search_regex(r'//vid\.ly/(.*?)/embed', webpage, 'id')
 
         embed_script = self._download_webpage('https://vid.ly/{0}/embed'.format(video_id), video_id)
-        jwconfig = self._search_regex(r'\svar jwconfig = (\{.*?\});\s', embed_script, 'metadata')
-        metadata = self._parse_json(jwconfig, video_id)
-
-        formats = [{
-            'url': source['file'],
-            'ext': source['type'],
-            'format_id': self._search_regex(r'&format=(.*)$', source['file'], 'video format'),
-            'format': source['label'],
-            'height': int(source['label'].rstrip('p')),
-        } for source in metadata['playlist'][0]['sources']]
-        self._sort_formats(formats)
+        jwconfig = self._parse_json(self._search_regex(
+            r'initCallback\((.*)\);', embed_script, 'metadata'), video_id)['config']
 
+        info_dict = self._parse_jwplayer_data(jwconfig, video_id)
         view_count = int_or_none(self._html_search_regex(
             r'Views since archived: ([0-9]+)',
             webpage, 'view count', fatal=False))
@@ -58,17 +50,17 @@ class AirMozillaIE(InfoExtractor):
             r'Duration:\s*(\d+\s*hours?\s*\d+\s*minutes?)',
             webpage, 'duration', fatal=False))
 
-        return {
+        info_dict.update({
             'id': video_id,
             'title': self._og_search_title(webpage),
-            'formats': formats,
             'url': self._og_search_url(webpage),
             'display_id': display_id,
-            'thumbnail': metadata['playlist'][0].get('image'),
             'description': self._og_search_description(webpage),
             'timestamp': timestamp,
             'location': self._html_search_regex(r'Location: (.*)', webpage, 'location', default=None),
             'duration': duration,
             'view_count': view_count,
             'categories': re.findall(r'<a href=".*?" class="channel">(.*?)</a>', webpage),
-        }
+        })
+
+        return info_dict

youtube_dl/extractor/allocine.py

@@ -2,9 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    remove_end,
+    int_or_none,
     qualities,
+    remove_end,
+    try_get,
+    unified_timestamp,
     url_basename,
 )
 
@@ -22,6 +26,10 @@ class AllocineIE(InfoExtractor):
             'title': 'Astérix - Le Domaine des Dieux Teaser VF',
             'description': 'md5:4a754271d9c6f16c72629a8a993ee884',
             'thumbnail': r're:http://.*\.jpg',
+            'duration': 39,
+            'timestamp': 1404273600,
+            'upload_date': '20140702',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/player_gen_cmedia=19540403&cfilm=222257.html',
@@ -33,6 +41,10 @@ class AllocineIE(InfoExtractor):
             'title': 'Planes 2 Bande-annonce VF',
             'description': 'Regardez la bande annonce du film Planes 2 (Planes 2 Bande-annonce VF). Planes 2, un film de Roberts Gannaway',
             'thumbnail': r're:http://.*\.jpg',
+            'duration': 69,
+            'timestamp': 1385659800,
+            'upload_date': '20131128',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/player_gen_cmedia=19544709&cfilm=181290.html',
@@ -44,6 +56,10 @@ class AllocineIE(InfoExtractor):
             'title': 'Dragons 2 - Bande annonce finale VF',
             'description': 'md5:6cdd2d7c2687d4c6aafe80a35e17267a',
             'thumbnail': r're:http://.*\.jpg',
+            'duration': 144,
+            'timestamp': 1397589900,
+            'upload_date': '20140415',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/video-19550147/',
@@ -69,34 +85,37 @@ class AllocineIE(InfoExtractor):
             r'data-model="([^"]+)"', webpage, 'data model', default=None)
         if model:
             model_data = self._parse_json(model, display_id)
-
-            for video_url in model_data['sources'].values():
+            video = model_data['videos'][0]
+            title = video['title']
+            for video_url in video['sources'].values():
                 video_id, format_id = url_basename(video_url).split('_')[:2]
                 formats.append({
                     'format_id': format_id,
                     'quality': quality(format_id),
                     'url': video_url,
                 })
-
-            title = model_data['title']
+            duration = int_or_none(video.get('duration'))
+            view_count = int_or_none(video.get('view_count'))
+            timestamp = unified_timestamp(try_get(
+                video, lambda x: x['added_at']['date'], compat_str))
         else:
             video_id = display_id
             media_data = self._download_json(
                 'http://www.allocine.fr/ws/AcVisiondataV5.ashx?media=%s' % video_id, display_id)
+            title = remove_end(
+                self._html_search_regex(
+                    r'(?s)<title>(.+?)</title>', webpage, 'title').strip(),
+                ' - AlloCiné')
             for key, value in media_data['video'].items():
                 if not key.endswith('Path'):
                     continue
-
                 format_id = key[:-len('Path')]
                 formats.append({
                     'format_id': format_id,
                     'quality': quality(format_id),
                     'url': value,
                 })
-
-            title = remove_end(self._html_search_regex(
-                r'(?s)<title>(.+?)</title>', webpage, 'title'
-            ).strip(), ' - AlloCiné')
+            duration, view_count, timestamp = [None] * 3
 
         self._sort_formats(formats)
 
@@ -104,7 +123,10 @@ class AllocineIE(InfoExtractor):
             'id': video_id,
             'display_id': display_id,
             'title': title,
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'formats': formats,
             'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'duration': duration,
+            'timestamp': timestamp,
+            'view_count': view_count,
+            'formats': formats,
         }

youtube_dl/extractor/amcnetworks.py

@@ -10,7 +10,7 @@ from ..utils import (
 
 
 class AMCNetworksIE(ThePlatformIE):
-    _VALID_URL = r'https?://(?:www\.)?(?:amc|bbcamerica|ifc|wetv)\.com/(?:movies/|shows/[^/]+/(?:full-episodes/)?[^/]+/episode-\d+(?:-(?:[^/]+/)?|/))(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:amc|bbcamerica|ifc|wetv)\.com/(?:movies|shows(?:/[^/]+)+)/(?P<id>[^/?#]+)'
     _TESTS = [{
         'url': 'http://www.ifc.com/shows/maron/season-04/episode-01/step-1',
         'md5': '',
@@ -44,6 +44,12 @@ class AMCNetworksIE(ThePlatformIE):
     }, {
         'url': 'http://www.bbcamerica.com/shows/doctor-who/full-episodes/the-power-of-the-daleks/episode-01-episode-1-color-version',
         'only_matching': True,
+    }, {
+        'url': 'http://www.wetv.com/shows/mama-june-from-not-to-hot/full-episode/season-01/thin-tervention',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.wetv.com/shows/la-hair/videos/season-05/episode-09-episode-9-2/episode-9-sneak-peek-3',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -53,20 +59,30 @@ class AMCNetworksIE(ThePlatformIE):
             'mbr': 'true',
             'manifest': 'm3u',
         }
-        media_url = self._search_regex(r'window\.platformLinkURL\s*=\s*[\'"]([^\'"]+)', webpage, 'media url')
+        media_url = self._search_regex(
+            r'window\.platformLinkURL\s*=\s*[\'"]([^\'"]+)',
+            webpage, 'media url')
         theplatform_metadata = self._download_theplatform_metadata(self._search_regex(
-            r'https?://link.theplatform.com/s/([^?]+)', media_url, 'theplatform_path'), display_id)
+            r'link\.theplatform\.com/s/([^?]+)',
+            media_url, 'theplatform_path'), display_id)
         info = self._parse_theplatform_metadata(theplatform_metadata)
         video_id = theplatform_metadata['pid']
         title = theplatform_metadata['title']
         rating = theplatform_metadata['ratings'][0]['rating']
-        auth_required = self._search_regex(r'window\.authRequired\s*=\s*(true|false);', webpage, 'auth required')
+        auth_required = self._search_regex(
+            r'window\.authRequired\s*=\s*(true|false);',
+            webpage, 'auth required')
         if auth_required == 'true':
-            requestor_id = self._search_regex(r'window\.requestor_id\s*=\s*[\'"]([^\'"]+)', webpage, 'requestor id')
-            resource = self._get_mvpd_resource(requestor_id, title, video_id, rating)
-            query['auth'] = self._extract_mvpd_auth(url, video_id, requestor_id, resource)
+            requestor_id = self._search_regex(
+                r'window\.requestor_id\s*=\s*[\'"]([^\'"]+)',
+                webpage, 'requestor id')
+            resource = self._get_mvpd_resource(
+                requestor_id, title, video_id, rating)
+            query['auth'] = self._extract_mvpd_auth(
+                url, video_id, requestor_id, resource)
         media_url = update_url_query(media_url, query)
-        formats, subtitles = self._extract_theplatform_smil(media_url, video_id)
+        formats, subtitles = self._extract_theplatform_smil(
+            media_url, video_id)
         self._sort_formats(formats)
         info.update({
             'id': video_id,
@@ -78,9 +94,11 @@ class AMCNetworksIE(ThePlatformIE):
         if ns_keys:
             ns = list(ns_keys)[0]
             series = theplatform_metadata.get(ns + '$show')
-            season_number = int_or_none(theplatform_metadata.get(ns + '$season'))
+            season_number = int_or_none(
+                theplatform_metadata.get(ns + '$season'))
             episode = theplatform_metadata.get(ns + '$episodeTitle')
-            episode_number = int_or_none(theplatform_metadata.get(ns + '$episode'))
+            episode_number = int_or_none(
+                theplatform_metadata.get(ns + '$episode'))
             if season_number:
                 title = 'Season %d - %s' % (season_number, title)
             if series:

youtube_dl/extractor/amp.py

@@ -7,15 +7,19 @@ from ..utils import (
     parse_iso8601,
     mimetype2ext,
     determine_ext,
+    ExtractorError,
 )
 
 
 class AMPIE(InfoExtractor):
     # parse Akamai Adaptive Media Player feed
     def _extract_feed_info(self, url):
-        item = self._download_json(
+        feed = self._download_json(
             url, None, 'Downloading Akamai AMP feed',
-            'Unable to download Akamai AMP feed')['channel']['item']
+            'Unable to download Akamai AMP feed')
+        item = feed.get('channel', {}).get('item')
+        if not item:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, feed['error']))
 
         video_id = item['guid']
 

youtube_dl/extractor/archiveorg.py

@@ -1,13 +1,13 @@
 from __future__ import unicode_literals
 
-from .jwplatform import JWPlatformBaseIE
+from .common import InfoExtractor
 from ..utils import (
     unified_strdate,
     clean_html,
 )
 
 
-class ArchiveOrgIE(JWPlatformBaseIE):
+class ArchiveOrgIE(InfoExtractor):
     IE_NAME = 'archive.org'
     IE_DESC = 'archive.org videos'
     _VALID_URL = r'https?://(?:www\.)?archive\.org/(?:details|embed)/(?P<id>[^/?#]+)(?:[?].*)?$'

youtube_dl/extractor/arkena.py

@@ -93,8 +93,7 @@ class ArkenaIE(InfoExtractor):
                 exts = (mimetype2ext(f.get('Type')), determine_ext(f_url, None))
                 if kind == 'm3u8' or 'm3u8' in exts:
                     formats.extend(self._extract_m3u8_formats(
-                        f_url, video_id, 'mp4',
-                        entry_protocol='m3u8' if is_live else 'm3u8_native',
+                        f_url, video_id, 'mp4', 'm3u8_native',
                         m3u8_id=kind, fatal=False, live=is_live))
                 elif kind == 'flash' or 'f4m' in exts:
                     formats.extend(self._extract_f4m_formats(

youtube_dl/extractor/atresplayer.py

@@ -90,7 +90,8 @@ class AtresPlayerIE(InfoExtractor):
             request, None, 'Logging in as %s' % username)
 
         error = self._html_search_regex(
-            r'(?s)<ul class="list_error">(.+?)</ul>', response, 'error', default=None)
+            r'(?s)<ul[^>]+class="[^"]*\blist_error\b[^"]*">(.+?)</ul>',
+            response, 'error', default=None)
         if error:
             raise ExtractorError(
                 'Unable to login: %s' % error, expected=True)
@@ -155,13 +156,17 @@ class AtresPlayerIE(InfoExtractor):
             if format_id == 'token' or not video_url.startswith('http'):
                 continue
             if 'geodeswowsmpra3player' in video_url:
-                f4m_path = video_url.split('smil:', 1)[-1].split('free_', 1)[0]
-                f4m_url = 'http://drg.antena3.com/{0}hds/es/sd.f4m'.format(f4m_path)
+                # f4m_path = video_url.split('smil:', 1)[-1].split('free_', 1)[0]
+                # f4m_url = 'http://drg.antena3.com/{0}hds/es/sd.f4m'.format(f4m_path)
                 # this videos are protected by DRM, the f4m downloader doesn't support them
                 continue
-            else:
-                f4m_url = video_url[:-9] + '/manifest.f4m'
-            formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False))
+            video_url_hd = video_url.replace('free_es', 'es')
+            formats.extend(self._extract_f4m_formats(
+                video_url_hd[:-9] + '/manifest.f4m', video_id, f4m_id='hds',
+                fatal=False))
+            formats.extend(self._extract_mpd_formats(
+                video_url_hd[:-9] + '/manifest.mpd', video_id, mpd_id='dash',
+                fatal=False))
         self._sort_formats(formats)
 
         path_data = player.get('pathData')

youtube_dl/extractor/atvat.py

@@ -0,0 +1,73 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    unescapeHTML,
+)
+
+
+class ATVAtIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?atv\.at/(?:[^/]+/){2}(?P<id>[dv]\d+)'
+    _TESTS = [{
+        'url': 'http://atv.at/aktuell/di-210317-2005-uhr/v1698449/',
+        'md5': 'c3b6b975fb3150fc628572939df205f2',
+        'info_dict': {
+            'id': '1698447',
+            'ext': 'mp4',
+            'title': 'DI, 21.03.17 | 20:05 Uhr 1/1',
+        }
+    }, {
+        'url': 'http://atv.at/aktuell/meinrad-knapp/d8416/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_data = self._parse_json(unescapeHTML(self._search_regex(
+            r'class="[^"]*jsb_video/FlashPlayer[^"]*"[^>]+data-jsb="([^"]+)"',
+            webpage, 'player data')), display_id)['config']['initial_video']
+
+        video_id = video_data['id']
+        video_title = video_data['title']
+
+        parts = []
+        for part in video_data.get('parts', []):
+            part_id = part['id']
+            part_title = part['title']
+
+            formats = []
+            for source in part.get('sources', []):
+                source_url = source.get('src')
+                if not source_url:
+                    continue
+                ext = determine_ext(source_url)
+                if ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        source_url, part_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
+                else:
+                    formats.append({
+                        'format_id': source.get('delivery'),
+                        'url': source_url,
+                    })
+            self._sort_formats(formats)
+
+            parts.append({
+                'id': part_id,
+                'title': part_title,
+                'thumbnail': part.get('preview_image_url'),
+                'duration': int_or_none(part.get('duration')),
+                'is_live': part.get('is_livestream'),
+                'formats': formats,
+            })
+
+        return {
+            '_type': 'multi_video',
+            'id': video_id,
+            'title': video_title,
+            'entries': parts,
+        }

youtube_dl/extractor/azmedien.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -5,6 +6,7 @@ import re
 from .common import InfoExtractor
 from .kaltura import KalturaIE
 from ..utils import (
+    get_element_by_class,
     get_element_by_id,
     strip_or_none,
     urljoin,
@@ -170,3 +172,42 @@ class AZMedienPlaylistIE(AZMedienBaseIE):
                 'video-title', webpage)), group='title')
 
         return self.playlist_result(entries, show_id, title)
+
+
+class AZMedienShowPlaylistIE(AZMedienBaseIE):
+    IE_DESC = 'AZ Medien show playlists'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?
+                        (?:
+                            telezueri\.ch|
+                            telebaern\.tv|
+                            telem1\.ch
+                        )/
+                        (?:
+                            all-episodes|
+                            alle-episoden
+                        )/
+                        (?P<id>[^/?#&]+)
+                    '''
+
+    _TEST = {
+        'url': 'http://www.telezueri.ch/all-episodes/astrotalk',
+        'info_dict': {
+            'id': 'astrotalk',
+            'title': 'TeleZüri: AstroTalk - alle episoden',
+            'description': 'md5:4c0f7e7d741d906004266e295ceb4a26',
+        },
+        'playlist_mincount': 13,
+    }
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+        webpage = self._download_webpage(url, playlist_id)
+        episodes = get_element_by_class('search-mobile-box', webpage)
+        entries = [self.url_result(
+            urljoin(url, m.group('url'))) for m in re.finditer(
+                r'<a[^>]+href=(["\'])(?P<url>(?:(?!\1).)+)\1', episodes)]
+        title = self._og_search_title(webpage, fatal=False)
+        description = self._og_search_description(webpage)
+        return self.playlist_result(entries, playlist_id, title, description)

youtube_dl/extractor/azubu.py

@@ -1,140 +0,0 @@
-from __future__ import unicode_literals
-
-import json
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    sanitized_Request,
-)
-
-
-class AzubuIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?azubu\.(?:tv|uol.com.br)/[^/]+#!/play/(?P<id>\d+)'
-    _TESTS = [
-        {
-            'url': 'http://www.azubu.tv/GSL#!/play/15575/2014-hot6-cup-last-big-match-ro8-day-1',
-            'md5': 'a88b42fcf844f29ad6035054bd9ecaf4',
-            'info_dict': {
-                'id': '15575',
-                'ext': 'mp4',
-                'title': '2014 HOT6 CUP LAST BIG MATCH Ro8 Day 1',
-                'description': 'md5:d06bdea27b8cc4388a90ad35b5c66c01',
-                'thumbnail': r're:^https?://.*\.jpe?g',
-                'timestamp': 1417523507.334,
-                'upload_date': '20141202',
-                'duration': 9988.7,
-                'uploader': 'GSL',
-                'uploader_id': 414310,
-                'view_count': int,
-            },
-        },
-        {
-            'url': 'http://www.azubu.tv/FnaticTV#!/play/9344/-fnatic-at-worlds-2014:-toyz---%22i-love-rekkles,-he-has-amazing-mechanics%22-',
-            'md5': 'b72a871fe1d9f70bd7673769cdb3b925',
-            'info_dict': {
-                'id': '9344',
-                'ext': 'mp4',
-                'title': 'Fnatic at Worlds 2014: Toyz - "I love Rekkles, he has amazing mechanics"',
-                'description': 'md5:4a649737b5f6c8b5c5be543e88dc62af',
-                'thumbnail': r're:^https?://.*\.jpe?g',
-                'timestamp': 1410530893.320,
-                'upload_date': '20140912',
-                'duration': 172.385,
-                'uploader': 'FnaticTV',
-                'uploader_id': 272749,
-                'view_count': int,
-            },
-            'skip': 'Channel offline',
-        },
-    ]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        data = self._download_json(
-            'http://www.azubu.tv/api/video/%s' % video_id, video_id)['data']
-
-        title = data['title'].strip()
-        description = data.get('description')
-        thumbnail = data.get('thumbnail')
-        view_count = data.get('view_count')
-        user = data.get('user', {})
-        uploader = user.get('username')
-        uploader_id = user.get('id')
-
-        stream_params = json.loads(data['stream_params'])
-
-        timestamp = float_or_none(stream_params.get('creationDate'), 1000)
-        duration = float_or_none(stream_params.get('length'), 1000)
-
-        renditions = stream_params.get('renditions') or []
-        video = stream_params.get('FLVFullLength') or stream_params.get('videoFullLength')
-        if video:
-            renditions.append(video)
-
-        if not renditions and not user.get('channel', {}).get('is_live', True):
-            raise ExtractorError('%s said: channel is offline.' % self.IE_NAME, expected=True)
-
-        formats = [{
-            'url': fmt['url'],
-            'width': fmt['frameWidth'],
-            'height': fmt['frameHeight'],
-            'vbr': float_or_none(fmt['encodingRate'], 1000),
-            'filesize': fmt['size'],
-            'vcodec': fmt['videoCodec'],
-            'container': fmt['videoContainer'],
-        } for fmt in renditions if fmt['url']]
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'timestamp': timestamp,
-            'duration': duration,
-            'uploader': uploader,
-            'uploader_id': uploader_id,
-            'view_count': view_count,
-            'formats': formats,
-        }
-
-
-class AzubuLiveIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?azubu\.(?:tv|uol.com.br)/(?P<id>[^/]+)$'
-
-    _TESTS = [{
-        'url': 'http://www.azubu.tv/MarsTVMDLen',
-        'only_matching': True,
-    }, {
-        'url': 'http://azubu.uol.com.br/adolfz',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        user = self._match_id(url)
-
-        info = self._download_json(
-            'http://api.azubu.tv/public/modules/last-video/{0}/info'.format(user),
-            user)['data']
-        if info['type'] != 'STREAM':
-            raise ExtractorError('{0} is not streaming live'.format(user), expected=True)
-
-        req = sanitized_Request(
-            'https://edge-elb.api.brightcove.com/playback/v1/accounts/3361910549001/videos/ref:' + info['reference_id'])
-        req.add_header('Accept', 'application/json;pk=BCpkADawqM1gvI0oGWg8dxQHlgT8HkdE2LnAlWAZkOlznO39bSZX726u4JqnDsK3MDXcO01JxXK2tZtJbgQChxgaFzEVdHRjaDoxaOu8hHOO8NYhwdxw9BzvgkvLUlpbDNUuDoc4E4wxDToV')
-        bc_info = self._download_json(req, user)
-        m3u8_url = next(source['src'] for source in bc_info['sources'] if source['container'] == 'M2TS')
-        formats = self._extract_m3u8_formats(m3u8_url, user, ext='mp4')
-        self._sort_formats(formats)
-
-        return {
-            'id': info['id'],
-            'title': self._live_title(info['title']),
-            'uploader_id': user,
-            'formats': formats,
-            'is_live': True,
-            'thumbnail': bc_info['poster'],
-        }

youtube_dl/extractor/bandcamp.py

@@ -209,6 +209,15 @@ class BandcampAlbumIE(InfoExtractor):
             'id': 'entropy-ep',
         },
         'playlist_mincount': 3,
+    }, {
+        # not all tracks have songs
+        'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
+        'info_dict': {
+            'id': 'we-are-the-plague',
+            'title': 'WE ARE THE PLAGUE',
+            'uploader_id': 'insulters',
+        },
+        'playlist_count': 2,
     }]
 
     def _real_extract(self, url):
@@ -217,12 +226,16 @@ class BandcampAlbumIE(InfoExtractor):
         album_id = mobj.group('album_id')
         playlist_id = album_id or uploader_id
         webpage = self._download_webpage(url, playlist_id)
-        tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
-        if not tracks_paths:
+        track_elements = re.findall(
+            r'(?s)<div[^>]*>(.*?<a[^>]+href="([^"]+?)"[^>]+itemprop="url"[^>]*>.*?)</div>', webpage)
+        if not track_elements:
             raise ExtractorError('The page doesn\'t contain any tracks')
+        # Only tracks with duration info have songs
         entries = [
             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
-            for t_path in tracks_paths]
+            for elem_content, t_path in track_elements
+            if self._html_search_meta('duration', elem_content, default=None)]
+
         title = self._html_search_regex(
             r'album_title\s*:\s*"((?:\\.|[^"\\])+?)"',
             webpage, 'title', fatal=False)

youtube_dl/extractor/bbc.py

@@ -225,6 +225,8 @@ class BBCCoUkIE(InfoExtractor):
         }
     ]
 
+    _USP_RE = r'/([^/]+?)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
+
     class MediaSelectionError(Exception):
         def __init__(self, id):
             self.id = id
@@ -336,6 +338,15 @@ class BBCCoUkIE(InfoExtractor):
                         formats.extend(self._extract_m3u8_formats(
                             href, programme_id, ext='mp4', entry_protocol='m3u8_native',
                             m3u8_id=format_id, fatal=False))
+                        if re.search(self._USP_RE, href):
+                            usp_formats = self._extract_m3u8_formats(
+                                re.sub(self._USP_RE, r'/\1.ism/\1.m3u8', href),
+                                programme_id, ext='mp4', entry_protocol='m3u8_native',
+                                m3u8_id=format_id, fatal=False)
+                            for f in usp_formats:
+                                if f.get('height') and f['height'] > 720:
+                                    continue
+                                formats.append(f)
                     elif transfer_format == 'hds':
                         formats.extend(self._extract_f4m_formats(
                             href, programme_id, f4m_id=format_id, fatal=False))
@@ -350,7 +361,7 @@ class BBCCoUkIE(InfoExtractor):
                             fmt.update({
                                 'width': width,
                                 'height': height,
-                                'vbr': bitrate,
+                                'tbr': bitrate,
                                 'vcodec': encoding,
                             })
                         else:
@@ -359,7 +370,7 @@ class BBCCoUkIE(InfoExtractor):
                                 'acodec': encoding,
                                 'vcodec': 'none',
                             })
-                        if protocol == 'http':
+                        if protocol in ('http', 'https'):
                             # Direct link
                             fmt.update({
                                 'url': href,
@@ -378,6 +389,8 @@ class BBCCoUkIE(InfoExtractor):
                                 'rtmp_live': False,
                                 'ext': 'flv',
                             })
+                        else:
+                            continue
                         formats.append(fmt)
             elif kind == 'captions':
                 subtitles = self.extract_subtitles(media, programme_id)
@@ -396,7 +409,7 @@ class BBCCoUkIE(InfoExtractor):
                 description = smp_config['summary']
                 for item in smp_config['items']:
                     kind = item['kind']
-                    if kind != 'programme' and kind != 'radioProgramme':
+                    if kind not in ('programme', 'radioProgramme'):
                         continue
                     programme_id = item.get('vpid')
                     duration = int_or_none(item.get('duration'))
@@ -437,7 +450,7 @@ class BBCCoUkIE(InfoExtractor):
 
         for item in self._extract_items(playlist):
             kind = item.get('kind')
-            if kind != 'programme' and kind != 'radioProgramme':
+            if kind not in ('programme', 'radioProgramme'):
                 continue
             title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
             description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)

youtube_dl/extractor/bellmedia.py

@@ -21,10 +21,11 @@ class BellMediaIE(InfoExtractor):
                 animalplanet|
                 bravo|
                 mtv|
-                space
+                space|
+                etalk
             )\.ca|
             much\.com
-        )/.*?(?:\bvid=|-vid|~|%7E|/(?:episode)?)(?P<id>[0-9]{6})'''
+        )/.*?(?:\bvid(?:eoid)?=|-vid|~|%7E|/(?:episode)?)(?P<id>[0-9]{6,})'''
     _TESTS = [{
         'url': 'http://www.ctv.ca/video/player?vid=706966',
         'md5': 'ff2ebbeae0aa2dcc32a830c3fd69b7b0',
@@ -55,6 +56,12 @@ class BellMediaIE(InfoExtractor):
     }, {
         'url': 'http://www.much.com/shows/the-almost-impossible-gameshow/928979/episode-6',
         'only_matching': True,
+    }, {
+        'url': 'http://www.ctv.ca/DCs-Legends-of-Tomorrow/Video/S2E11-Turncoat-vid1051430',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.etalk.ca/video?videoid=663455',
+        'only_matching': True,
     }]
     _DOMAINS = {
         'thecomedynetwork': 'comedy',
@@ -62,6 +69,7 @@ class BellMediaIE(InfoExtractor):
         'sciencechannel': 'discsci',
         'investigationdiscovery': 'invdisc',
         'animalplanet': 'aniplan',
+        'etalk': 'ctv',
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/bloomberg.py

@@ -33,6 +33,10 @@ class BloombergIE(InfoExtractor):
         'params': {
             'format': 'best[format_id^=hds]',
         },
+    }, {
+        # data-bmmrid=
+        'url': 'https://www.bloomberg.com/politics/articles/2017-02-08/le-pen-aide-briefed-french-central-banker-on-plan-to-print-money',
+        'only_matching': True,
     }, {
         'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
         'only_matching': True,
@@ -45,9 +49,10 @@ class BloombergIE(InfoExtractor):
         name = self._match_id(url)
         webpage = self._download_webpage(url, name)
         video_id = self._search_regex(
-            (r'["\']bmmrId["\']\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1',
-             r'videoId\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1'),
-            webpage, 'id', group='url', default=None)
+            (r'["\']bmmrId["\']\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
+             r'videoId\s*:\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
+             r'data-bmmrid=(["\'])(?P<id>(?:(?!\1).)+)\1'),
+            webpage, 'id', group='id', default=None)
         if not video_id:
             bplayer_data = self._parse_json(self._search_regex(
                 r'BPlayer\(null,\s*({[^;]+})\);', webpage, 'id'), name)

youtube_dl/extractor/bostonglobe.py

@@ -0,0 +1,72 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+from ..utils import (
+    extract_attributes,
+)
+
+
+class BostonGlobeIE(InfoExtractor):
+    _VALID_URL = r'(?i)https?://(?:www\.)?bostonglobe\.com/.*/(?P<id>[^/]+)/\w+(?:\.html)?'
+    _TESTS = [
+        {
+            'url': 'http://www.bostonglobe.com/metro/2017/02/11/tree-finally-succumbs-disease-leaving-hole-neighborhood/h1b4lviqzMTIn9sVy8F3gP/story.html',
+            'md5': '0a62181079c85c2d2b618c9a738aedaf',
+            'info_dict': {
+                'title': 'A tree finally succumbs to disease, leaving a hole in a neighborhood',
+                'id': '5320421710001',
+                'ext': 'mp4',
+                'description': 'It arrived as a sapling when the Back Bay was in its infancy, a spindly American elm tamped down into a square of dirt cut into the brick sidewalk of 1880s Marlborough Street, no higher than the first bay window of the new brownstone behind it.',
+                'timestamp': 1486877593,
+                'upload_date': '20170212',
+                'uploader_id': '245991542',
+            },
+        },
+        {
+            # Embedded youtube video; we hand it off to the Generic extractor.
+            'url': 'https://www.bostonglobe.com/lifestyle/names/2017/02/17/does-ben-affleck-play-matt-damon-favorite-version-batman/ruqkc9VxKBYmh5txn1XhSI/story.html',
+            'md5': '582b40327089d5c0c949b3c54b13c24b',
+            'info_dict': {
+                'title': "Who Is Matt Damon's Favorite Batman?",
+                'id': 'ZW1QCnlA6Qc',
+                'ext': 'mp4',
+                'upload_date': '20170217',
+                'description': 'md5:3b3dccb9375867e0b4d527ed87d307cb',
+                'uploader': 'The Late Late Show with James Corden',
+                'uploader_id': 'TheLateLateShow',
+            },
+            'expected_warnings': ['404'],
+        },
+    ]
+
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+        webpage = self._download_webpage(url, page_id)
+
+        page_title = self._og_search_title(webpage, default=None)
+
+        # <video data-brightcove-video-id="5320421710001" data-account="245991542" data-player="SJWAiyYWg" data-embed="default" class="video-js" controls itemscope itemtype="http://schema.org/VideoObject">
+        entries = []
+        for video in re.findall(r'(?i)(<video[^>]+>)', webpage):
+            attrs = extract_attributes(video)
+
+            video_id = attrs.get('data-brightcove-video-id')
+            account_id = attrs.get('data-account')
+            player_id = attrs.get('data-player')
+            embed = attrs.get('data-embed')
+
+            if video_id and account_id and player_id and embed:
+                entries.append(
+                    'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
+                    % (account_id, player_id, embed, video_id))
+
+        if len(entries) == 0:
+            return self.url_result(url, 'Generic')
+        elif len(entries) == 1:
+            return self.url_result(entries[0], 'BrightcoveNew')
+        else:
+            return self.playlist_from_matches(entries, page_id, page_title, ie='BrightcoveNew')

youtube_dl/extractor/brightcove.py

@@ -17,6 +17,7 @@ from ..compat import (
 from ..utils import (
     determine_ext,
     ExtractorError,
+    extract_attributes,
     find_xpath_attr,
     fix_xml_ampersands,
     float_or_none,
@@ -109,6 +110,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                 'upload_date': '20140827',
                 'uploader_id': '710858724001',
             },
+            'skip': 'Video gone',
         },
         {
             # playlist with 'videoList'
@@ -129,6 +131,12 @@ class BrightcoveLegacyIE(InfoExtractor):
             },
             'playlist_mincount': 10,
         },
+        {
+            # playerID inferred from bcpid
+            # from http://www.un.org/chinese/News/story.asp?NewsID=27724
+            'url': 'https://link.brightcove.com/services/player/bcpid1722935254001/?bctid=5360463607001&autoStart=false&secureConnections=true&width=650&height=350',
+            'only_matching': True,  # Tested in GenericIE
+        }
     ]
     FLV_VCODECS = {
         1: 'SORENSON',
@@ -191,6 +199,16 @@ class BrightcoveLegacyIE(InfoExtractor):
         # These fields hold the id of the video
         videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList')
         if videoPlayer is not None:
+            if isinstance(videoPlayer, list):
+                videoPlayer = videoPlayer[0]
+            videoPlayer = videoPlayer.strip()
+            # UUID is also possible for videoPlayer (e.g.
+            # http://www.popcornflix.com/hoodies-vs-hooligans/7f2d2b87-bbf2-4623-acfb-ea942b4f01dd
+            # or http://www8.hp.com/cn/zh/home.html)
+            if not (re.match(
+                    r'^(?:\d+|[\da-fA-F]{8}-?[\da-fA-F]{4}-?[\da-fA-F]{4}-?[\da-fA-F]{4}-?[\da-fA-F]{12})$',
+                    videoPlayer) or videoPlayer.startswith('ref:')):
+                return None
             params['@videoPlayer'] = videoPlayer
         linkBase = find_param('linkBaseURL')
         if linkBase is not None:
@@ -254,9 +272,13 @@ class BrightcoveLegacyIE(InfoExtractor):
         if matches:
             return list(filter(None, [cls._build_brighcove_url(m) for m in matches]))
 
-        return list(filter(None, [
-            cls._build_brighcove_url_from_js(custom_bc)
-            for custom_bc in re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)]))
+        matches = re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)
+        if matches:
+            return list(filter(None, [
+                cls._build_brighcove_url_from_js(custom_bc)
+                for custom_bc in matches]))
+        return [src for _, src in re.findall(
+            r'<iframe[^>]+src=([\'"])((?:https?:)?//link\.brightcove\.com/services/player/(?!\1).+)\1', webpage)]
 
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
@@ -273,6 +295,10 @@ class BrightcoveLegacyIE(InfoExtractor):
         if videoPlayer:
             # We set the original url as the default 'Referer' header
             referer = smuggled_data.get('Referer', url)
+            if 'playerID' not in query:
+                mobj = re.search(r'/bcpid(\d+)', url)
+                if mobj is not None:
+                    query['playerID'] = [mobj.group(1)]
             return self._get_video_info(
                 videoPlayer[0], query, referer=referer)
         elif 'playerKey' in query:
@@ -472,17 +498,18 @@ class BrightcoveNewIE(InfoExtractor):
     }]
 
     @staticmethod
-    def _extract_url(webpage):
-        urls = BrightcoveNewIE._extract_urls(webpage)
+    def _extract_url(ie, webpage):
+        urls = BrightcoveNewIE._extract_urls(ie, webpage)
         return urls[0] if urls else None
 
     @staticmethod
-    def _extract_urls(webpage):
+    def _extract_urls(ie, webpage):
         # Reference:
         # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe
-        # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript
-        # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
-        # 4. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player
+        # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#tag
+        # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript
+        # 4. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/in-page-embed-player-implementation.html
+        # 5. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player
 
         entries = []
 
@@ -491,26 +518,55 @@ class BrightcoveNewIE(InfoExtractor):
                 r'<iframe[^>]+src=(["\'])((?:https?:)?//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage):
             entries.append(url if url.startswith('http') else 'http:' + url)
 
-        # Look for embed_in_page embeds [2]
-        for video_id, account_id, player_id, embed in re.findall(
-                # According to examples from [3] it's unclear whether video id
-                # may be optional and what to do when it is
-                # According to [4] data-video-id may be prefixed with ref:
-                r'''(?sx)
-                    <video[^>]+
-                        data-video-id=["\'](\d+|ref:[^"\']+)["\'][^>]*>.*?
-                    </video>.*?
-                    <script[^>]+
-                        src=["\'](?:https?:)?//players\.brightcove\.net/
-                        (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js
+        # Look for <video> tags [2] and embed_in_page embeds [3]
+        # [2] looks like:
+        for video, script_tag, account_id, player_id, embed in re.findall(
+                r'''(?isx)
+                    (<video\s+[^>]*data-video-id=['"]?[^>]+>)
+                    (?:.*?
+                        (<script[^>]+
+                            src=["\'](?:https?:)?//players\.brightcove\.net/
+                            (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js
+                        )
+                    )?
                 ''', webpage):
-            entries.append(
-                'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
-                % (account_id, player_id, embed, video_id))
+            attrs = extract_attributes(video)
+
+            # According to examples from [4] it's unclear whether video id
+            # may be optional and what to do when it is
+            video_id = attrs.get('data-video-id')
+            if not video_id:
+                continue
+
+            account_id = account_id or attrs.get('data-account')
+            if not account_id:
+                continue
+
+            player_id = player_id or attrs.get('data-player') or 'default'
+            embed = embed or attrs.get('data-embed') or 'default'
+
+            bc_url = 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s' % (
+                account_id, player_id, embed, video_id)
+
+            # Some brightcove videos may be embedded with video tag only and
+            # without script tag or any mentioning of brightcove at all. Such
+            # embeds are considered ambiguous since they are matched based only
+            # on data-video-id and data-account attributes and in the wild may
+            # not be brightcove embeds at all. Let's check reconstructed
+            # brightcove URLs in case of such embeds and only process valid
+            # ones. By this we ensure there is indeed a brightcove embed.
+            if not script_tag and not ie._is_valid_url(
+                    bc_url, video_id, 'possible brightcove video'):
+                continue
+
+            entries.append(bc_url)
 
         return entries
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
+        self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
+
         account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
 
         webpage = self._download_webpage(
@@ -540,8 +596,10 @@ class BrightcoveNewIE(InfoExtractor):
         except ExtractorError as e:
             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
                 json_data = self._parse_json(e.cause.read().decode(), video_id)[0]
-                raise ExtractorError(
-                    json_data.get('message') or json_data['error_code'], expected=True)
+                message = json_data.get('message') or json_data['error_code']
+                if json_data.get('error_subcode') == 'CLIENT_GEO':
+                    self.raise_geo_restricted(msg=message)
+                raise ExtractorError(message, expected=True)
             raise
 
         title = json_data['name'].strip()

youtube_dl/extractor/canalplus.py

@@ -7,8 +7,8 @@ from .common import InfoExtractor
 from ..compat import compat_urllib_parse_urlparse
 from ..utils import (
     dict_get,
-    ExtractorError,
-    HEADRequest,
+    # ExtractorError,
+    # HEADRequest,
     int_or_none,
     qualities,
     remove_end,
@@ -27,6 +27,7 @@ class CanalplusIE(InfoExtractor):
                                     (?:www\.)?d8\.tv|
                                     (?:www\.)?c8\.fr|
                                     (?:www\.)?d17\.tv|
+                                    (?:(?:football|www)\.)?cstar\.fr|
                                     (?:www\.)?itele\.fr
                                 )/(?:(?:[^/]+/)*(?P<display_id>[^/?#&]+))?(?:\?.*\bvid=(?P<vid>\d+))?|
                                 player\.canalplus\.fr/#/(?P<id>\d+)
@@ -40,9 +41,13 @@ class CanalplusIE(InfoExtractor):
         'd8': 'd8',
         'c8': 'd8',
         'd17': 'd17',
+        'cstar': 'd17',
         'itele': 'itele',
     }
 
+    # Only works for direct mp4 URLs
+    _GEO_COUNTRIES = ['FR']
+
     _TESTS = [{
         'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1192814',
         'info_dict': {
@@ -54,6 +59,7 @@ class CanalplusIE(InfoExtractor):
             'upload_date': '20160702',
         },
     }, {
+        # geo restricted, bypassed
         'url': 'http://www.piwiplus.fr/videos-piwi/pid1405-le-labyrinthe-boing-super-ranger.html?vid=1108190',
         'info_dict': {
             'id': '1108190',
@@ -63,19 +69,20 @@ class CanalplusIE(InfoExtractor):
             'description': 'md5:4cea7a37153be42c1ba2c1d3064376ff',
             'upload_date': '20140724',
         },
-        'skip': 'Only works from France',
+        'expected_warnings': ['HTTP Error 403: Forbidden'],
     }, {
-        'url': 'http://www.c8.fr/c8-divertissement/ms-touche-pas-a-mon-poste/pid6318-videos-integrales.html',
-        'md5': '4b47b12b4ee43002626b97fad8fb1de5',
+        # geo restricted, bypassed
+        'url': 'http://www.c8.fr/c8-divertissement/ms-touche-pas-a-mon-poste/pid6318-videos-integrales.html?vid=1443684',
+        'md5': 'bb6f9f343296ab7ebd88c97b660ecf8d',
         'info_dict': {
-            'id': '1420213',
+            'id': '1443684',
             'display_id': 'pid6318-videos-integrales',
             'ext': 'mp4',
-            'title': 'TPMP ! Même le matin - Les 35H de Baba - 14/10/2016',
-            'description': 'md5:f96736c1b0ffaa96fd5b9e60ad871799',
-            'upload_date': '20161014',
+            'title': 'Guess my iep ! - TPMP - 07/04/2017',
+            'description': 'md5:6f005933f6e06760a9236d9b3b5f17fa',
+            'upload_date': '20170407',
         },
-        'skip': 'Only works from France',
+        'expected_warnings': ['HTTP Error 403: Forbidden'],
     }, {
         'url': 'http://www.itele.fr/chroniques/invite-michael-darmon/rachida-dati-nicolas-sarkozy-est-le-plus-en-phase-avec-les-inquietudes-des-francais-171510',
         'info_dict': {
@@ -86,6 +93,19 @@ class CanalplusIE(InfoExtractor):
             'description': 'Chaque matin du lundi au vendredi, Michaël Darmon reçoit un invité politique à 8h25.',
             'upload_date': '20161014',
         },
+    }, {
+        'url': 'http://football.cstar.fr/cstar-minisite-foot/pid7566-feminines-videos.html?vid=1416769',
+        'info_dict': {
+            'id': '1416769',
+            'display_id': 'pid7566-feminines-videos',
+            'ext': 'mp4',
+            'title': 'France - Albanie : les temps forts de la soirée - 20/09/2016',
+            'description': 'md5:c3f30f2aaac294c1c969b3294de6904e',
+            'upload_date': '20160921',
+        },
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'http://m.canalplus.fr/?vid=1398231',
         'only_matching': True,
@@ -119,15 +139,15 @@ class CanalplusIE(InfoExtractor):
 
         preference = qualities(['MOBILE', 'BAS_DEBIT', 'HAUT_DEBIT', 'HD'])
 
-        fmt_url = next(iter(media.get('VIDEOS')))
-        if '/geo' in fmt_url.lower():
-            response = self._request_webpage(
-                HEADRequest(fmt_url), video_id,
-                'Checking if the video is georestricted')
-            if '/blocage' in response.geturl():
-                raise ExtractorError(
-                    'The video is not available in your country',
-                    expected=True)
+        # _, fmt_url = next(iter(media['VIDEOS'].items()))
+        # if '/geo' in fmt_url.lower():
+        #     response = self._request_webpage(
+        #         HEADRequest(fmt_url), video_id,
+        #         'Checking if the video is georestricted')
+        #     if '/blocage' in response.geturl():
+        #         raise ExtractorError(
+        #             'The video is not available in your country',
+        #             expected=True)
 
         formats = []
         for format_id, format_url in media['VIDEOS'].items():

youtube_dl/extractor/canvas.py

@@ -7,6 +7,7 @@ from ..utils import float_or_none
 
 
 class CanvasIE(InfoExtractor):
+    IE_DESC = 'canvas.be and een.be'
     _VALID_URL = r'https?://(?:www\.)?(?P<site_id>canvas|een)\.be/(?:[^/]+/)*(?P<id>[^/?#&]+)'
     _TESTS = [{
         'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',

youtube_dl/extractor/cda.py

@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import codecs
 import re
 
 from .common import InfoExtractor
@@ -96,6 +97,10 @@ class CDAIE(InfoExtractor):
             if not video or 'file' not in video:
                 self.report_warning('Unable to extract %s version information' % version)
                 return
+            if video['file'].startswith('uggc'):
+                video['file'] = codecs.decode(video['file'], 'rot_13')
+                if video['file'].endswith('adc.mp4'):
+                    video['file'] = video['file'].replace('adc.mp4', '.mp4')
             f = {
                 'url': video['file'],
             }

youtube_dl/extractor/ceskatelevize.py

@@ -12,19 +12,21 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     sanitized_Request,
+    unescapeHTML,
     urlencode_postdata,
+    USER_AGENTS,
 )
 
 
 class CeskaTelevizeIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/(porady|ivysilani)/(?:[^/]+/)*(?P<id>[^/#?]+)/*(?:[#?].*)?$'
+    _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/ivysilani/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
     _TESTS = [{
         'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
         'info_dict': {
-            'id': '61924494876951776',
+            'id': '61924494877246241',
             'ext': 'mp4',
-            'title': 'Hyde Park Civilizace',
-            'description': 'md5:fe93f6eda372d150759d11644ebbfb4a',
+            'title': 'Hyde Park Civilizace: Život v Grónsku',
+            'description': 'md5:3fec8f6bb497be5cdb0c9e8781076626',
             'thumbnail': r're:^https?://.*\.jpg',
             'duration': 3350,
         },
@@ -61,40 +63,12 @@ class CeskaTelevizeIE(InfoExtractor):
         },
         'skip': 'Georestricted to Czech Republic',
     }, {
-        # video with 18+ caution trailer
-        'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
-        'info_dict': {
-            'id': '215562210900007-bogotart',
-            'title': 'Queer: Bogotart',
-            'description': 'Alternativní průvodce současným queer světem',
-        },
-        'playlist': [{
-            'info_dict': {
-                'id': '61924494876844842',
-                'ext': 'mp4',
-                'title': 'Queer: Bogotart (Varování 18+)',
-                'duration': 10.2,
-            },
-        }, {
-            'info_dict': {
-                'id': '61924494877068022',
-                'ext': 'mp4',
-                'title': 'Queer: Bogotart (Queer)',
-                'thumbnail': r're:^https?://.*\.jpg',
-                'duration': 1558.3,
-            },
-        }],
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
+        'url': 'http://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php?hash=d6a3e1370d2e4fa76296b90bad4dfc19673b641e&IDEC=217 562 22150/0004&channelID=1&width=100%25',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
-
-        mobj = re.match(self._VALID_URL, url)
-        playlist_id = mobj.group('id')
+        playlist_id = self._match_id(url)
 
         webpage = self._download_webpage(url, playlist_id)
 
@@ -102,82 +76,126 @@ class CeskaTelevizeIE(InfoExtractor):
         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 
-        typ = self._html_search_regex(
-            r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
-        episode_id = self._html_search_regex(
-            r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
+        type_ = None
+        episode_id = None
+
+        playlist = self._parse_json(
+            self._search_regex(
+                r'getPlaylistUrl\(\[({.+?})\]', webpage, 'playlist',
+                default='{}'), playlist_id)
+        if playlist:
+            type_ = playlist.get('type')
+            episode_id = playlist.get('id')
+
+        if not type_:
+            type_ = self._html_search_regex(
+                r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],',
+                webpage, 'type')
+        if not episode_id:
+            episode_id = self._html_search_regex(
+                r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],',
+                webpage, 'episode_id')
 
         data = {
-            'playlist[0][type]': typ,
+            'playlist[0][type]': type_,
             'playlist[0][id]': episode_id,
             'requestUrl': compat_urllib_parse_urlparse(url).path,
             'requestSource': 'iVysilani',
         }
 
-        req = sanitized_Request(
-            'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
-            data=urlencode_postdata(data))
-
-        req.add_header('Content-type', 'application/x-www-form-urlencoded')
-        req.add_header('x-addr', '127.0.0.1')
-        req.add_header('X-Requested-With', 'XMLHttpRequest')
-        req.add_header('Referer', url)
-
-        playlistpage = self._download_json(req, playlist_id)
-
-        playlist_url = playlistpage['url']
-        if playlist_url == 'error_region':
-            raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
-
-        req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
-        req.add_header('Referer', url)
-
-        playlist_title = self._og_search_title(webpage, default=None)
-        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' if is_live else 'm3u8_native',
-                    fatal=False))
-            self._sort_formats(formats)
 
-            item_id = item.get('id') or item['assetId']
-            title = item['title']
+        for user_agent in (None, USER_AGENTS['Safari']):
+            req = sanitized_Request(
+                'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
+                data=urlencode_postdata(data))
 
-            duration = float_or_none(item.get('duration'))
-            thumbnail = item.get('previewImageUrl')
+            req.add_header('Content-type', 'application/x-www-form-urlencoded')
+            req.add_header('x-addr', '127.0.0.1')
+            req.add_header('X-Requested-With', 'XMLHttpRequest')
+            if user_agent:
+                req.add_header('User-Agent', user_agent)
+            req.add_header('Referer', url)
 
-            subtitles = {}
-            if item.get('type') == 'VOD':
-                subs = item.get('subtitles')
-                if subs:
-                    subtitles = self.extract_subtitles(episode_id, subs)
+            playlistpage = self._download_json(req, playlist_id, fatal=False)
 
-            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)
+            if not playlistpage:
+                continue
 
-            entries.append({
-                'id': item_id,
-                'title': final_title,
-                'description': playlist_description if playlist_len == 1 else None,
-                'thumbnail': thumbnail,
-                'duration': duration,
-                'formats': formats,
-                'subtitles': subtitles,
-                'is_live': is_live,
-            })
+            playlist_url = playlistpage['url']
+            if playlist_url == 'error_region':
+                raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
+
+            req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
+            req.add_header('Referer', url)
+
+            playlist_title = self._og_search_title(webpage, default=None)
+            playlist_description = self._og_search_description(webpage, default=None)
+
+            playlist = self._download_json(req, playlist_id, fatal=False)
+            if not playlist:
+                continue
+
+            playlist = playlist.get('playlist')
+            if not isinstance(playlist, list):
+                continue
+
+            playlist_len = len(playlist)
+
+            for num, item in enumerate(playlist):
+                is_live = item.get('type') == 'LIVE'
+                formats = []
+                for format_id, stream_url in item.get('streamUrls', {}).items():
+                    if 'playerType=flash' in stream_url:
+                        stream_formats = self._extract_m3u8_formats(
+                            stream_url, playlist_id, 'mp4', 'm3u8_native',
+                            m3u8_id='hls-%s' % format_id, fatal=False)
+                    else:
+                        stream_formats = self._extract_mpd_formats(
+                            stream_url, playlist_id,
+                            mpd_id='dash-%s' % format_id, fatal=False)
+                    # See https://github.com/rg3/youtube-dl/issues/12119#issuecomment-280037031
+                    if format_id == 'audioDescription':
+                        for f in stream_formats:
+                            f['source_preference'] = -10
+                    formats.extend(stream_formats)
+
+                if user_agent and len(entries) == playlist_len:
+                    entries[num]['formats'].extend(formats)
+                    continue
+
+                item_id = item.get('id') or item['assetId']
+                title = item['title']
+
+                duration = float_or_none(item.get('duration'))
+                thumbnail = item.get('previewImageUrl')
+
+                subtitles = {}
+                if item.get('type') == 'VOD':
+                    subs = item.get('subtitles')
+                    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': final_title,
+                    'description': playlist_description if playlist_len == 1 else None,
+                    'thumbnail': thumbnail,
+                    'duration': duration,
+                    'formats': formats,
+                    'subtitles': subtitles,
+                    'is_live': is_live,
+                })
+
+        for e in entries:
+            self._sort_formats(e['formats'])
 
         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 
@@ -215,3 +233,47 @@ class CeskaTelevizeIE(InfoExtractor):
                     yield line
 
         return '\r\n'.join(_fix_subtitle(subtitles))
+
+
+class CeskaTelevizePoradyIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/porady/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
+    _TESTS = [{
+        # video with 18+ caution trailer
+        'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
+        'info_dict': {
+            'id': '215562210900007-bogotart',
+            'title': 'Queer: Bogotart',
+            'description': 'Alternativní průvodce současným queer světem',
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': '61924494876844842',
+                'ext': 'mp4',
+                'title': 'Queer: Bogotart (Varování 18+)',
+                'duration': 10.2,
+            },
+        }, {
+            'info_dict': {
+                'id': '61924494877068022',
+                'ext': 'mp4',
+                'title': 'Queer: Bogotart (Queer)',
+                'thumbnail': r're:^https?://.*\.jpg',
+                'duration': 1558.3,
+            },
+        }],
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        data_url = unescapeHTML(self._search_regex(
+            r'<span[^>]*\bdata-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'iframe player url', group='url'))
+
+        return self.url_result(data_url, ie=CeskaTelevizeIE.ie_key())

youtube_dl/extractor/channel9.py

@@ -4,62 +4,62 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
     ExtractorError,
-    parse_filesize,
+    int_or_none,
+    parse_iso8601,
     qualities,
+    unescapeHTML,
 )
 
 
 class Channel9IE(InfoExtractor):
-    '''
-    Common extractor for channel9.msdn.com.
-
-    The type of provided URL (video or playlist) is determined according to
-    meta Search.PageType from web page HTML rather than URL itself, as it is
-    not always possible to do.
-    '''
     IE_DESC = 'Channel 9'
     IE_NAME = 'channel9'
-    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)'
+    _VALID_URL = r'https?://(?:www\.)?(?:channel9\.msdn\.com|s\.ch9\.ms)/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)'
 
     _TESTS = [{
         'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
-        'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
+        'md5': '32083d4eaf1946db6d454313f44510ca',
         'info_dict': {
-            'id': 'Events/TechEd/Australia/2013/KOS002',
-            'ext': 'mp4',
+            'id': '6c413323-383a-49dc-88f9-a22800cab024',
+            'ext': 'wmv',
             'title': 'Developer Kick-Off Session: Stuff We Love',
-            'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
+            'description': 'md5:b80bf9355a503c193aff7ec6cd5a7731',
             'duration': 4576,
-            'thumbnail': r're:http://.*\.jpg',
+            'thumbnail': r're:https?://.*\.jpg',
+            'timestamp': 1377717420,
+            'upload_date': '20130828',
             'session_code': 'KOS002',
-            'session_day': 'Day 1',
             'session_room': 'Arena 1A',
-            'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug',
-                                 'Mads Kristensen'],
+            'session_speakers': ['Andrew Coates', 'Brady Gaster', 'Mads Kristensen', 'Ed Blankenship', 'Patrick Klug'],
         },
     }, {
         'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
-        'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
+        'md5': 'dcf983ee6acd2088e7188c3cf79b46bc',
         'info_dict': {
-            'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
-            'ext': 'mp4',
+            'id': 'fe8e435f-bb93-4e01-8e97-a28c01887024',
+            'ext': 'wmv',
             'title': 'Self-service BI with Power BI - nuclear testing',
-            'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
+            'description': 'md5:2d17fec927fc91e9e17783b3ecc88f54',
             'duration': 1540,
-            'thumbnail': r're:http://.*\.jpg',
+            'thumbnail': r're:https?://.*\.jpg',
+            'timestamp': 1386381991,
+            'upload_date': '20131207',
             'authors': ['Mike Wilmot'],
         },
     }, {
         # low quality mp4 is best
         'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
         'info_dict': {
-            'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+            'id': '33ad69d2-6a4e-4172-83a1-a523013dec76',
             'ext': 'mp4',
             'title': 'Ranges for the Standard Library',
-            'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
+            'description': 'md5:9895e0a9fd80822d2f01c454b8f4a372',
             'duration': 5646,
-            'thumbnail': r're:http://.*\.jpg',
+            'thumbnail': r're:https?://.*\.jpg',
+            'upload_date': '20150930',
+            'timestamp': 1443640735,
         },
         'params': {
             'skip_download': True,
@@ -70,7 +70,7 @@ class Channel9IE(InfoExtractor):
             'id': 'Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b',
             'title': 'Channel 9',
         },
-        'playlist_count': 2,
+        'playlist_mincount': 100,
     }, {
         'url': 'https://channel9.msdn.com/Events/DEVintersection/DEVintersection-2016/RSS',
         'only_matching': True,
@@ -81,189 +81,6 @@ class Channel9IE(InfoExtractor):
 
     _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
 
-    def _formats_from_html(self, html):
-        FORMAT_REGEX = r'''
-            (?x)
-            <a\s+href="(?P<url>[^"]+)">(?P<quality>[^<]+)</a>\s*
-            <span\s+class="usage">\((?P<note>[^\)]+)\)</span>\s*
-            (?:<div\s+class="popup\s+rounded">\s*
-            <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s*
-            </div>)?                                                # File size part may be missing
-        '''
-        quality = qualities((
-            'MP3', 'MP4',
-            'Low Quality WMV', 'Low Quality MP4',
-            'Mid Quality WMV', 'Mid Quality MP4',
-            'High Quality WMV', 'High Quality MP4'))
-        formats = [{
-            'url': x.group('url'),
-            'format_id': x.group('quality'),
-            'format_note': x.group('note'),
-            'format': '%s (%s)' % (x.group('quality'), x.group('note')),
-            'filesize_approx': parse_filesize(x.group('filesize')),
-            'quality': quality(x.group('quality')),
-            'vcodec': 'none' if x.group('note') == 'Audio only' else None,
-        } for x in list(re.finditer(FORMAT_REGEX, html))]
-
-        self._sort_formats(formats)
-
-        return formats
-
-    def _extract_title(self, html):
-        title = self._html_search_meta('title', html, 'title')
-        if title is None:
-            title = self._og_search_title(html)
-            TITLE_SUFFIX = ' (Channel 9)'
-            if title is not None and title.endswith(TITLE_SUFFIX):
-                title = title[:-len(TITLE_SUFFIX)]
-        return title
-
-    def _extract_description(self, html):
-        DESCRIPTION_REGEX = r'''(?sx)
-            <div\s+class="entry-content">\s*
-            <div\s+id="entry-body">\s*
-            (?P<description>.+?)\s*
-            </div>\s*
-            </div>
-        '''
-        m = re.search(DESCRIPTION_REGEX, html)
-        if m is not None:
-            return m.group('description')
-        return self._html_search_meta('description', html, 'description')
-
-    def _extract_duration(self, html):
-        m = re.search(r'"length": *"(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html)
-        return ((int(m.group('hours')) * 60 * 60) + (int(m.group('minutes')) * 60) + int(m.group('seconds'))) if m else None
-
-    def _extract_slides(self, html):
-        m = re.search(r'<a href="(?P<slidesurl>[^"]+)" class="slides">Slides</a>', html)
-        return m.group('slidesurl') if m is not None else None
-
-    def _extract_zip(self, html):
-        m = re.search(r'<a href="(?P<zipurl>[^"]+)" class="zip">Zip</a>', html)
-        return m.group('zipurl') if m is not None else None
-
-    def _extract_avg_rating(self, html):
-        m = re.search(r'<p class="avg-rating">Avg Rating: <span>(?P<avgrating>[^<]+)</span></p>', html)
-        return float(m.group('avgrating')) if m is not None else 0
-
-    def _extract_rating_count(self, html):
-        m = re.search(r'<div class="rating-count">\((?P<ratingcount>[^<]+)\)</div>', html)
-        return int(self._fix_count(m.group('ratingcount'))) if m is not None else 0
-
-    def _extract_view_count(self, html):
-        m = re.search(r'<li class="views">\s*<span class="count">(?P<viewcount>[^<]+)</span> Views\s*</li>', html)
-        return int(self._fix_count(m.group('viewcount'))) if m is not None else 0
-
-    def _extract_comment_count(self, html):
-        m = re.search(r'<li class="comments">\s*<a href="#comments">\s*<span class="count">(?P<commentcount>[^<]+)</span> Comments\s*</a>\s*</li>', html)
-        return int(self._fix_count(m.group('commentcount'))) if m is not None else 0
-
-    def _fix_count(self, count):
-        return int(str(count).replace(',', '')) if count is not None else None
-
-    def _extract_authors(self, html):
-        m = re.search(r'(?s)<li class="author">(.*?)</li>', html)
-        if m is None:
-            return None
-        return re.findall(r'<a href="/Niners/[^"]+">([^<]+)</a>', m.group(1))
-
-    def _extract_session_code(self, html):
-        m = re.search(r'<li class="code">\s*(?P<code>.+?)\s*</li>', html)
-        return m.group('code') if m is not None else None
-
-    def _extract_session_day(self, html):
-        m = re.search(r'<li class="day">\s*<a href="/Events/[^"]+">(?P<day>[^<]+)</a>\s*</li>', html)
-        return m.group('day').strip() if m is not None else None
-
-    def _extract_session_room(self, html):
-        m = re.search(r'<li class="room">\s*(?P<room>.+?)\s*</li>', html)
-        return m.group('room') if m is not None else None
-
-    def _extract_session_speakers(self, html):
-        return re.findall(r'<a href="/Events/Speakers/[^"]+">([^<]+)</a>', html)
-
-    def _extract_content(self, html, content_path):
-        # Look for downloadable content
-        formats = self._formats_from_html(html)
-        slides = self._extract_slides(html)
-        zip_ = self._extract_zip(html)
-
-        # Nothing to download
-        if len(formats) == 0 and slides is None and zip_ is None:
-            self._downloader.report_warning('None of recording, slides or zip are available for %s' % content_path)
-            return
-
-        # Extract meta
-        title = self._extract_title(html)
-        description = self._extract_description(html)
-        thumbnail = self._og_search_thumbnail(html)
-        duration = self._extract_duration(html)
-        avg_rating = self._extract_avg_rating(html)
-        rating_count = self._extract_rating_count(html)
-        view_count = self._extract_view_count(html)
-        comment_count = self._extract_comment_count(html)
-
-        common = {
-            '_type': 'video',
-            'id': content_path,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'avg_rating': avg_rating,
-            'rating_count': rating_count,
-            'view_count': view_count,
-            'comment_count': comment_count,
-        }
-
-        result = []
-
-        if slides is not None:
-            d = common.copy()
-            d.update({'title': title + '-Slides', 'url': slides})
-            result.append(d)
-
-        if zip_ is not None:
-            d = common.copy()
-            d.update({'title': title + '-Zip', 'url': zip_})
-            result.append(d)
-
-        if len(formats) > 0:
-            d = common.copy()
-            d.update({'title': title, 'formats': formats})
-            result.append(d)
-
-        return result
-
-    def _extract_entry_item(self, html, content_path):
-        contents = self._extract_content(html, content_path)
-        if contents is None:
-            return contents
-
-        if len(contents) > 1:
-            raise ExtractorError('Got more than one entry')
-        result = contents[0]
-        result['authors'] = self._extract_authors(html)
-
-        return result
-
-    def _extract_session(self, html, content_path):
-        contents = self._extract_content(html, content_path)
-        if contents is None:
-            return contents
-
-        session_meta = {
-            'session_code': self._extract_session_code(html),
-            'session_day': self._extract_session_day(html),
-            'session_room': self._extract_session_room(html),
-            'session_speakers': self._extract_session_speakers(html),
-        }
-
-        for content in contents:
-            content.update(session_meta)
-
-        return self.playlist_result(contents)
-
     def _extract_list(self, video_id, rss_url=None):
         if not rss_url:
             rss_url = self._RSS_URL % video_id
@@ -274,9 +91,7 @@ class Channel9IE(InfoExtractor):
         return self.playlist_result(entries, video_id, title_text)
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        content_path = mobj.group('contentpath')
-        rss = mobj.group('rss')
+        content_path, rss = re.match(self._VALID_URL, url).groups()
 
         if rss:
             return self._extract_list(content_path, url)
@@ -284,17 +99,158 @@ class Channel9IE(InfoExtractor):
         webpage = self._download_webpage(
             url, content_path, 'Downloading web page')
 
-        page_type = self._search_regex(
-            r'<meta[^>]+name=(["\'])WT\.entryid\1[^>]+content=(["\'])(?P<pagetype>[^:]+).+?\2',
-            webpage, 'page type', default=None, group='pagetype')
-        if page_type:
-            if page_type == 'Entry':      # Any 'item'-like page, may contain downloadable content
-                return self._extract_entry_item(webpage, content_path)
-            elif page_type == 'Session':  # Event session page, may contain downloadable content
-                return self._extract_session(webpage, content_path)
-            elif page_type == 'Event':
-                return self._extract_list(content_path)
+        episode_data = self._search_regex(
+            r"data-episode='([^']+)'", webpage, 'episode data', default=None)
+        if episode_data:
+            episode_data = self._parse_json(unescapeHTML(
+                episode_data), content_path)
+            content_id = episode_data['contentId']
+            is_session = '/Sessions(' in episode_data['api']
+            content_url = 'https://channel9.msdn.com/odata' + episode_data['api']
+            if is_session:
+                content_url += '?$expand=Speakers'
             else:
-                raise ExtractorError('Unexpected WT.entryid %s' % page_type, expected=True)
-        else:  # Assuming list
+                content_url += '?$expand=Authors'
+            content_data = self._download_json(content_url, content_id)
+            title = content_data['Title']
+
+            QUALITIES = (
+                'mp3',
+                'wmv', 'mp4',
+                'wmv-low', 'mp4-low',
+                'wmv-mid', 'mp4-mid',
+                'wmv-high', 'mp4-high',
+            )
+
+            quality_key = qualities(QUALITIES)
+
+            def quality(quality_id, format_url):
+                return (len(QUALITIES) if '_Source.' in format_url
+                        else quality_key(quality_id))
+
+            formats = []
+            urls = set()
+
+            SITE_QUALITIES = {
+                'MP3': 'mp3',
+                'MP4': 'mp4',
+                'Low Quality WMV': 'wmv-low',
+                'Low Quality MP4': 'mp4-low',
+                'Mid Quality WMV': 'wmv-mid',
+                'Mid Quality MP4': 'mp4-mid',
+                'High Quality WMV': 'wmv-high',
+                'High Quality MP4': 'mp4-high',
+            }
+
+            formats_select = self._search_regex(
+                r'(?s)<select[^>]+name=["\']format[^>]+>(.+?)</select', webpage,
+                'formats select', default=None)
+            if formats_select:
+                for mobj in re.finditer(
+                        r'<option\b[^>]+\bvalue=(["\'])(?P<url>(?:(?!\1).)+)\1[^>]*>\s*(?P<format>[^<]+?)\s*<',
+                        formats_select):
+                    format_url = mobj.group('url')
+                    if format_url in urls:
+                        continue
+                    urls.add(format_url)
+                    format_id = mobj.group('format')
+                    quality_id = SITE_QUALITIES.get(format_id, format_id)
+                    formats.append({
+                        'url': format_url,
+                        'format_id': quality_id,
+                        'quality': quality(quality_id, format_url),
+                        'vcodec': 'none' if quality_id == 'mp3' else None,
+                    })
+
+            API_QUALITIES = {
+                'VideoMP4Low': 'mp4-low',
+                'VideoWMV': 'wmv-mid',
+                'VideoMP4Medium': 'mp4-mid',
+                'VideoMP4High': 'mp4-high',
+                'VideoWMVHQ': 'wmv-hq',
+            }
+
+            for format_id, q in API_QUALITIES.items():
+                q_url = content_data.get(format_id)
+                if not q_url or q_url in urls:
+                    continue
+                urls.add(q_url)
+                formats.append({
+                    'url': q_url,
+                    'format_id': q,
+                    'quality': quality(q, q_url),
+                })
+
+            self._sort_formats(formats)
+
+            slides = content_data.get('Slides')
+            zip_file = content_data.get('ZipFile')
+
+            if not formats and not slides and not zip_file:
+                raise ExtractorError(
+                    'None of recording, slides or zip are available for %s' % content_path)
+
+            subtitles = {}
+            for caption in content_data.get('Captions', []):
+                caption_url = caption.get('Url')
+                if not caption_url:
+                    continue
+                subtitles.setdefault(caption.get('Language', 'en'), []).append({
+                    'url': caption_url,
+                    'ext': 'vtt',
+                })
+
+            common = {
+                'id': content_id,
+                'title': title,
+                'description': clean_html(content_data.get('Description') or content_data.get('Body')),
+                'thumbnail': content_data.get('Thumbnail') or content_data.get('VideoPlayerPreviewImage'),
+                'duration': int_or_none(content_data.get('MediaLengthInSeconds')),
+                'timestamp': parse_iso8601(content_data.get('PublishedDate')),
+                'avg_rating': int_or_none(content_data.get('Rating')),
+                'rating_count': int_or_none(content_data.get('RatingCount')),
+                'view_count': int_or_none(content_data.get('Views')),
+                'comment_count': int_or_none(content_data.get('CommentCount')),
+                'subtitles': subtitles,
+            }
+            if is_session:
+                speakers = []
+                for s in content_data.get('Speakers', []):
+                    speaker_name = s.get('FullName')
+                    if not speaker_name:
+                        continue
+                    speakers.append(speaker_name)
+
+                common.update({
+                    'session_code': content_data.get('Code'),
+                    'session_room': content_data.get('Room'),
+                    'session_speakers': speakers,
+                })
+            else:
+                authors = []
+                for a in content_data.get('Authors', []):
+                    author_name = a.get('DisplayName')
+                    if not author_name:
+                        continue
+                    authors.append(author_name)
+                common['authors'] = authors
+
+            contents = []
+
+            if slides:
+                d = common.copy()
+                d.update({'title': title + '-Slides', 'url': slides})
+                contents.append(d)
+
+            if zip_file:
+                d = common.copy()
+                d.update({'title': title + '-Zip', 'url': zip_file})
+                contents.append(d)
+
+            if formats:
+                d = common.copy()
+                d.update({'title': title, 'formats': formats})
+                contents.append(d)
+            return self.playlist_result(contents)
+        else:
             return self._extract_list(content_path)

youtube_dl/extractor/chaturbate.py

@@ -33,10 +33,17 @@ class ChaturbateIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        m3u8_formats = [(m.group('id').lower(), m.group('url')) for m in re.finditer(
-            r'hlsSource(?P<id>.+?)\s*=\s*(?P<q>["\'])(?P<url>http.+?)(?P=q)', webpage)]
+        m3u8_urls = []
 
-        if not m3u8_formats:
+        for m in re.finditer(
+                r'(["\'])(?P<url>http.+?\.m3u8.*?)\1', webpage):
+            m3u8_fast_url, m3u8_no_fast_url = m.group('url'), m.group(
+                'url').replace('_fast', '')
+            for m3u8_url in (m3u8_fast_url, m3u8_no_fast_url):
+                if m3u8_url not in m3u8_urls:
+                    m3u8_urls.append(m3u8_url)
+
+        if not m3u8_urls:
             error = self._search_regex(
                 [r'<span[^>]+class=(["\'])desc_span\1[^>]*>(?P<error>[^<]+)</span>',
                  r'<div[^>]+id=(["\'])defchat\1[^>]*>\s*<p><strong>(?P<error>[^<]+)<'],
@@ -50,7 +57,8 @@ class ChaturbateIE(InfoExtractor):
             raise ExtractorError('Unable to find stream URL')
 
         formats = []
-        for m3u8_id, m3u8_url in m3u8_formats:
+        for m3u8_url in m3u8_urls:
+            m3u8_id = 'fast' if '_fast' in m3u8_url else 'slow'
             formats.extend(self._extract_m3u8_formats(
                 m3u8_url, video_id, ext='mp4',
                 # ffmpeg skips segments for fast m3u8

youtube_dl/extractor/cloudy.py

@@ -1,97 +1,56 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..compat import (
-    compat_parse_qs,
-    compat_HTTPError,
-)
 from ..utils import (
-    ExtractorError,
-    HEADRequest,
-    remove_end,
+    str_to_int,
+    unified_strdate,
 )
 
 
 class CloudyIE(InfoExtractor):
     _IE_DESC = 'cloudy.ec'
-    _VALID_URL = r'''(?x)
-        https?://(?:www\.)?cloudy\.ec/
-        (?:v/|embed\.php\?id=)
-        (?P<id>[A-Za-z0-9]+)
-        '''
-    _EMBED_URL = 'http://www.cloudy.ec/embed.php?id=%s'
-    _API_URL = 'http://www.cloudy.ec/api/player.api.php'
-    _MAX_TRIES = 2
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?cloudy\.ec/(?:v/|embed\.php\?.*?\bid=)(?P<id>[A-Za-z0-9]+)'
+    _TESTS = [{
         'url': 'https://www.cloudy.ec/v/af511e2527aac',
-        'md5': '5cb253ace826a42f35b4740539bedf07',
+        'md5': '29832b05028ead1b58be86bf319397ca',
         'info_dict': {
             'id': 'af511e2527aac',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Funny Cats and Animals Compilation june 2013',
+            'upload_date': '20130913',
+            'view_count': int,
         }
-    }
-
-    def _extract_video(self, video_id, file_key, error_url=None, try_num=0):
-
-        if try_num > self._MAX_TRIES - 1:
-            raise ExtractorError('Unable to extract video URL', expected=True)
-
-        form = {
-            'file': video_id,
-            'key': file_key,
-        }
-
-        if error_url:
-            form.update({
-                'numOfErrors': try_num,
-                'errorCode': '404',
-                'errorUrl': error_url,
-            })
-
-        player_data = self._download_webpage(
-            self._API_URL, video_id, 'Downloading player data', query=form)
-        data = compat_parse_qs(player_data)
-
-        try_num += 1
-
-        if 'error' in data:
-            raise ExtractorError(
-                '%s error: %s' % (self.IE_NAME, ' '.join(data['error_msg'])),
-                expected=True)
-
-        title = data.get('title', [None])[0]
-        if title:
-            title = remove_end(title, '&asdasdas').strip()
-
-        video_url = data.get('url', [None])[0]
-
-        if video_url:
-            try:
-                self._request_webpage(HEADRequest(video_url), video_id, 'Checking video URL')
-            except ExtractorError as e:
-                if isinstance(e.cause, compat_HTTPError) and e.cause.code in [404, 410]:
-                    self.report_warning('Invalid video URL, requesting another', video_id)
-                    return self._extract_video(video_id, file_key, video_url, try_num)
-
-        return {
-            'id': video_id,
-            'url': video_url,
-            'title': title,
-        }
+    }, {
+        'url': 'http://www.cloudy.ec/embed.php?autoplay=1&id=af511e2527aac',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
-        url = self._EMBED_URL % video_id
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(
+            'http://www.cloudy.ec/embed.php?id=%s' % video_id, video_id)
 
-        file_key = self._search_regex(
-            [r'key\s*:\s*"([^"]+)"', r'filekey\s*=\s*"([^"]+)"'],
-            webpage, 'file_key')
+        info = self._parse_html5_media_entries(url, webpage, video_id)[0]
 
-        return self._extract_video(video_id, file_key)
+        webpage = self._download_webpage(
+            'https://www.cloudy.ec/v/%s' % video_id, video_id, fatal=False)
+
+        if webpage:
+            info.update({
+                'title': self._search_regex(
+                    r'<h\d[^>]*>([^<]+)<', webpage, 'title'),
+                'upload_date': unified_strdate(self._search_regex(
+                    r'>Published at (\d{4}-\d{1,2}-\d{1,2})', webpage,
+                    'upload date', fatal=False)),
+                'view_count': str_to_int(self._search_regex(
+                    r'([\d,.]+) views<', webpage, 'view count', fatal=False)),
+            })
+
+        if not info.get('title'):
+            info['title'] = video_id
+
+        info['id'] = video_id
+
+        return info

youtube_dl/extractor/common.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import base64
@@ -6,6 +7,7 @@ import hashlib
 import json
 import netrc
 import os
+import random
 import re
 import socket
 import sys
@@ -35,31 +37,35 @@ from ..utils import (
     clean_html,
     compiled_regex_type,
     determine_ext,
+    determine_protocol,
     error_to_compat_str,
     ExtractorError,
+    extract_attributes,
     fix_xml_ampersands,
     float_or_none,
+    GeoRestrictedError,
+    GeoUtils,
     int_or_none,
+    js_to_json,
+    mimetype2ext,
+    orderedSet,
+    parse_codecs,
+    parse_duration,
     parse_iso8601,
+    parse_m3u8_attributes,
     RegexNotFoundError,
-    sanitize_filename,
     sanitized_Request,
+    sanitize_filename,
     unescapeHTML,
     unified_strdate,
     unified_timestamp,
+    update_Request,
+    update_url_query,
+    urljoin,
     url_basename,
     xpath_element,
     xpath_text,
     xpath_with_ns,
-    determine_protocol,
-    parse_duration,
-    mimetype2ext,
-    update_Request,
-    update_url_query,
-    parse_m3u8_attributes,
-    extract_attributes,
-    parse_codecs,
-    urljoin,
 )
 
 
@@ -319,17 +325,34 @@ class InfoExtractor(object):
     _real_extract() methods and define a _VALID_URL regexp.
     Probably, they should also be added to the list of extractors.
 
+    _GEO_BYPASS attribute may be set to False in order to disable
+    geo restriction bypass mechanisms for a particular extractor.
+    Though it won't disable explicit geo restriction bypass based on
+    country code provided with geo_bypass_country. (experimental)
+
+    _GEO_COUNTRIES attribute may contain a list of presumably geo unrestricted
+    countries for this extractor. One of these countries will be used by
+    geo restriction bypass mechanism right away in order to bypass
+    geo restriction, of course, if the mechanism is not disabled. (experimental)
+
+    NB: both these geo attributes are experimental and may change in future
+    or be completely removed.
+
     Finally, the _WORKING attribute should be set to False for broken IEs
     in order to warn the users and skip the tests.
     """
 
     _ready = False
     _downloader = None
+    _x_forwarded_for_ip = None
+    _GEO_BYPASS = True
+    _GEO_COUNTRIES = None
     _WORKING = True
 
     def __init__(self, downloader=None):
         """Constructor. Receives an optional downloader."""
         self._ready = False
+        self._x_forwarded_for_ip = None
         self.set_downloader(downloader)
 
     @classmethod
@@ -358,15 +381,59 @@ class InfoExtractor(object):
 
     def initialize(self):
         """Initializes an instance (authentication, etc)."""
+        self._initialize_geo_bypass(self._GEO_COUNTRIES)
         if not self._ready:
             self._real_initialize()
             self._ready = True
 
+    def _initialize_geo_bypass(self, countries):
+        """
+        Initialize geo restriction bypass mechanism.
+
+        This method is used to initialize geo bypass mechanism based on faking
+        X-Forwarded-For HTTP header. A random country from provided country list
+        is selected and a random IP belonging to this country is generated. This
+        IP will be passed as X-Forwarded-For HTTP header in all subsequent
+        HTTP requests.
+
+        This method will be used for initial geo bypass mechanism initialization
+        during the instance initialization with _GEO_COUNTRIES.
+
+        You may also manually call it from extractor's code if geo countries
+        information is not available beforehand (e.g. obtained during
+        extraction) or due to some another reason.
+        """
+        if not self._x_forwarded_for_ip:
+            country_code = self._downloader.params.get('geo_bypass_country', None)
+            # If there is no explicit country for geo bypass specified and
+            # the extractor is known to be geo restricted let's fake IP
+            # as X-Forwarded-For right away.
+            if (not country_code and
+                    self._GEO_BYPASS and
+                    self._downloader.params.get('geo_bypass', True) and
+                    countries):
+                country_code = random.choice(countries)
+            if country_code:
+                self._x_forwarded_for_ip = GeoUtils.random_ipv4(country_code)
+                if self._downloader.params.get('verbose', False):
+                    self._downloader.to_stdout(
+                        '[debug] Using fake IP %s (%s) as X-Forwarded-For.'
+                        % (self._x_forwarded_for_ip, country_code.upper()))
+
     def extract(self, url):
         """Extracts URL information and returns it in list of dicts."""
         try:
-            self.initialize()
-            return self._real_extract(url)
+            for _ in range(2):
+                try:
+                    self.initialize()
+                    ie_result = self._real_extract(url)
+                    if self._x_forwarded_for_ip:
+                        ie_result['__x_forwarded_for_ip'] = self._x_forwarded_for_ip
+                    return ie_result
+                except GeoRestrictedError as e:
+                    if self.__maybe_fake_ip_and_retry(e.countries):
+                        continue
+                    raise
         except ExtractorError:
             raise
         except compat_http_client.IncompleteRead as e:
@@ -374,6 +441,21 @@ class InfoExtractor(object):
         except (KeyError, StopIteration) as e:
             raise ExtractorError('An extractor error has occurred.', cause=e)
 
+    def __maybe_fake_ip_and_retry(self, countries):
+        if (not self._downloader.params.get('geo_bypass_country', None) and
+                self._GEO_BYPASS and
+                self._downloader.params.get('geo_bypass', True) and
+                not self._x_forwarded_for_ip and
+                countries):
+            country_code = random.choice(countries)
+            self._x_forwarded_for_ip = GeoUtils.random_ipv4(country_code)
+            if self._x_forwarded_for_ip:
+                self.report_warning(
+                    'Video is geo restricted. Retrying extraction with fake IP %s (%s) as X-Forwarded-For.'
+                    % (self._x_forwarded_for_ip, country_code.upper()))
+                return True
+        return False
+
     def set_downloader(self, downloader):
         """Sets the downloader for this IE."""
         self._downloader = downloader
@@ -433,6 +515,15 @@ class InfoExtractor(object):
         if isinstance(url_or_request, (compat_str, str)):
             url_or_request = url_or_request.partition('#')[0]
 
+        # Some sites check X-Forwarded-For HTTP header in order to figure out
+        # the origin of the client behind proxy. This allows bypassing geo
+        # restriction by faking this header's value to IP that belongs to some
+        # geo unrestricted country. We will do so once we encounter any
+        # geo restriction error.
+        if self._x_forwarded_for_ip:
+            if 'X-Forwarded-For' not in headers:
+                headers['X-Forwarded-For'] = self._x_forwarded_for_ip
+
         urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal, data=data, headers=headers, query=query)
         if urlh is False:
             assert not fatal
@@ -457,6 +548,34 @@ class InfoExtractor(object):
 
         return encoding
 
+    def __check_blocked(self, content):
+        first_block = content[:512]
+        if ('<title>Access to this site is blocked</title>' in content and
+                'Websense' in first_block):
+            msg = 'Access to this webpage has been blocked by Websense filtering software in your network.'
+            blocked_iframe = self._html_search_regex(
+                r'<iframe src="([^"]+)"', content,
+                'Websense information URL', default=None)
+            if blocked_iframe:
+                msg += ' Visit %s for more details' % blocked_iframe
+            raise ExtractorError(msg, expected=True)
+        if '<title>The URL you requested has been blocked</title>' in first_block:
+            msg = (
+                'Access to this webpage has been blocked by Indian censorship. '
+                'Use a VPN or proxy server (with --proxy) to route around it.')
+            block_msg = self._html_search_regex(
+                r'</h1><p>(.*?)</p>',
+                content, 'block message', default=None)
+            if block_msg:
+                msg += ' (Message: "%s")' % block_msg.replace('\n', ' ')
+            raise ExtractorError(msg, expected=True)
+        if ('<title>TTK :: Доступ к ресурсу ограничен</title>' in content and
+                'blocklist.rkn.gov.ru' in content):
+            raise ExtractorError(
+                'Access to this webpage has been blocked by decision of the Russian government. '
+                'Visit http://blocklist.rkn.gov.ru/ for a block reason.',
+                expected=True)
+
     def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None):
         content_type = urlh.headers.get('Content-Type', '')
         webpage_bytes = urlh.read()
@@ -498,25 +617,7 @@ class InfoExtractor(object):
         except LookupError:
             content = webpage_bytes.decode('utf-8', 'replace')
 
-        if ('<title>Access to this site is blocked</title>' in content and
-                'Websense' in content[:512]):
-            msg = 'Access to this webpage has been blocked by Websense filtering software in your network.'
-            blocked_iframe = self._html_search_regex(
-                r'<iframe src="([^"]+)"', content,
-                'Websense information URL', default=None)
-            if blocked_iframe:
-                msg += ' Visit %s for more details' % blocked_iframe
-            raise ExtractorError(msg, expected=True)
-        if '<title>The URL you requested has been blocked</title>' in content[:512]:
-            msg = (
-                'Access to this webpage has been blocked by Indian censorship. '
-                'Use a VPN or proxy server (with --proxy) to route around it.')
-            block_msg = self._html_search_regex(
-                r'</h1><p>(.*?)</p>',
-                content, 'block message', default=None)
-            if block_msg:
-                msg += ' (Message: "%s")' % block_msg.replace('\n', ' ')
-            raise ExtractorError(msg, expected=True)
+        self.__check_blocked(content)
 
         return content
 
@@ -608,10 +709,8 @@ class InfoExtractor(object):
             expected=True)
 
     @staticmethod
-    def raise_geo_restricted(msg='This video is not available from your location due to geo restriction'):
-        raise ExtractorError(
-            '%s. You might want to use --proxy to workaround.' % msg,
-            expected=True)
+    def raise_geo_restricted(msg='This video is not available from your location due to geo restriction', countries=None):
+        raise GeoRestrictedError(msg, countries=countries)
 
     # Methods for following #608
     @staticmethod
@@ -627,6 +726,13 @@ class InfoExtractor(object):
             video_info['title'] = video_title
         return video_info
 
+    def playlist_from_matches(self, matches, video_id, video_title, getter=None, ie=None):
+        urlrs = orderedSet(
+            self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
+            for m in matches)
+        return self.playlist_result(
+            urlrs, playlist_id=video_id, playlist_title=video_title)
+
     @staticmethod
     def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):
         """Returns a playlist"""
@@ -870,6 +976,22 @@ class InfoExtractor(object):
             return info
         if isinstance(json_ld, dict):
             json_ld = [json_ld]
+
+        def extract_video_object(e):
+            assert e['@type'] == 'VideoObject'
+            info.update({
+                'url': e.get('contentUrl'),
+                'title': unescapeHTML(e.get('name')),
+                'description': unescapeHTML(e.get('description')),
+                'thumbnail': e.get('thumbnailUrl') or e.get('thumbnailURL'),
+                'duration': parse_duration(e.get('duration')),
+                'timestamp': unified_timestamp(e.get('uploadDate')),
+                'filesize': float_or_none(e.get('contentSize')),
+                'tbr': int_or_none(e.get('bitrate')),
+                'width': int_or_none(e.get('width')),
+                'height': int_or_none(e.get('height')),
+            })
+
         for e in json_ld:
             if e.get('@context') == 'http://schema.org':
                 item_type = e.get('@type')
@@ -894,18 +1016,11 @@ class InfoExtractor(object):
                         'description': unescapeHTML(e.get('articleBody')),
                     })
                 elif item_type == 'VideoObject':
-                    info.update({
-                        'url': e.get('contentUrl'),
-                        'title': unescapeHTML(e.get('name')),
-                        'description': unescapeHTML(e.get('description')),
-                        'thumbnail': e.get('thumbnailUrl') or e.get('thumbnailURL'),
-                        'duration': parse_duration(e.get('duration')),
-                        'timestamp': unified_timestamp(e.get('uploadDate')),
-                        'filesize': float_or_none(e.get('contentSize')),
-                        'tbr': int_or_none(e.get('bitrate')),
-                        'width': int_or_none(e.get('width')),
-                        'height': int_or_none(e.get('height')),
-                    })
+                    extract_video_object(e)
+                elif item_type == 'WebPage':
+                    video = e.get('video')
+                    if isinstance(video, dict) and video.get('@type') == 'VideoObject':
+                        extract_video_object(video)
                 break
         return dict((k, v) for k, v in info.items() if v is not None)
 
@@ -1197,17 +1312,28 @@ class InfoExtractor(object):
                               entry_protocol='m3u8', preference=None,
                               m3u8_id=None, note=None, errnote=None,
                               fatal=True, live=False):
-
         res = self._download_webpage_handle(
             m3u8_url, video_id,
             note=note or 'Downloading m3u8 information',
             errnote=errnote or 'Failed to download m3u8 information',
             fatal=fatal)
+
         if res is False:
             return []
+
         m3u8_doc, urlh = res
         m3u8_url = urlh.geturl()
 
+        return self._parse_m3u8_formats(
+            m3u8_doc, m3u8_url, ext=ext, entry_protocol=entry_protocol,
+            preference=preference, m3u8_id=m3u8_id, live=live)
+
+    def _parse_m3u8_formats(self, m3u8_doc, m3u8_url, ext=None,
+                            entry_protocol='m3u8', preference=None,
+                            m3u8_id=None, live=False):
+        if '#EXT-X-FAXS-CM:' in m3u8_doc:  # Adobe Flash Access
+            return []
+
         formats = [self._m3u8_meta_format(m3u8_url, ext, preference, m3u8_id)]
 
         format_url = lambda u: (
@@ -1215,19 +1341,21 @@ class InfoExtractor(object):
             if re.match(r'^https?://', u)
             else compat_urlparse.urljoin(m3u8_url, u))
 
-        # We should try extracting formats only from master playlists [1], i.e.
-        # playlists that describe available qualities. On the other hand media
-        # playlists [2] should be returned as is since they contain just the media
-        # without qualities renditions.
+        # References:
+        # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-21
+        # 2. https://github.com/rg3/youtube-dl/issues/12211
+
+        # We should try extracting formats only from master playlists [1, 4.3.4],
+        # i.e. playlists that describe available qualities. On the other hand
+        # media playlists [1, 4.3.3] should be returned as is since they contain
+        # just the media without qualities renditions.
         # Fortunately, master playlist can be easily distinguished from media
-        # playlist based on particular tags availability. As of [1, 2] master
-        # playlist tags MUST NOT appear in a media playist and vice versa.
-        # As of [3] #EXT-X-TARGETDURATION tag is REQUIRED for every media playlist
-        # and MUST NOT appear in master playlist thus we can clearly detect media
-        # playlist with this criterion.
-        # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.4
-        # 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3
-        # 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.1
+        # playlist based on particular tags availability. As of [1, 4.3.3, 4.3.4]
+        # master playlist tags MUST NOT appear in a media playist and vice versa.
+        # As of [1, 4.3.3.1] #EXT-X-TARGETDURATION tag is REQUIRED for every
+        # media playlist and MUST NOT appear in master playlist thus we can
+        # clearly detect media playlist with this criterion.
+
         if '#EXT-X-TARGETDURATION' in m3u8_doc:  # media playlist, return as is
             return [{
                 'url': m3u8_url,
@@ -1236,52 +1364,71 @@ class InfoExtractor(object):
                 'protocol': entry_protocol,
                 'preference': preference,
             }]
-        audio_in_video_stream = {}
-        last_info = {}
-        last_media = {}
+
+        groups = {}
+        last_stream_inf = {}
+
+        def extract_media(x_media_line):
+            media = parse_m3u8_attributes(x_media_line)
+            # As per [1, 4.3.4.1] TYPE, GROUP-ID and NAME are REQUIRED
+            media_type, group_id, name = media.get('TYPE'), media.get('GROUP-ID'), media.get('NAME')
+            if not (media_type and group_id and name):
+                return
+            groups.setdefault(group_id, []).append(media)
+            if media_type not in ('VIDEO', 'AUDIO'):
+                return
+            media_url = media.get('URI')
+            if media_url:
+                format_id = []
+                for v in (group_id, name):
+                    if v:
+                        format_id.append(v)
+                f = {
+                    'format_id': '-'.join(format_id),
+                    'url': format_url(media_url),
+                    'language': media.get('LANGUAGE'),
+                    'ext': ext,
+                    'protocol': entry_protocol,
+                    'preference': preference,
+                }
+                if media_type == 'AUDIO':
+                    f['vcodec'] = 'none'
+                formats.append(f)
+
+        def build_stream_name():
+            # Despite specification does not mention NAME attribute for
+            # EXT-X-STREAM-INF tag it still sometimes may be present (see [1]
+            # or vidio test in TestInfoExtractor.test_parse_m3u8_formats)
+            # 1. http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015
+            stream_name = last_stream_inf.get('NAME')
+            if stream_name:
+                return stream_name
+            # If there is no NAME in EXT-X-STREAM-INF it will be obtained
+            # from corresponding rendition group
+            stream_group_id = last_stream_inf.get('VIDEO')
+            if not stream_group_id:
+                return
+            stream_group = groups.get(stream_group_id)
+            if not stream_group:
+                return stream_group_id
+            rendition = stream_group[0]
+            return rendition.get('NAME') or stream_group_id
+
         for line in m3u8_doc.splitlines():
             if line.startswith('#EXT-X-STREAM-INF:'):
-                last_info = parse_m3u8_attributes(line)
+                last_stream_inf = parse_m3u8_attributes(line)
             elif line.startswith('#EXT-X-MEDIA:'):
-                media = parse_m3u8_attributes(line)
-                media_type = media.get('TYPE')
-                if media_type in ('VIDEO', 'AUDIO'):
-                    group_id = media.get('GROUP-ID')
-                    media_url = media.get('URI')
-                    if media_url:
-                        format_id = []
-                        for v in (group_id, media.get('NAME')):
-                            if v:
-                                format_id.append(v)
-                        f = {
-                            'format_id': '-'.join(format_id),
-                            'url': format_url(media_url),
-                            'language': media.get('LANGUAGE'),
-                            'ext': ext,
-                            'protocol': entry_protocol,
-                            'preference': preference,
-                        }
-                        if media_type == 'AUDIO':
-                            f['vcodec'] = 'none'
-                            if group_id and not audio_in_video_stream.get(group_id):
-                                audio_in_video_stream[group_id] = False
-                        formats.append(f)
-                    else:
-                        # When there is no URI in EXT-X-MEDIA let this tag's
-                        # data be used by regular URI lines below
-                        last_media = media
-                        if media_type == 'AUDIO' and group_id:
-                            audio_in_video_stream[group_id] = True
+                extract_media(line)
             elif line.startswith('#') or not line.strip():
                 continue
             else:
-                tbr = int_or_none(last_info.get('AVERAGE-BANDWIDTH') or last_info.get('BANDWIDTH'), scale=1000)
+                tbr = float_or_none(
+                    last_stream_inf.get('AVERAGE-BANDWIDTH') or
+                    last_stream_inf.get('BANDWIDTH'), scale=1000)
                 format_id = []
                 if m3u8_id:
                     format_id.append(m3u8_id)
-                # Despite specification does not mention NAME attribute for
-                # EXT-X-STREAM-INF it still sometimes may be present
-                stream_name = last_info.get('NAME') or last_media.get('NAME')
+                stream_name = build_stream_name()
                 # Bandwidth of live streams may differ over time thus making
                 # format_id unpredictable. So it's better to keep provided
                 # format_id intact.
@@ -1294,11 +1441,11 @@ class InfoExtractor(object):
                     'manifest_url': manifest_url,
                     'tbr': tbr,
                     'ext': ext,
-                    'fps': float_or_none(last_info.get('FRAME-RATE')),
+                    'fps': float_or_none(last_stream_inf.get('FRAME-RATE')),
                     'protocol': entry_protocol,
                     'preference': preference,
                 }
-                resolution = last_info.get('RESOLUTION')
+                resolution = last_stream_inf.get('RESOLUTION')
                 if resolution:
                     mobj = re.search(r'(?P<width>\d+)[xX](?P<height>\d+)', resolution)
                     if mobj:
@@ -1314,13 +1461,26 @@ class InfoExtractor(object):
                         'vbr': vbr,
                         'abr': abr,
                     })
-                f.update(parse_codecs(last_info.get('CODECS')))
-                if audio_in_video_stream.get(last_info.get('AUDIO')) is False:
-                    # TODO: update acodec for for audio only formats with the same GROUP-ID
-                    f['acodec'] = 'none'
+                codecs = parse_codecs(last_stream_inf.get('CODECS'))
+                f.update(codecs)
+                audio_group_id = last_stream_inf.get('AUDIO')
+                # As per [1, 4.3.4.1.1] any EXT-X-STREAM-INF tag which
+                # references a rendition group MUST have a CODECS attribute.
+                # However, this is not always respected, for example, [2]
+                # contains EXT-X-STREAM-INF tag which references AUDIO
+                # rendition group but does not have CODECS and despite
+                # referencing audio group an audio group, it represents
+                # a complete (with audio and video) format. So, for such cases
+                # we will ignore references to rendition groups and treat them
+                # as complete formats.
+                if audio_group_id and codecs and f.get('vcodec') != 'none':
+                    audio_group = groups.get(audio_group_id)
+                    if audio_group and audio_group[0].get('URI'):
+                        # TODO: update acodec for audio only formats with
+                        # the same GROUP-ID
+                        f['acodec'] = 'none'
                 formats.append(f)
-                last_info = {}
-                last_media = {}
+                last_stream_inf = {}
         return formats
 
     @staticmethod
@@ -1670,7 +1830,7 @@ class InfoExtractor(object):
                     if content_type == 'text':
                         # TODO implement WebVTT downloading
                         pass
-                    elif content_type == 'video' or content_type == 'audio':
+                    elif content_type in ('video', 'audio'):
                         base_url = ''
                         for element in (representation, adaptation_set, period, mpd_doc):
                             base_url_e = element.find(_add_ns('BaseURL'))
@@ -1694,7 +1854,7 @@ class InfoExtractor(object):
                             'ext': mimetype2ext(mime_type),
                             'width': int_or_none(representation_attrib.get('width')),
                             'height': int_or_none(representation_attrib.get('height')),
-                            'tbr': int_or_none(bandwidth, 1000),
+                            'tbr': float_or_none(bandwidth, 1000),
                             'asr': int_or_none(representation_attrib.get('audioSamplingRate')),
                             'fps': int_or_none(representation_attrib.get('frameRate')),
                             'language': lang if lang not in ('mul', 'und', 'zxx', 'mis') else None,
@@ -1920,7 +2080,7 @@ class InfoExtractor(object):
                 })
         return formats
 
-    def _parse_html5_media_entries(self, base_url, webpage, video_id, m3u8_id=None, m3u8_entry_protocol='m3u8', mpd_id=None):
+    def _parse_html5_media_entries(self, base_url, webpage, video_id, m3u8_id=None, m3u8_entry_protocol='m3u8', mpd_id=None, preference=None):
         def absolute_url(video_url):
             return compat_urlparse.urljoin(base_url, video_url)
 
@@ -1942,7 +2102,8 @@ class InfoExtractor(object):
                 is_plain_url = False
                 formats = self._extract_m3u8_formats(
                     full_url, video_id, ext='mp4',
-                    entry_protocol=m3u8_entry_protocol, m3u8_id=m3u8_id)
+                    entry_protocol=m3u8_entry_protocol, m3u8_id=m3u8_id,
+                    preference=preference)
             elif ext == 'mpd':
                 is_plain_url = False
                 formats = self._extract_mpd_formats(
@@ -1959,7 +2120,12 @@ class InfoExtractor(object):
         media_tags = [(media_tag, media_type, '')
                       for media_tag, media_type
                       in re.findall(r'(?s)(<(video|audio)[^>]*/>)', webpage)]
-        media_tags.extend(re.findall(r'(?s)(<(?P<tag>video|audio)[^>]*>)(.*?)</(?P=tag)>', webpage))
+        media_tags.extend(re.findall(
+            # We only allow video|audio followed by a whitespace or '>'.
+            # Allowing more characters may end up in significant slow down (see
+            # https://github.com/rg3/youtube-dl/issues/11979, example URL:
+            # http://www.porntrex.com/maps/videositemap.xml).
+            r'(?s)(<(?P<tag>video|audio)(?:\s+[^>]*)?>)(.*?)</(?P=tag)>', webpage))
         for media_tag, media_type, media_content in media_tags:
             media_info = {
                 'formats': [],
@@ -2065,6 +2231,147 @@ class InfoExtractor(object):
                     })
         return formats
 
+    def _find_jwplayer_data(self, webpage, video_id=None, transform_source=js_to_json):
+        mobj = re.search(
+            r'(?s)jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)(?!</script>).*?\.setup\s*\((?P<options>[^)]+)\)',
+            webpage)
+        if mobj:
+            try:
+                jwplayer_data = self._parse_json(mobj.group('options'),
+                                                 video_id=video_id,
+                                                 transform_source=transform_source)
+            except ExtractorError:
+                pass
+            else:
+                if isinstance(jwplayer_data, dict):
+                    return jwplayer_data
+
+    def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
+        jwplayer_data = self._find_jwplayer_data(
+            webpage, video_id, transform_source=js_to_json)
+        return self._parse_jwplayer_data(
+            jwplayer_data, video_id, *args, **kwargs)
+
+    def _parse_jwplayer_data(self, jwplayer_data, video_id=None, require_title=True,
+                             m3u8_id=None, mpd_id=None, rtmp_params=None, base_url=None):
+        # JWPlayer backward compatibility: flattened playlists
+        # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/api/config.js#L81-L96
+        if 'playlist' not in jwplayer_data:
+            jwplayer_data = {'playlist': [jwplayer_data]}
+
+        entries = []
+
+        # JWPlayer backward compatibility: single playlist item
+        # https://github.com/jwplayer/jwplayer/blob/v7.7.0/src/js/playlist/playlist.js#L10
+        if not isinstance(jwplayer_data['playlist'], list):
+            jwplayer_data['playlist'] = [jwplayer_data['playlist']]
+
+        for video_data in jwplayer_data['playlist']:
+            # JWPlayer backward compatibility: flattened sources
+            # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35
+            if 'sources' not in video_data:
+                video_data['sources'] = [video_data]
+
+            this_video_id = video_id or video_data['mediaid']
+
+            formats = self._parse_jwplayer_formats(
+                video_data['sources'], video_id=this_video_id, m3u8_id=m3u8_id,
+                mpd_id=mpd_id, rtmp_params=rtmp_params, base_url=base_url)
+            self._sort_formats(formats)
+
+            subtitles = {}
+            tracks = video_data.get('tracks')
+            if tracks and isinstance(tracks, list):
+                for track in tracks:
+                    if track.get('kind') != 'captions':
+                        continue
+                    track_url = urljoin(base_url, track.get('file'))
+                    if not track_url:
+                        continue
+                    subtitles.setdefault(track.get('label') or 'en', []).append({
+                        'url': self._proto_relative_url(track_url)
+                    })
+
+            entries.append({
+                'id': this_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') or video_data.get('duration')),
+                'subtitles': subtitles,
+                'formats': formats,
+            })
+        if len(entries) == 1:
+            return entries[0]
+        else:
+            return self.playlist_result(entries)
+
+    def _parse_jwplayer_formats(self, jwplayer_sources_data, video_id=None,
+                                m3u8_id=None, mpd_id=None, rtmp_params=None, base_url=None):
+        urls = []
+        formats = []
+        for source in jwplayer_sources_data:
+            source_url = self._proto_relative_url(source.get('file'))
+            if not source_url:
+                continue
+            if base_url:
+                source_url = compat_urlparse.urljoin(base_url, source_url)
+            if source_url in urls:
+                continue
+            urls.append(source_url)
+            source_type = source.get('type') or ''
+            ext = mimetype2ext(source_type) or determine_ext(source_url)
+            if source_type == 'hls' or ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    source_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id=m3u8_id, fatal=False))
+            elif ext == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    source_url, video_id, mpd_id=mpd_id, fatal=False))
+            elif ext == 'smil':
+                formats.extend(self._extract_smil_formats(
+                    source_url, video_id, fatal=False))
+            # https://github.com/jwplayer/jwplayer/blob/master/src/js/providers/default.js#L67
+            elif source_type.startswith('audio') or ext in (
+                    'oga', 'aac', 'mp3', 'mpeg', 'vorbis'):
+                formats.append({
+                    'url': source_url,
+                    'vcodec': 'none',
+                    'ext': ext,
+                })
+            else:
+                height = int_or_none(source.get('height'))
+                if height is None:
+                    # Often no height is provided but there is a label in
+                    # format like "1080p", "720p SD", or 1080.
+                    height = int_or_none(self._search_regex(
+                        r'^(\d{3,4})[pP]?(?:\b|$)', compat_str(source.get('label') or ''),
+                        'height', default=None))
+                a_format = {
+                    'url': source_url,
+                    'width': int_or_none(source.get('width')),
+                    'height': height,
+                    'tbr': int_or_none(source.get('bitrate')),
+                    'ext': ext,
+                }
+                if source_url.startswith('rtmp'):
+                    a_format['ext'] = 'flv'
+                    # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as
+                    # of jwplayer.flash.swf
+                    rtmp_url_parts = re.split(
+                        r'((?:mp4|mp3|flv):)', source_url, 1)
+                    if len(rtmp_url_parts) == 3:
+                        rtmp_url, prefix, play_path = rtmp_url_parts
+                        a_format.update({
+                            'url': rtmp_url,
+                            'play_path': prefix + play_path,
+                        })
+                    if rtmp_params:
+                        a_format.update(rtmp_params)
+                formats.append(a_format)
+        return formats
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()

youtube_dl/extractor/commonmistakes.py

@@ -1,5 +1,7 @@
 from __future__ import unicode_literals
 
+import sys
+
 from .common import InfoExtractor
 from ..utils import ExtractorError
 
@@ -7,7 +9,7 @@ from ..utils import ExtractorError
 class CommonMistakesIE(InfoExtractor):
     IE_DESC = False  # Do not list
     _VALID_URL = r'''(?x)
-        (?:url|URL)
+        (?:url|URL)$
     '''
 
     _TESTS = [{
@@ -33,7 +35,9 @@ class UnicodeBOMIE(InfoExtractor):
         IE_DESC = False
         _VALID_URL = r'(?P<bom>\ufeff)(?P<id>.*)$'
 
-        _TESTS = [{
+        # Disable test for python 3.2 since BOM is broken in re in this version
+        # (see https://github.com/rg3/youtube-dl/issues/9751)
+        _TESTS = [] if (3, 0) < sys.version_info <= (3, 3) else [{
             'url': '\ufeffhttp://www.youtube.com/watch?v=BaW_jenozKc',
             'only_matching': True,
         }]

youtube_dl/extractor/condenast.py

@@ -9,13 +9,14 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
-    orderedSet,
-    remove_end,
-    extract_attributes,
-    mimetype2ext,
     determine_ext,
+    extract_attributes,
     int_or_none,
+    js_to_json,
+    mimetype2ext,
+    orderedSet,
     parse_iso8601,
+    remove_end,
 )
 
 
@@ -66,6 +67,16 @@ class CondeNastIE(InfoExtractor):
             'upload_date': '20130314',
             'timestamp': 1363219200,
         }
+    }, {
+        'url': 'http://video.gq.com/watch/the-closer-with-keith-olbermann-the-only-true-surprise-trump-s-an-idiot?c=series',
+        'info_dict': {
+            'id': '58d1865bfd2e6126e2000015',
+            'ext': 'mp4',
+            'title': 'The Only True Surprise? Trump’s an Idiot',
+            'uploader': 'gq',
+            'upload_date': '20170321',
+            'timestamp': 1490126427,
+        },
     }, {
         # JS embed
         'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js',
@@ -114,26 +125,33 @@ class CondeNastIE(InfoExtractor):
             })
         video_id = query['videoId']
         video_info = None
-        info_page = self._download_webpage(
+        info_page = self._download_json(
             'http://player.cnevids.com/player/video.js',
-            video_id, 'Downloading video info', query=query, fatal=False)
+            video_id, 'Downloading video info', fatal=False, query=query)
         if info_page:
-            video_info = self._parse_json(self._search_regex(
-                r'loadCallback\(({.+})\)', info_page, 'video info'), video_id)['video']
-        else:
+            video_info = info_page.get('video')
+        if not video_info:
             info_page = self._download_webpage(
                 'http://player.cnevids.com/player/loader.js',
                 video_id, 'Downloading loader info', query=query)
-            video_info = self._parse_json(self._search_regex(
-                r'var\s+video\s*=\s*({.+?});', info_page, 'video info'), video_id)
+            video_info = self._parse_json(
+                self._search_regex(
+                    r'(?s)var\s+config\s*=\s*({.+?});', info_page, 'config'),
+                video_id, transform_source=js_to_json)['video']
+
         title = video_info['title']
 
         formats = []
-        for fdata in video_info.get('sources', [{}])[0]:
+        for fdata in video_info['sources']:
             src = fdata.get('src')
             if not src:
                 continue
             ext = mimetype2ext(fdata.get('type')) or determine_ext(src)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    src, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+                continue
             quality = fdata.get('quality')
             formats.append({
                 'format_id': ext + ('-%s' % quality if quality else ''),
@@ -169,7 +187,6 @@ class CondeNastIE(InfoExtractor):
                 path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/')))
             url_type = 'embed'
 
-        self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site])
         webpage = self._download_webpage(url, item_id)
 
         if url_type == 'series':

youtube_dl/extractor/corus.py

@@ -0,0 +1,72 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .theplatform import ThePlatformFeedIE
+from ..utils import int_or_none
+
+
+class CorusIE(ThePlatformFeedIE):
+    _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:globaltv|etcanada)\.com|(?:hgtv|foodnetwork|slice)\.ca)/(?:video/|(?:[^/]+/)+(?:videos/[a-z0-9-]+-|video\.html\?.*?\bv=))(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://www.hgtv.ca/shows/bryan-inc/videos/movie-night-popcorn-with-bryan-870923331648/',
+        'md5': '05dcbca777bf1e58c2acbb57168ad3a6',
+        'info_dict': {
+            'id': '870923331648',
+            'ext': 'mp4',
+            'title': 'Movie Night Popcorn with Bryan',
+            'description': 'Bryan whips up homemade popcorn, the old fashion way for Jojo and Lincoln.',
+            'uploader': 'SHWM-NEW',
+            'upload_date': '20170206',
+            'timestamp': 1486392197,
+        },
+    }, {
+        'url': 'http://www.foodnetwork.ca/shows/chopped/video/episode/chocolate-obsession/video.html?v=872683587753',
+        'only_matching': True,
+    }, {
+        'url': 'http://etcanada.com/video/873675331955/meet-the-survivor-game-changers-castaways-part-2/',
+        'only_matching': True,
+    }]
+
+    _TP_FEEDS = {
+        'globaltv': {
+            'feed_id': 'ChQqrem0lNUp',
+            'account_id': 2269680845,
+        },
+        'etcanada': {
+            'feed_id': 'ChQqrem0lNUp',
+            'account_id': 2269680845,
+        },
+        'hgtv': {
+            'feed_id': 'L0BMHXi2no43',
+            'account_id': 2414428465,
+        },
+        'foodnetwork': {
+            'feed_id': 'ukK8o58zbRmJ',
+            'account_id': 2414429569,
+        },
+        'slice': {
+            'feed_id': '5tUJLgV2YNJ5',
+            'account_id': 2414427935,
+        },
+    }
+
+    def _real_extract(self, url):
+        domain, video_id = re.match(self._VALID_URL, url).groups()
+        feed_info = self._TP_FEEDS[domain.split('.')[0]]
+        return self._extract_feed_info('dtjsEC', feed_info['feed_id'], 'byId=' + video_id, video_id, lambda e: {
+            'episode_number': int_or_none(e.get('pl1$episode')),
+            'season_number': int_or_none(e.get('pl1$season')),
+            'series': e.get('pl1$show'),
+        }, {
+            'HLS': {
+                'manifest': 'm3u',
+            },
+            'DesktopHLS Default': {
+                'manifest': 'm3u',
+            },
+            'MP4 MBR': {
+                'manifest': 'm3u',
+            },
+        }, feed_info['account_id'])

youtube_dl/extractor/crackle.py

@@ -6,6 +6,7 @@ from ..utils import int_or_none
 
 
 class CrackleIE(InfoExtractor):
+    _GEO_COUNTRIES = ['US']
     _VALID_URL = r'(?:crackle:|https?://(?:(?:www|m)\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
     _TEST = {
         'url': 'http://www.crackle.com/comedians-in-cars-getting-coffee/2498934',

youtube_dl/extractor/crunchyroll.py

@@ -123,7 +123,7 @@ class CrunchyrollIE(CrunchyrollBaseIE):
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
         'info_dict': {
             'id': '645513',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
             'description': 'md5:2d17137920c64f2f49981a7797d275ef',
             'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
@@ -177,6 +177,7 @@ class CrunchyrollIE(CrunchyrollBaseIE):
             'uploader': 'Kadokawa Pictures Inc.',
             'upload_date': '20170118',
             'series': "KONOSUBA -God's blessing on this wonderful world!",
+            'season': "KONOSUBA -God's blessing on this wonderful world! 2",
             'season_number': 2,
             'episode': 'Give Me Deliverance from this Judicial Injustice!',
             'episode_number': 1,
@@ -192,6 +193,53 @@ class CrunchyrollIE(CrunchyrollBaseIE):
         # geo-restricted (US), 18+ maturity wall, non-premium available
         'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617',
         'only_matching': True,
+    }, {
+        # A description with double quotes
+        'url': 'http://www.crunchyroll.com/11eyes/episode-1-piros-jszaka-red-night-535080',
+        'info_dict': {
+            'id': '535080',
+            'ext': 'mp4',
+            'title': '11eyes Episode 1 – Piros éjszaka - Red Night',
+            'description': 'Kakeru and Yuka are thrown into an alternate nightmarish world they call "Red Night".',
+            'uploader': 'Marvelous AQL Inc.',
+            'upload_date': '20091021',
+        },
+        'params': {
+            # Just test metadata extraction
+            'skip_download': True,
+        },
+    }, {
+        # make sure we can extract an uploader name that's not a link
+        'url': 'http://www.crunchyroll.com/hakuoki-reimeiroku/episode-1-dawn-of-the-divine-warriors-606899',
+        'info_dict': {
+            'id': '606899',
+            'ext': 'mp4',
+            'title': 'Hakuoki Reimeiroku Episode 1 – Dawn of the Divine Warriors',
+            'description': 'Ryunosuke was left to die, but Serizawa-san asked him a simple question "Do you want to live?"',
+            'uploader': 'Geneon Entertainment',
+            'upload_date': '20120717',
+        },
+        'params': {
+            # just test metadata extraction
+            'skip_download': True,
+        },
+    }, {
+        # A video with a vastly different season name compared to the series name
+        'url': 'http://www.crunchyroll.com/nyarko-san-another-crawling-chaos/episode-1-test-590532',
+        'info_dict': {
+            'id': '590532',
+            'ext': 'mp4',
+            'title': 'Haiyoru! Nyaruani (ONA) Episode 1 – Test',
+            'description': 'Mahiro and Nyaruko talk about official certification.',
+            'uploader': 'TV TOKYO',
+            'upload_date': '20120305',
+            'series': 'Nyarko-san: Another Crawling Chaos',
+            'season': 'Haiyoru! Nyaruani (ONA)',
+        },
+        'params': {
+            # Just test metadata extraction
+            'skip_download': True,
+        },
     }]
 
     _FORMAT_IDS = {
@@ -342,7 +390,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         else:
             webpage_url = 'http://www.' + mobj.group('url')
 
-        webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage')
+        webpage = self._download_webpage(
+            self._add_skip_wall(webpage_url), video_id,
+            headers=self.geo_verification_headers())
         note_m = self._html_search_regex(
             r'<div class="showmedia-trailer-notice">(.+?)</div>',
             webpage, 'trailer-notice', default='')
@@ -362,9 +412,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
             webpage, 'video_title')
         video_title = re.sub(r' {2,}', ' ', video_title)
-        video_description = self._html_search_regex(
-            r'<script[^>]*>\s*.+?\[media_id=%s\].+?"description"\s*:\s*"([^"]+)' % video_id,
-            webpage, 'description', default=None)
+        video_description = self._parse_json(self._html_search_regex(
+            r'<script[^>]*>\s*.+?\[media_id=%s\].+?({.+?"description"\s*:.+?})\);' % video_id,
+            webpage, 'description', default='{}'), video_id).get('description')
         if video_description:
             video_description = lowercase_escape(video_description.replace(r'\r\n', '\n'))
         video_upload_date = self._html_search_regex(
@@ -373,8 +423,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         if video_upload_date:
             video_upload_date = unified_strdate(video_upload_date)
         video_uploader = self._html_search_regex(
-            r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage,
-            'video_uploader', fatal=False)
+            # try looking for both an uploader that's a link and one that's not
+            [r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', r'<div>\s*Publisher:\s*<span>\s*(.+?)\s*</span>\s*</div>'],
+            webpage, 'video_uploader', fatal=False)
 
         available_fmts = []
         for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
@@ -460,7 +511,8 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         # webpage provide more accurate data than series_title from XML
         series = self._html_search_regex(
             r'id=["\']showmedia_about_episode_num[^>]+>\s*<a[^>]+>([^<]+)',
-            webpage, 'series', default=xpath_text(metadata, 'series_title'))
+            webpage, 'series', fatal=False)
+        season = xpath_text(metadata, 'series_title')
 
         episode = xpath_text(metadata, 'episode_title')
         episode_number = int_or_none(xpath_text(metadata, 'episode_number'))
@@ -477,6 +529,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             'uploader': video_uploader,
             'upload_date': video_upload_date,
             'series': series,
+            'season': season,
             'season_number': season_number,
             'episode': episode,
             'episode_number': episode_number,
@@ -514,16 +567,18 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
     def _real_extract(self, url):
         show_id = self._match_id(url)
 
-        webpage = self._download_webpage(self._add_skip_wall(url), show_id)
+        webpage = self._download_webpage(
+            self._add_skip_wall(url), show_id,
+            headers=self.geo_verification_headers())
         title = self._html_search_regex(
             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
             webpage, 'title')
         episode_paths = re.findall(
-            r'(?s)<li id="showview_videos_media_[0-9]+"[^>]+>.*?<a href="([^"]+)"',
+            r'(?s)<li id="showview_videos_media_(\d+)"[^>]+>.*?<a href="([^"]+)"',
             webpage)
         entries = [
-            self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
-            for ep in episode_paths
+            self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll', ep_id)
+            for ep_id, ep in episode_paths
         ]
         entries.reverse()
 

youtube_dl/extractor/curiositystream.py

@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -46,9 +48,50 @@ class CuriosityStreamBaseIE(InfoExtractor):
 
     def _extract_media_info(self, media):
         video_id = compat_str(media['id'])
-        limelight_media_id = media['limelight_media_id']
         title = media['title']
 
+        formats = []
+        for encoding in media.get('encodings', []):
+            m3u8_url = encoding.get('master_playlist_url')
+            if m3u8_url:
+                formats.extend(self._extract_m3u8_formats(
+                    m3u8_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
+            encoding_url = encoding.get('url')
+            file_url = encoding.get('file_url')
+            if not encoding_url and not file_url:
+                continue
+            f = {
+                'width': int_or_none(encoding.get('width')),
+                'height': int_or_none(encoding.get('height')),
+                'vbr': int_or_none(encoding.get('video_bitrate')),
+                'abr': int_or_none(encoding.get('audio_bitrate')),
+                'filesize': int_or_none(encoding.get('size_in_bytes')),
+                'vcodec': encoding.get('video_codec'),
+                'acodec': encoding.get('audio_codec'),
+                'container': encoding.get('container_type'),
+            }
+            for f_url in (encoding_url, file_url):
+                if not f_url:
+                    continue
+                fmt = f.copy()
+                rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', f_url)
+                if rtmp:
+                    fmt.update({
+                        'url': rtmp.group('url'),
+                        'play_path': rtmp.group('playpath'),
+                        'app': rtmp.group('app'),
+                        'ext': 'flv',
+                        'format_id': 'rtmp',
+                    })
+                else:
+                    fmt.update({
+                        'url': f_url,
+                        'format_id': 'http',
+                    })
+                formats.append(fmt)
+        self._sort_formats(formats)
+
         subtitles = {}
         for closed_caption in media.get('closed_captions', []):
             sub_url = closed_caption.get('file')
@@ -60,16 +103,14 @@ class CuriosityStreamBaseIE(InfoExtractor):
             })
 
         return {
-            '_type': 'url_transparent',
             'id': video_id,
-            'url': 'limelight:media:' + limelight_media_id,
+            'formats': formats,
             'title': title,
             'description': media.get('description'),
             'thumbnail': media.get('image_large') or media.get('image_medium') or media.get('image_small'),
             'duration': int_or_none(media.get('duration')),
             'tags': media.get('tags'),
             'subtitles': subtitles,
-            'ie_key': 'LimelightMedia',
         }
 
 
@@ -78,14 +119,12 @@ class CuriosityStreamIE(CuriosityStreamBaseIE):
     _VALID_URL = r'https?://app\.curiositystream\.com/video/(?P<id>\d+)'
     _TEST = {
         'url': 'https://app.curiositystream.com/video/2',
-        'md5': 'a0074c190e6cddaf86900b28d3e9ee7a',
+        'md5': '262bb2f257ff301115f1973540de8983',
         'info_dict': {
             'id': '2',
             'ext': 'mp4',
             'title': 'How Did You Develop The Internet?',
             'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.',
-            'timestamp': 1448388615,
-            'upload_date': '20151124',
         }
     }
 
@@ -105,7 +144,7 @@ class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
             'title': 'Curious Minds: The Internet',
             'description': 'How is the internet shaping our lives in the 21st Century?',
         },
-        'playlist_mincount': 17,
+        'playlist_mincount': 12,
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/cwtv.py

@@ -82,6 +82,11 @@ class CWTVIE(InfoExtractor):
                             'url': quality_url,
                             'tbr': tbr,
                         })
+        video_metadata = video_data['assetFields']
+        ism_url = video_metadata.get('smoothStreamingUrl')
+        if ism_url:
+            formats.extend(self._extract_ism_formats(
+                ism_url, video_id, ism_id='mss', fatal=False))
         self._sort_formats(formats)
 
         thumbnails = [{
@@ -90,8 +95,6 @@ class CWTVIE(InfoExtractor):
             'height': image.get('height'),
         } for image_id, image in video_data['images'].items() if image.get('uri')] if video_data.get('images') else None
 
-        video_metadata = video_data['assetFields']
-
         subtitles = {
             'en': [{
                 'url': video_metadata['UnicornCcUrl'],

youtube_dl/extractor/dailymotion.py

@@ -66,7 +66,6 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                 'uploader_id': 'xijv66',
                 'age_limit': 0,
                 'view_count': int,
-                'comment_count': int,
             }
         },
         # Vevo video
@@ -140,7 +139,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         view_count = str_to_int(view_count_str)
         comment_count = int_or_none(self._search_regex(
             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
-            webpage, 'comment count', fatal=False))
+            webpage, 'comment count', default=None))
 
         player_v5 = self._search_regex(
             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
@@ -283,9 +282,14 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         }
 
     def _check_error(self, info):
+        error = info.get('error')
         if info.get('error') is not None:
+            title = error['title']
+            # See https://developer.dailymotion.com/api#access-error
+            if error.get('code') == 'DM007':
+                self.raise_geo_restricted(msg=title)
             raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
+                '%s said: %s' % (self.IE_NAME, title), expected=True)
 
     def _get_subtitles(self, video_id, webpage):
         try:

youtube_dl/extractor/daisuki.py

@@ -0,0 +1,159 @@
+from __future__ import unicode_literals
+
+import base64
+import json
+import random
+import re
+
+from .common import InfoExtractor
+from ..aes import (
+    aes_cbc_decrypt,
+    aes_cbc_encrypt,
+)
+from ..utils import (
+    bytes_to_intlist,
+    bytes_to_long,
+    clean_html,
+    ExtractorError,
+    intlist_to_bytes,
+    get_element_by_id,
+    js_to_json,
+    int_or_none,
+    long_to_bytes,
+    pkcs1pad,
+    remove_end,
+)
+
+
+class DaisukiIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?daisuki\.net/[^/]+/[^/]+/[^/]+/watch\.[^.]+\.(?P<id>\d+)\.html'
+
+    _TEST = {
+        'url': 'http://www.daisuki.net/tw/en/anime/watch.TheIdolMasterCG.11213.html',
+        'info_dict': {
+            'id': '11213',
+            'ext': 'mp4',
+            'title': '#01 Who is in the pumpkin carriage? - THE IDOLM@STER CINDERELLA GIRLS',
+            'subtitles': {
+                'mul': [{
+                    'ext': 'ttml',
+                }],
+            },
+            'creator': 'BANDAI NAMCO Entertainment',
+        },
+        'params': {
+            'skip_download': True,  # AES-encrypted HLS stream
+        },
+    }
+
+    # The public key in PEM format can be found in clientlibs_anime_watch.min.js
+    _RSA_KEY = (0xc5524c25e8e14b366b3754940beeb6f96cb7e2feef0b932c7659a0c5c3bf173d602464c2df73d693b513ae06ff1be8f367529ab30bf969c5640522181f2a0c51ea546ae120d3d8d908595e4eff765b389cde080a1ef7f1bbfb07411cc568db73b7f521cedf270cbfbe0ddbc29b1ac9d0f2d8f4359098caffee6d07915020077d, 65537)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        flashvars = self._parse_json(self._search_regex(
+            r'(?s)var\s+flashvars\s*=\s*({.+?});', webpage, 'flashvars'),
+            video_id, transform_source=js_to_json)
+
+        iv = [0] * 16
+
+        data = {}
+        for key in ('device_cd', 'mv_id', 'ss1_prm', 'ss2_prm', 'ss3_prm', 'ss_id'):
+            data[key] = flashvars.get(key, '')
+
+        encrypted_rtn = None
+
+        # Some AES keys are rejected. Try it with different AES keys
+        for idx in range(5):
+            aes_key = [random.randint(0, 254) for _ in range(32)]
+            padded_aeskey = intlist_to_bytes(pkcs1pad(aes_key, 128))
+
+            n, e = self._RSA_KEY
+            encrypted_aeskey = long_to_bytes(pow(bytes_to_long(padded_aeskey), e, n))
+            init_data = self._download_json('http://www.daisuki.net/bin/bgn/init', video_id, query={
+                's': flashvars.get('s', ''),
+                'c': flashvars.get('ss3_prm', ''),
+                'e': url,
+                'd': base64.b64encode(intlist_to_bytes(aes_cbc_encrypt(
+                    bytes_to_intlist(json.dumps(data)),
+                    aes_key, iv))).decode('ascii'),
+                'a': base64.b64encode(encrypted_aeskey).decode('ascii'),
+            }, note='Downloading JSON metadata' + (' (try #%d)' % (idx + 1) if idx > 0 else ''))
+
+            if 'rtn' in init_data:
+                encrypted_rtn = init_data['rtn']
+                break
+
+            self._sleep(5, video_id)
+
+        if encrypted_rtn is None:
+            raise ExtractorError('Failed to fetch init data')
+
+        rtn = self._parse_json(
+            intlist_to_bytes(aes_cbc_decrypt(bytes_to_intlist(
+                base64.b64decode(encrypted_rtn)),
+                aes_key, iv)).decode('utf-8').rstrip('\0'),
+            video_id)
+
+        formats = self._extract_m3u8_formats(
+            rtn['play_url'], video_id, ext='mp4', entry_protocol='m3u8_native')
+
+        title = remove_end(self._og_search_title(webpage), ' - DAISUKI')
+
+        creator = self._html_search_regex(
+            r'Creator\s*:\s*([^<]+)', webpage, 'creator', fatal=False)
+
+        subtitles = {}
+        caption_url = rtn.get('caption_url')
+        if caption_url:
+            # mul: multiple languages
+            subtitles['mul'] = [{
+                'url': caption_url,
+                'ext': 'ttml',
+            }]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'subtitles': subtitles,
+            'creator': creator,
+        }
+
+
+class DaisukiPlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)daisuki\.net/[^/]+/[^/]+/[^/]+/detail\.(?P<id>[a-zA-Z0-9]+)\.html'
+
+    _TEST = {
+        'url': 'http://www.daisuki.net/tw/en/anime/detail.TheIdolMasterCG.html',
+        'info_dict': {
+            'id': 'TheIdolMasterCG',
+            'title': 'THE IDOLM@STER CINDERELLA GIRLS',
+            'description': 'md5:0f2c028a9339f7a2c7fbf839edc5c5d8',
+        },
+        'playlist_count': 26,
+    }
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        episode_pattern = r'''(?sx)
+            <img[^>]+delay="[^"]+/(\d+)/movie\.jpg".+?
+            <p[^>]+class=".*?\bepisodeNumber\b.*?">(?:<a[^>]+>)?([^<]+)'''
+        entries = [{
+            '_type': 'url_transparent',
+            'url': url.replace('detail', 'watch').replace('.html', '.' + movie_id + '.html'),
+            'episode_id': episode_id,
+            'episode_number': int_or_none(episode_id),
+        } for movie_id, episode_id in re.findall(episode_pattern, webpage)]
+
+        playlist_title = remove_end(
+            self._og_search_title(webpage, fatal=False), ' - Anime - DAISUKI')
+        playlist_description = clean_html(get_element_by_id('synopsisTxt', webpage))
+
+        return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)

youtube_dl/extractor/discoverygo.py

@@ -1,17 +1,21 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
     extract_attributes,
+    ExtractorError,
     int_or_none,
     parse_age_limit,
-    ExtractorError,
+    remove_end,
+    unescapeHTML,
 )
 
 
-class DiscoveryGoIE(InfoExtractor):
-    _VALID_URL = r'''(?x)https?://(?:www\.)?(?:
+class DiscoveryGoBaseIE(InfoExtractor):
+    _VALID_URL_TEMPLATE = r'''(?x)https?://(?:www\.)?(?:
             discovery|
             investigationdiscovery|
             discoverylife|
@@ -21,18 +25,23 @@ class DiscoveryGoIE(InfoExtractor):
             sciencechannel|
             tlc|
             velocitychannel
-        )go\.com/(?:[^/]+/)*(?P<id>[^/?#&]+)'''
+        )go\.com/%s(?P<id>[^/?#&]+)'''
+
+
+class DiscoveryGoIE(DiscoveryGoBaseIE):
+    _VALID_URL = DiscoveryGoBaseIE._VALID_URL_TEMPLATE % r'(?:[^/]+/)+'
+    _GEO_COUNTRIES = ['US']
     _TEST = {
-        'url': 'https://www.discoverygo.com/love-at-first-kiss/kiss-first-ask-questions-later/',
+        'url': 'https://www.discoverygo.com/bering-sea-gold/reaper-madness/',
         'info_dict': {
-            'id': '57a33c536b66d1cd0345eeb1',
+            'id': '58c167d86b66d12f2addeb01',
             'ext': 'mp4',
-            'title': 'Kiss First, Ask Questions Later!',
-            'description': 'md5:fe923ba34050eae468bffae10831cb22',
-            'duration': 2579,
-            'series': 'Love at First Kiss',
-            'season_number': 1,
-            'episode_number': 1,
+            'title': 'Reaper Madness',
+            'description': 'md5:09f2c625c99afb8946ed4fb7865f6e78',
+            'duration': 2519,
+            'series': 'Bering Sea Gold',
+            'season_number': 8,
+            'episode_number': 6,
             'age_limit': 14,
         },
     }
@@ -113,3 +122,46 @@ class DiscoveryGoIE(InfoExtractor):
             'formats': formats,
             'subtitles': subtitles,
         }
+
+
+class DiscoveryGoPlaylistIE(DiscoveryGoBaseIE):
+    _VALID_URL = DiscoveryGoBaseIE._VALID_URL_TEMPLATE % ''
+    _TEST = {
+        'url': 'https://www.discoverygo.com/bering-sea-gold/',
+        'info_dict': {
+            'id': 'bering-sea-gold',
+            'title': 'Bering Sea Gold',
+            'description': 'md5:cc5c6489835949043c0cc3ad66c2fa0e',
+        },
+        'playlist_mincount': 6,
+    }
+
+    @classmethod
+    def suitable(cls, url):
+        return False if DiscoveryGoIE.suitable(url) else super(
+            DiscoveryGoPlaylistIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        entries = []
+        for mobj in re.finditer(r'data-json=(["\'])(?P<json>{.+?})\1', webpage):
+            data = self._parse_json(
+                mobj.group('json'), display_id,
+                transform_source=unescapeHTML, fatal=False)
+            if not isinstance(data, dict) or data.get('type') != 'episode':
+                continue
+            episode_url = data.get('socialUrl')
+            if not episode_url:
+                continue
+            entries.append(self.url_result(
+                episode_url, ie=DiscoveryGoIE.ie_key(),
+                video_id=data.get('id')))
+
+        return self.playlist_result(
+            entries, display_id,
+            remove_end(self._og_search_title(
+                webpage, fatal=False), ' | Discovery GO'),
+            self._og_search_description(webpage))

youtube_dl/extractor/discoverynetworks.py

@@ -9,13 +9,13 @@ from ..compat import (
     compat_parse_qs,
     compat_urlparse,
 )
+from ..utils import smuggle_url
 
 
-class TlcDeIE(InfoExtractor):
-    IE_NAME = 'tlc.de'
-    _VALID_URL = r'https?://(?:www\.)?tlc\.de/(?:[^/]+/)*videos/(?P<title>[^/?#]+)?(?:.*#(?P<id>\d+))?'
+class DiscoveryNetworksDeIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?(?:discovery|tlc|animalplanet|dmax)\.de/(?:.*#(?P<id>\d+)|(?:[^/]+/)*videos/(?P<title>[^/?#]+))'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.tlc.de/sendungen/breaking-amish/videos/#3235167922001',
         'info_dict': {
             'id': '3235167922001',
@@ -29,7 +29,13 @@ class TlcDeIE(InfoExtractor):
             'upload_date': '20140404',
             'uploader_id': '1659832546',
         },
-    }
+    }, {
+        'url': 'http://www.dmax.de/programme/storage-hunters-uk/videos/storage-hunters-uk-episode-6/',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.discovery.de/#5332316765001',
+        'only_matching': True,
+    }]
     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1659832546/default_default/index.html?videoId=%s'
 
     def _real_extract(self, url):
@@ -39,5 +45,8 @@ class TlcDeIE(InfoExtractor):
             title = mobj.group('title')
             webpage = self._download_webpage(url, title)
             brightcove_legacy_url = BrightcoveLegacyIE._extract_brightcove_url(webpage)
-            brightcove_id = compat_parse_qs(compat_urlparse.urlparse(brightcove_legacy_url).query)['@videoPlayer'][0]
-        return self.url_result(self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew', brightcove_id)
+            brightcove_id = compat_parse_qs(compat_urlparse.urlparse(
+                brightcove_legacy_url).query)['@videoPlayer'][0]
+        return self.url_result(smuggle_url(
+            self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, {'geo_countries': ['DE']}),
+            'BrightcoveNew', brightcove_id)

youtube_dl/extractor/discoveryvr.py

@@ -0,0 +1,59 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import parse_duration
+
+
+class DiscoveryVRIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?discoveryvr\.com/watch/(?P<id>[^/?#]+)'
+    _TEST = {
+        'url': 'http://www.discoveryvr.com/watch/discovery-vr-an-introduction',
+        'md5': '32b1929798c464a54356378b7912eca4',
+        'info_dict': {
+            'id': 'discovery-vr-an-introduction',
+            'ext': 'mp4',
+            'title': 'Discovery VR - An Introduction',
+            'description': 'md5:80d418a10efb8899d9403e61d8790f06',
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        bootstrap_data = self._search_regex(
+            r'root\.DVR\.bootstrapData\s+=\s+"({.+?})";',
+            webpage, 'bootstrap data')
+        bootstrap_data = self._parse_json(
+            bootstrap_data.encode('utf-8').decode('unicode_escape'),
+            display_id)
+        videos = self._parse_json(bootstrap_data['videos'], display_id)['allVideos']
+        video_data = next(video for video in videos if video.get('slug') == display_id)
+
+        series = video_data.get('showTitle')
+        title = episode = video_data.get('title') or series
+        if series and series != title:
+            title = '%s - %s' % (series, title)
+
+        formats = []
+        for f, format_id in (('cdnUriM3U8', 'mobi'), ('webVideoUrlSd', 'sd'), ('webVideoUrlHd', 'hd')):
+            f_url = video_data.get(f)
+            if not f_url:
+                continue
+            formats.append({
+                'format_id': format_id,
+                'url': f_url,
+            })
+
+        return {
+            'id': display_id,
+            'display_id': display_id,
+            'title': title,
+            'description': video_data.get('description'),
+            'thumbnail': video_data.get('thumbnail'),
+            'duration': parse_duration(video_data.get('runTime')),
+            'formats': formats,
+            'episode': episode,
+            'series': series,
+        }

youtube_dl/extractor/disney.py

@@ -9,13 +9,15 @@ from ..utils import (
     unified_strdate,
     compat_str,
     determine_ext,
+    ExtractorError,
 )
 
 
 class DisneyIE(InfoExtractor):
     _VALID_URL = r'''(?x)
-        https?://(?P<domain>(?:[^/]+\.)?(?:disney\.[a-z]{2,3}(?:\.[a-z]{2})?|disney(?:(?:me|latino)\.com|turkiye\.com\.tr)|starwars\.com))/(?:embed/|(?:[^/]+/)+[\w-]+-)(?P<id>[a-z0-9]{24})'''
+        https?://(?P<domain>(?:[^/]+\.)?(?:disney\.[a-z]{2,3}(?:\.[a-z]{2})?|disney(?:(?:me|latino)\.com|turkiye\.com\.tr)|(?:starwars|marvelkids)\.com))/(?:(?:embed/|(?:[^/]+/)+[\w-]+-)(?P<id>[a-z0-9]{24})|(?:[^/]+/)?(?P<display_id>[^/?#]+))'''
     _TESTS = [{
+        # Disney.EmbedVideo
         'url': 'http://video.disney.com/watch/moana-trailer-545ed1857afee5a0ec239977',
         'info_dict': {
             'id': '545ed1857afee5a0ec239977',
@@ -28,6 +30,20 @@ class DisneyIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        # Grill.burger
+        'url': 'http://www.starwars.com/video/rogue-one-a-star-wars-story-intro-featurette',
+        'info_dict': {
+            'id': '5454e9f4e9804a552e3524c8',
+            'ext': 'mp4',
+            'title': '"Intro" Featurette: Rogue One: A Star Wars Story',
+            'upload_date': '20170104',
+            'description': 'Go behind-the-scenes of Rogue One: A Star Wars Story in this featurette with Director Gareth Edwards and the cast of the film.',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
     }, {
         'url': 'http://videos.disneylatino.com/ver/spider-man-de-regreso-a-casa-primer-adelanto-543a33a1850bdcfcca13bae2',
         'only_matching': True,
@@ -43,31 +59,55 @@ class DisneyIE(InfoExtractor):
     }, {
         'url': 'http://www.starwars.com/embed/54690d1e6c42e5f09a0fb097',
         'only_matching': True,
+    }, {
+        'url': 'http://spiderman.marvelkids.com/embed/522900d2ced3c565e4cc0677',
+        'only_matching': True,
+    }, {
+        'url': 'http://spiderman.marvelkids.com/videos/contest-of-champions-part-four-clip-1',
+        'only_matching': True,
+    }, {
+        'url': 'http://disneyjunior.en.disneyme.com/dj/watch-my-friends-tigger-and-pooh-promo',
+        'only_matching': True,
+    }, {
+        'url': 'http://disneyjunior.disney.com/galactech-the-galactech-grab-galactech-an-admiral-rescue',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        domain, video_id = re.match(self._VALID_URL, url).groups()
-        webpage = self._download_webpage(
-            'http://%s/embed/%s' % (domain, video_id), video_id)
-        video_data = self._parse_json(self._search_regex(
-            r'Disney\.EmbedVideo=({.+});', webpage, 'embed data'), video_id)['video']
+        domain, video_id, display_id = re.match(self._VALID_URL, url).groups()
+        if not video_id:
+            webpage = self._download_webpage(url, display_id)
+            grill = re.sub(r'"\s*\+\s*"', '', self._search_regex(
+                r'Grill\.burger\s*=\s*({.+})\s*:',
+                webpage, 'grill data'))
+            page_data = next(s for s in self._parse_json(grill, display_id)['stack'] if s.get('type') == 'video')
+            video_data = page_data['data'][0]
+        else:
+            webpage = self._download_webpage(
+                'http://%s/embed/%s' % (domain, video_id), video_id)
+            page_data = self._parse_json(self._search_regex(
+                r'Disney\.EmbedVideo\s*=\s*({.+});',
+                webpage, 'embed data'), video_id)
+            video_data = page_data['video']
 
         for external in video_data.get('externals', []):
             if external.get('source') == 'vevo':
                 return self.url_result('vevo:' + external['data_id'], 'Vevo')
 
+        video_id = video_data['id']
         title = video_data['title']
 
         formats = []
         for flavor in video_data.get('flavors', []):
             flavor_format = flavor.get('format')
             flavor_url = flavor.get('url')
-            if not flavor_url or not re.match(r'https?://', flavor_url):
+            if not flavor_url or not re.match(r'https?://', flavor_url) or flavor_format == 'mp4_access':
                 continue
             tbr = int_or_none(flavor.get('bitrate'))
             if tbr == 99999:
                 formats.extend(self._extract_m3u8_formats(
-                    flavor_url, video_id, 'mp4', m3u8_id=flavor_format, fatal=False))
+                    flavor_url, video_id, 'mp4',
+                    m3u8_id=flavor_format, fatal=False))
                 continue
             format_id = []
             if flavor_format:
@@ -88,6 +128,10 @@ class DisneyIE(InfoExtractor):
                 'ext': ext,
                 'vcodec': 'none' if (width == 0 and height == 0) else None,
             })
+        if not formats and video_data.get('expired'):
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, page_data['translations']['video_expired']),
+                expected=True)
         self._sort_formats(formats)
 
         subtitles = {}

youtube_dl/extractor/douyutv.py

@@ -1,15 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import hashlib
 import time
-import uuid
+import hashlib
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_str,
-    compat_urllib_parse_urlencode,
-)
 from ..utils import (
     ExtractorError,
     unescapeHTML,
@@ -25,7 +20,7 @@ class DouyuTVIE(InfoExtractor):
             'id': '17732',
             '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}$',
+            'title': 're:^清晨醒脑！T-ARA根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
             'description': r're:.*m7show@163\.com.*',
             'thumbnail': r're:^https?://.*\.jpg$',
             'uploader': '7师傅',
@@ -56,7 +51,7 @@ class DouyuTVIE(InfoExtractor):
             'id': '17732',
             'display_id': '17732',
             'ext': 'flv',
-            'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            '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}$',
             'description': r're:.*m7show@163\.com.*',
             'thumbnail': r're:^https?://.*\.jpg$',
             'uploader': '7师傅',
@@ -74,10 +69,6 @@ class DouyuTVIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    # Decompile core.swf in webpage by ffdec "Search SWFs in memory". core.swf
-    # is encrypted originally, but ffdec can dump memory to get the decrypted one.
-    _API_KEY = 'A12Svb&%1UUmf@hC'
-
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -88,6 +79,7 @@ class DouyuTVIE(InfoExtractor):
             room_id = self._html_search_regex(
                 r'"room_id\\?"\s*:\s*(\d+),', page, 'room id')
 
+        # Grab metadata from mobile API
         room = self._download_json(
             'http://m.douyu.com/html5/live?roomId=%s' % room_id, video_id,
             note='Downloading room info')['data']
@@ -96,38 +88,22 @@ class DouyuTVIE(InfoExtractor):
         if room.get('show_status') == '2':
             raise ExtractorError('Live stream is offline', expected=True)
 
-        tt = compat_str(int(time.time() / 60))
-        did = uuid.uuid4().hex.upper()
-
-        sign_content = ''.join((room_id, did, self._API_KEY, tt))
-        sign = hashlib.md5((sign_content).encode('utf-8')).hexdigest()
-
-        flv_data = compat_urllib_parse_urlencode({
-            'cdn': 'ws',
-            'rate': '0',
-            'tt': tt,
-            'did': did,
-            'sign': sign,
-        })
-
-        video_info = self._download_json(
-            'http://www.douyu.com/lapi/live/getPlay/%s' % room_id, video_id,
-            data=flv_data, note='Downloading video info',
-            headers={'Content-Type': 'application/x-www-form-urlencoded'})
-
-        error_code = video_info.get('error', 0)
-        if error_code is not 0:
-            raise ExtractorError(
-                '%s reported error %i' % (self.IE_NAME, error_code),
-                expected=True)
-
-        base_url = video_info['data']['rtmp_url']
-        live_path = video_info['data']['rtmp_live']
-
-        video_url = '%s/%s' % (base_url, live_path)
+        # Grab the URL from PC client API
+        # The m3u8 url from mobile API requires re-authentication every 5 minutes
+        tt = int(time.time())
+        signContent = 'lapi/live/thirdPart/getPlay/%s?aid=pcclient&rate=0&time=%d9TUk5fjjUjg9qIMH3sdnh' % (room_id, tt)
+        sign = hashlib.md5(signContent.encode('ascii')).hexdigest()
+        video_url = self._download_json(
+            'http://coapi.douyucdn.cn/lapi/live/thirdPart/getPlay/' + room_id,
+            video_id, note='Downloading video URL info',
+            query={'rate': 0}, headers={
+                'auth': sign,
+                'time': str(tt),
+                'aid': 'pcclient'
+            })['data']['live_url']
 
         title = self._live_title(unescapeHTML(room['room_name']))
-        description = room.get('notice')
+        description = room.get('show_details')
         thumbnail = room.get('room_src')
         uploader = room.get('nickname')
 

youtube_dl/extractor/dplay.py

@@ -6,37 +6,24 @@ import re
 import time
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_urlparse,
+    compat_HTTPError,
+)
 from ..utils import (
     USER_AGENTS,
+    ExtractorError,
     int_or_none,
+    unified_strdate,
+    remove_end,
     update_url_query,
 )
 
 
 class DPlayIE(InfoExtractor):
-    _VALID_URL = r'https?://(?P<domain>it\.dplay\.com|www\.dplay\.(?:dk|se|no))/[^/]+/(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?P<domain>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',
-            'display_id': 'stagione-1-episodio-25',
-            'ext': 'mp4',
-            'title': 'Episodio 25',
-            'description': 'md5:cae5f40ad988811b197d2d27a53227eb',
-            'duration': 2761,
-            'timestamp': 1454701800,
-            'upload_date': '20160205',
-            'creator': 'RTIT',
-            'series': 'Take me out',
-            'season_number': 1,
-            'episode_number': 25,
-            'age_limit': 0,
-        },
-        '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': {
@@ -168,3 +155,90 @@ class DPlayIE(InfoExtractor):
             'formats': formats,
             'subtitles': subtitles,
         }
+
+
+class DPlayItIE(InfoExtractor):
+    _VALID_URL = r'https?://it\.dplay\.com/[^/]+/[^/]+/(?P<id>[^/?#]+)'
+    _GEO_COUNTRIES = ['IT']
+    _TEST = {
+        'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
+        'md5': '2b808ffb00fc47b884a172ca5d13053c',
+        'info_dict': {
+            'id': '6918',
+            'display_id': 'luigi-di-maio-la-psicosi-di-stanislawskij',
+            'ext': 'mp4',
+            'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij',
+            'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
+            'thumbnail': r're:^https?://.*\.jpe?g',
+            'upload_date': '20160524',
+            'series': 'Biografie imbarazzanti',
+            'season_number': 1,
+            'episode': 'Luigi Di Maio: la psicosi di Stanislawskij',
+            'episode_number': 1,
+        },
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        info_url = self._search_regex(
+            r'url\s*:\s*["\']((?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)',
+            webpage, 'video id')
+
+        title = remove_end(self._og_search_title(webpage), ' | Dplay')
+
+        try:
+            info = self._download_json(
+                info_url, display_id, headers={
+                    'Authorization': 'Bearer %s' % self._get_cookies(url).get(
+                        'dplayit_token').value,
+                    'Referer': url,
+                })
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 403):
+                info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
+                error = info['errors'][0]
+                if error.get('code') == 'access.denied.geoblocked':
+                    self.raise_geo_restricted(
+                        msg=error.get('detail'), countries=self._GEO_COUNTRIES)
+                raise ExtractorError(info['errors'][0]['detail'], expected=True)
+            raise
+
+        hls_url = info['data']['attributes']['streaming']['hls']['url']
+
+        formats = self._extract_m3u8_formats(
+            hls_url, display_id, ext='mp4', entry_protocol='m3u8_native',
+            m3u8_id='hls')
+
+        series = self._html_search_regex(
+            r'(?s)<h1[^>]+class=["\'].*?\bshow_title\b.*?["\'][^>]*>(.+?)</h1>',
+            webpage, 'series', fatal=False)
+        episode = self._search_regex(
+            r'<p[^>]+class=["\'].*?\bdesc_ep\b.*?["\'][^>]*>\s*<br/>\s*<b>([^<]+)',
+            webpage, 'episode', fatal=False)
+
+        mobj = re.search(
+            r'(?s)<span[^>]+class=["\']dates["\'][^>]*>.+?\bS\.(?P<season_number>\d+)\s+E\.(?P<episode_number>\d+)\s*-\s*(?P<upload_date>\d{2}/\d{2}/\d{4})',
+            webpage)
+        if mobj:
+            season_number = int(mobj.group('season_number'))
+            episode_number = int(mobj.group('episode_number'))
+            upload_date = unified_strdate(mobj.group('upload_date'))
+        else:
+            season_number = episode_number = upload_date = None
+
+        return {
+            'id': info_url.rpartition('/')[-1],
+            'display_id': display_id,
+            'title': title,
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'series': series,
+            'season_number': season_number,
+            'episode': episode,
+            'episode_number': episode_number,
+            'upload_date': upload_date,
+            'formats': formats,
+        }

youtube_dl/extractor/dramafever.py

@@ -20,6 +20,7 @@ from ..utils import (
 class DramaFeverBaseIE(AMPIE):
     _LOGIN_URL = 'https://www.dramafever.com/accounts/login/'
     _NETRC_MACHINE = 'dramafever'
+    _GEO_COUNTRIES = ['US', 'CA']
 
     _CONSUMER_SECRET = 'DA59dtVXYLxajktV'
 
@@ -116,8 +117,9 @@ class DramaFeverIE(DramaFeverBaseIE):
                 'http://www.dramafever.com/amp/episode/feed.json?guid=%s' % video_id)
         except ExtractorError as e:
             if isinstance(e.cause, compat_HTTPError):
-                raise ExtractorError(
-                    'Currently unavailable in your country.', expected=True)
+                self.raise_geo_restricted(
+                    msg='Currently unavailable in your country',
+                    countries=self._GEO_COUNTRIES)
             raise
 
         series_id, episode_number = video_id.split('.')

youtube_dl/extractor/drtv.py

@@ -15,6 +15,8 @@ from ..utils import (
 
 class DRTVIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv/se|nyheder|radio/ondemand)/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
+    _GEO_BYPASS = False
+    _GEO_COUNTRIES = ['DK']
     IE_NAME = 'drtv'
     _TESTS = [{
         'url': 'https://www.dr.dk/tv/se/boern/ultra/klassen-ultra/klassen-darlig-taber-10',
@@ -137,7 +139,7 @@ class DRTVIE(InfoExtractor):
         if not formats and restricted_to_denmark:
             self.raise_geo_restricted(
                 'Unfortunately, DR is not allowed to show this program outside Denmark.',
-                expected=True)
+                countries=self._GEO_COUNTRIES)
 
         self._sort_formats(formats)
 
@@ -156,6 +158,7 @@ class DRTVIE(InfoExtractor):
 class DRTVLiveIE(InfoExtractor):
     IE_NAME = 'drtv:live'
     _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv|TV)/live/(?P<id>[\da-z-]+)'
+    _GEO_COUNTRIES = ['DK']
     _TEST = {
         'url': 'https://www.dr.dk/tv/live/dr1',
         'info_dict': {

youtube_dl/extractor/einthusan.py

@@ -1,67 +1,97 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import base64
+import json
+
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_urlparse,
+    compat_str,
+)
 from ..utils import (
-    remove_start,
-    sanitized_Request,
+    extract_attributes,
+    ExtractorError,
+    get_elements_by_class,
+    urlencode_postdata,
 )
 
 
 class EinthusanIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?einthusan\.com/movies/watch.php\?([^#]*?)id=(?P<id>[0-9]+)'
-    _TESTS = [
-        {
-            'url': 'http://www.einthusan.com/movies/watch.php?id=2447',
-            'md5': 'd71379996ff5b7f217eca034c34e3461',
-            'info_dict': {
-                'id': '2447',
-                'ext': 'mp4',
-                'title': 'Ek Villain',
-                'thumbnail': r're:^https?://.*\.jpg$',
-                'description': 'md5:9d29fc91a7abadd4591fb862fa560d93',
-            }
-        },
-        {
-            'url': 'http://www.einthusan.com/movies/watch.php?id=1671',
-            'md5': 'b16a6fd3c67c06eb7c79c8a8615f4213',
-            'info_dict': {
-                'id': '1671',
-                'ext': 'mp4',
-                'title': 'Soodhu Kavvuum',
-                'thumbnail': r're:^https?://.*\.jpg$',
-                'description': 'md5:b40f2bf7320b4f9414f3780817b2af8c',
-            }
-        },
-    ]
+    _VALID_URL = r'https?://einthusan\.tv/movie/watch/(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://einthusan.tv/movie/watch/9097/',
+        'md5': 'ff0f7f2065031b8a2cf13a933731c035',
+        'info_dict': {
+            'id': '9097',
+            'ext': 'mp4',
+            'title': 'Ae Dil Hai Mushkil',
+            'description': 'md5:33ef934c82a671a94652a9b4e54d931b',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        }
+    }, {
+        'url': 'https://einthusan.tv/movie/watch/51MZ/?lang=hindi',
+        'only_matching': True,
+    }]
+
+    # reversed from jsoncrypto.prototype.decrypt() in einthusan-PGMovieWatcher.js
+    def _decrypt(self, encrypted_data, video_id):
+        return self._parse_json(base64.b64decode((
+            encrypted_data[:10] + encrypted_data[-1] + encrypted_data[12:-1]
+        ).encode('ascii')).decode('utf-8'), video_id)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        request = sanitized_Request(url)
-        request.add_header('User-Agent', 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:43.0) Gecko/20100101 Firefox/43.0')
-        webpage = self._download_webpage(request, video_id)
+        webpage = self._download_webpage(url, video_id)
 
-        title = self._html_search_regex(
-            r'<h1><a[^>]+class=["\']movie-title["\'][^>]*>(.+?)</a></h1>',
-            webpage, 'title')
+        title = self._html_search_regex(r'<h3>([^<]+)</h3>', webpage, 'title')
 
-        video_id = self._search_regex(
-            r'data-movieid=["\'](\d+)', webpage, 'video id', default=video_id)
+        player_params = extract_attributes(self._search_regex(
+            r'(<section[^>]+id="UIVideoPlayer"[^>]+>)', webpage, 'player parameters'))
 
-        m3u8_url = self._download_webpage(
-            'http://cdn.einthusan.com/geturl/%s/hd/London,Washington,Toronto,Dallas,San,Sydney/'
-            % video_id, video_id, headers={'Referer': url})
-        formats = self._extract_m3u8_formats(
-            m3u8_url, video_id, ext='mp4', entry_protocol='m3u8_native')
+        page_id = self._html_search_regex(
+            '<html[^>]+data-pageid="([^"]+)"', webpage, 'page ID')
+        video_data = self._download_json(
+            'https://einthusan.tv/ajax/movie/watch/%s/' % video_id, video_id,
+            data=urlencode_postdata({
+                'xEvent': 'UIVideoPlayer.PingOutcome',
+                'xJson': json.dumps({
+                    'EJOutcomes': player_params['data-ejpingables'],
+                    'NativeHLS': False
+                }),
+                'arcVersion': 3,
+                'appVersion': 59,
+                'gorilla.csrf.Token': page_id,
+            }))['Data']
 
-        description = self._html_search_meta('description', webpage)
+        if isinstance(video_data, compat_str) and video_data.startswith('/ratelimited/'):
+            raise ExtractorError(
+                'Download rate reached. Please try again later.', expected=True)
+
+        ej_links = self._decrypt(video_data['EJLinks'], video_id)
+
+        formats = []
+
+        m3u8_url = ej_links.get('HLSLink')
+        if m3u8_url:
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, video_id, ext='mp4', entry_protocol='m3u8_native'))
+
+        mp4_url = ej_links.get('MP4Link')
+        if mp4_url:
+            formats.append({
+                'url': mp4_url,
+            })
+
+        self._sort_formats(formats)
+
+        description = get_elements_by_class('synopsis', webpage)[0]
         thumbnail = self._html_search_regex(
-            r'''<a class="movie-cover-wrapper".*?><img src=["'](.*?)["'].*?/></a>''',
-            webpage, "thumbnail url", fatal=False)
+            r'''<img[^>]+src=(["'])(?P<url>(?!\1).+?/moviecovers/(?!\1).+?)\1''',
+            webpage, 'thumbnail url', fatal=False, group='url')
         if thumbnail is not None:
-            thumbnail = compat_urlparse.urljoin(url, remove_start(thumbnail, '..'))
+            thumbnail = compat_urlparse.urljoin(url, thumbnail)
 
         return {
             'id': video_id,

youtube_dl/extractor/ellentv.py

@@ -1,13 +1,9 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import json
-
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    NO_DEFAULT,
-)
+from .kaltura import KalturaIE
+from ..utils import NO_DEFAULT
 
 
 class EllenTVIE(InfoExtractor):
@@ -65,7 +61,7 @@ class EllenTVIE(InfoExtractor):
             if partner_id and kaltura_id:
                 break
 
-        return self.url_result('kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura')
+        return self.url_result('kaltura:%s:%s' % (partner_id, kaltura_id), KalturaIE.ie_key())
 
 
 class EllenTVClipsIE(InfoExtractor):
@@ -77,14 +73,14 @@ class EllenTVClipsIE(InfoExtractor):
             'id': 'meryl-streep-vanessa-hudgens',
             'title': 'Meryl Streep, Vanessa Hudgens',
         },
-        'playlist_mincount': 7,
+        'playlist_mincount': 5,
     }
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
 
         webpage = self._download_webpage(url, playlist_id)
-        playlist = self._extract_playlist(webpage)
+        playlist = self._extract_playlist(webpage, playlist_id)
 
         return {
             '_type': 'playlist',
@@ -93,16 +89,13 @@ class EllenTVClipsIE(InfoExtractor):
             'entries': self._extract_entries(playlist)
         }
 
-    def _extract_playlist(self, webpage):
+    def _extract_playlist(self, webpage, playlist_id):
         json_string = self._search_regex(r'playerView.addClips\(\[\{(.*?)\}\]\);', webpage, 'json')
-        try:
-            return json.loads('[{' + json_string + '}]')
-        except ValueError as ve:
-            raise ExtractorError('Failed to download JSON', cause=ve)
+        return self._parse_json('[{' + json_string + '}]', playlist_id)
 
     def _extract_entries(self, playlist):
         return [
             self.url_result(
                 'kaltura:%s:%s' % (item['kaltura_partner_id'], item['kaltura_entry_id']),
-                'Kaltura')
+                KalturaIE.ie_key(), video_id=item['kaltura_entry_id'])
             for item in playlist]

youtube_dl/extractor/elpais.py

@@ -39,6 +39,18 @@ class ElPaisIE(InfoExtractor):
             'description': 'La nave portaba cientos de ánforas y se hundió cerca de la isla de Cabrera por razones desconocidas',
             'upload_date': '20170127',
         },
+    }, {
+        'url': 'http://epv.elpais.com/epv/2017/02/14/programa_la_voz_de_inaki/1487062137_075943.html',
+        'info_dict': {
+            'id': '1487062137_075943',
+            'ext': 'mp4',
+            'title': 'Disyuntivas',
+            'description': 'md5:a0fb1485c4a6a8a917e6f93878e66218',
+            'upload_date': '20170214',
+        },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
@@ -59,14 +71,15 @@ class ElPaisIE(InfoExtractor):
         video_url = prefix + video_suffix
         thumbnail_suffix = self._search_regex(
             r"(?:URLMediaStill|urlFotogramaFijo_\d+)\s*=\s*url_cache\s*\+\s*'([^']+)'",
-            webpage, 'thumbnail URL', fatal=False)
+            webpage, 'thumbnail URL', default=None)
         thumbnail = (
             None if thumbnail_suffix is None
-            else prefix + thumbnail_suffix)
+            else prefix + thumbnail_suffix) or self._og_search_thumbnail(webpage)
         title = self._html_search_regex(
-            (r"tituloVideo\s*=\s*'([^']+)'", webpage, 'title',
-             r'<h2 class="entry-header entry-title.*?>(.*?)</h2>'),
-            webpage, 'title')
+            (r"tituloVideo\s*=\s*'([^']+)'",
+             r'<h2 class="entry-header entry-title.*?>(.*?)</h2>',
+             r'<h1[^>]+class="titulo"[^>]*>([^<]+)'),
+            webpage, 'title', default=None) or self._og_search_title(webpage)
         upload_date = unified_strdate(self._search_regex(
             r'<p class="date-header date-int updated"\s+title="([^"]+)">',
             webpage, 'upload date', default=None) or self._html_search_meta(

youtube_dl/extractor/etonline.py

@@ -0,0 +1,39 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class ETOnlineIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?etonline\.com/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'http://www.etonline.com/tv/211130_dove_cameron_liv_and_maddie_emotional_episode_series_finale/',
+        'info_dict': {
+            'id': '211130_dove_cameron_liv_and_maddie_emotional_episode_series_finale',
+            'title': 'md5:a21ec7d3872ed98335cbd2a046f34ee6',
+            'description': 'md5:8b94484063f463cca709617c79618ccd',
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'http://www.etonline.com/media/video/here_are_the_stars_who_love_bringing_their_moms_as_dates_to_the_oscars-211359/',
+        'only_matching': True,
+    }]
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1242911076001/default_default/index.html?videoId=ref:%s'
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        entries = [
+            self.url_result(
+                self.BRIGHTCOVE_URL_TEMPLATE % video_id, 'BrightcoveNew', video_id)
+            for video_id in re.findall(
+                r'site\.brightcove\s*\([^,]+,\s*["\'](title_\d+)', webpage)]
+
+        return self.playlist_result(
+            entries, playlist_id,
+            self._og_search_title(webpage, fatal=False),
+            self._og_search_description(webpage))

youtube_dl/extractor/extractors.py

@@ -19,6 +19,7 @@ from .acast import (
     ACastChannelIE,
 )
 from .addanime import AddAnimeIE
+from .adn import ADNIE
 from .adobetv import (
     AdobeTVIE,
     AdobeTVShowIE,
@@ -71,6 +72,7 @@ from .arte import (
 )
 from .atresplayer import AtresPlayerIE
 from .atttechchannel import ATTTechChannelIE
+from .atvat import ATVAtIE
 from .audimedia import AudiMediaIE
 from .audioboom import AudioBoomIE
 from .audiomack import AudiomackIE, AudiomackAlbumIE
@@ -83,8 +85,8 @@ from .awaan import (
 from .azmedien import (
     AZMedienIE,
     AZMedienPlaylistIE,
+    AZMedienShowPlaylistIE,
 )
-from .azubu import AzubuIE, AzubuLiveIE
 from .baidu import BaiduVideoIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
@@ -116,6 +118,7 @@ from .bleacherreport import (
 from .blinkx import BlinkxIE
 from .bloomberg import BloombergIE
 from .bokecc import BokeCCIE
+from .bostonglobe import BostonGlobeIE
 from .bpb import BpbIE
 from .br import BRIE
 from .bravotv import BravoTVIE
@@ -161,7 +164,10 @@ from .ccc import CCCIE
 from .ccma import CCMAIE
 from .cctv import CCTVIE
 from .cda import CDAIE
-from .ceskatelevize import CeskaTelevizeIE
+from .ceskatelevize import (
+    CeskaTelevizeIE,
+    CeskaTelevizePoradyIE,
+)
 from .channel9 import Channel9IE
 from .charlierose import CharlieRoseIE
 from .chaturbate import ChaturbateIE
@@ -202,6 +208,7 @@ from .commonprotocols import (
     RtmpIE,
 )
 from .condenast import CondeNastIE
+from .corus import CorusIE
 from .cracked import CrackedIE
 from .crackle import CrackleIE
 from .criterion import CriterionIE
@@ -226,6 +233,10 @@ from .dailymotion import (
     DailymotionUserIE,
     DailymotionCloudIE,
 )
+from .daisuki import (
+    DaisukiIE,
+    DaisukiPlaylistIE,
+)
 from .daum import (
     DaumIE,
     DaumClipIE,
@@ -240,7 +251,10 @@ from .dfb import DFBIE
 from .dhm import DHMIE
 from .dotsub import DotsubIE
 from .douyutv import DouyuTVIE
-from .dplay import DPlayIE
+from .dplay import (
+    DPlayIE,
+    DPlayItIE,
+)
 from .dramafever import (
     DramaFeverIE,
     DramaFeverSeriesIE,
@@ -256,7 +270,12 @@ from .dvtv import DVTVIE
 from .dumpert import DumpertIE
 from .defense import DefenseGouvFrIE
 from .discovery import DiscoveryIE
-from .discoverygo import DiscoveryGoIE
+from .discoverygo import (
+    DiscoveryGoIE,
+    DiscoveryGoPlaylistIE,
+)
+from .discoverynetworks import DiscoveryNetworksDeIE
+from .discoveryvr import DiscoveryVRIE
 from .disney import DisneyIE
 from .dispeak import DigitallySpeakingIE
 from .dropbox import DropboxIE
@@ -287,6 +306,7 @@ from .espn import (
     ESPNArticleIE,
 )
 from .esri import EsriVideoIE
+from .etonline import ETOnlineIE
 from .europa import EuropaIE
 from .everyonesmixtape import EveryonesMixtapeIE
 from .expotv import ExpoTVIE
@@ -337,6 +357,7 @@ from .francetv import (
 )
 from .freesound import FreesoundIE
 from .freespeech import FreespeechIE
+from .freshlive import FreshLiveIE
 from .funimation import FunimationIE
 from .funnyordie import FunnyOrDieIE
 from .fusion import FusionIE
@@ -349,6 +370,7 @@ from .gameone import (
 from .gamersyde import GamersydeIE
 from .gamespot import GameSpotIE
 from .gamestar import GameStarIE
+from .gaskrank import GaskrankIE
 from .gazeta import GazetaIE
 from .gdcvault import GDCVaultIE
 from .generic import GenericIE
@@ -361,6 +383,7 @@ from .globo import (
     GloboArticleIE,
 )
 from .go import GoIE
+from .go90 import Go90IE
 from .godtube import GodTubeIE
 from .godtv import GodTVIE
 from .golem import GolemIE
@@ -380,10 +403,7 @@ from .heise import HeiseIE
 from .hellporno import HellPornoIE
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
-from .hgtv import (
-    HGTVIE,
-    HGTVComShowIE,
-)
+from .hgtv import HGTVComShowIE
 from .historicfilms import HistoricFilmsIE
 from .hitbox import HitboxIE, HitboxLiveIE
 from .hitrecord import HitRecordIE
@@ -521,6 +541,7 @@ from .mangomolo import (
 )
 from .matchtv import MatchTVIE
 from .mdr import MDRIE
+from .medici import MediciIE
 from .meipai import MeipaiIE
 from .melonvod import MelonVODIE
 from .meta import METAIE
@@ -638,6 +659,7 @@ from .ninecninemedia import (
 from .ninegag import NineGagIE
 from .ninenow import NineNowIE
 from .nintendo import NintendoIE
+from .njpwworld import NJPWWorldIE
 from .nobelprize import NobelPrizeIE
 from .noco import NocoIE
 from .normalboots import NormalbootsIE
@@ -667,6 +689,7 @@ from .npo import (
     NPORadioIE,
     NPORadioFragmentIE,
     SchoolTVIE,
+    HetKlokhuisIE,
     VPROIE,
     WNLIE,
 )
@@ -695,6 +718,8 @@ from .ondemandkorea import OnDemandKoreaIE
 from .onet import (
     OnetIE,
     OnetChannelIE,
+    OnetMVPIE,
+    OnetPlIE,
 )
 from .onionstudios import OnionStudiosIE
 from .ooyala import (
@@ -709,6 +734,10 @@ from .orf import (
     ORFFM4IE,
     ORFIPTVIE,
 )
+from .packtpub import (
+    PacktPubIE,
+    PacktPubCourseIE,
+)
 from .pandatv import PandaTVIE
 from .pandoratv import PandoraTVIE
 from .parliamentliveuk import ParliamentLiveUKIE
@@ -778,11 +807,12 @@ from .radiojavan import RadioJavanIE
 from .radiobremen import RadioBremenIE
 from .radiofrance import RadioFranceIE
 from .rai import (
-    RaiTVIE,
+    RaiPlayIE,
     RaiIE,
 )
 from .rbmaradio import RBMARadioIE
 from .rds import RDSIE
+from .redbulltv import RedBullTVIE
 from .redtube import RedTubeIE
 from .regiotv import RegioTVIE
 from .rentv import (
@@ -808,7 +838,11 @@ from .rozhlas import RozhlasIE
 from .rtbf import RTBFIE
 from .rte import RteIE, RteRadioIE
 from .rtlnl import RtlNlIE
-from .rtl2 import RTL2IE
+from .rtl2 import (
+    RTL2IE,
+    RTL2YouIE,
+    RTL2YouSeriesIE,
+)
 from .rtp import RTPIE
 from .rts import RTSIE
 from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE, RTVELiveIE, RTVETelevisionIE
@@ -834,9 +868,9 @@ from .safari import (
 from .sapo import SapoIE
 from .savefrom import SaveFromIE
 from .sbs import SBSIE
-from .scivee import SciVeeIE
 from .screencast import ScreencastIE
 from .screencastomatic import ScreencastOMaticIE
+from .scrippsnetworks import ScrippsNetworksWatchIE
 from .seeker import SeekerIE
 from .senateisvp import SenateISVPIE
 from .sendtonews import SendtoNewsIE
@@ -850,6 +884,7 @@ from .shared import (
 from .showroomlive import ShowRoomLiveIE
 from .sina import SinaIE
 from .sixplay import SixPlayIE
+from .skylinewebcams import SkylineWebcamsIE
 from .skynewsarabia import (
     SkyNewsArabiaIE,
     SkyNewsArabiaArticleIE,
@@ -894,6 +929,7 @@ from .sport5 import Sport5IE
 from .sportbox import SportBoxEmbedIE
 from .sportdeutschland import SportDeutschlandIE
 from .sportschau import SportschauIE
+from .sprout import SproutIE
 from .srgssr import (
     SRGSSRIE,
     SRGSSRPlayIE,
@@ -902,6 +938,7 @@ from .srmediathek import SRMediathekIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .steam import SteamIE
 from .streamable import StreamableIE
+from .streamango import StreamangoIE
 from .streamcloud import StreamcloudIE
 from .streamcz import StreamCZIE
 from .streetvoice import StreetVoiceIE
@@ -948,13 +985,13 @@ from .theplatform import (
 from .thescene import TheSceneIE
 from .thesixtyone import TheSixtyOneIE
 from .thestar import TheStarIE
+from .thesun import TheSunIE
 from .theweatherchannel import TheWeatherChannelIE
 from .thisamericanlife import ThisAmericanLifeIE
 from .thisav import ThisAVIE
 from .thisoldhouse import ThisOldHouseIE
 from .threeqsdn import ThreeQSDNIE
 from .tinypic import TinyPicIE
-from .tlc import TlcDeIE
 from .tmz import (
     TMZIE,
     TMZArticleIE,
@@ -967,6 +1004,7 @@ from .tnaflix import (
 )
 from .toggle import ToggleIE
 from .tonline import TOnlineIE
+from .toongoggles import ToonGogglesIE
 from .toutv import TouTvIE
 from .toypics import ToypicsUserIE, ToypicsIE
 from .traileraddict import TrailerAddictIE
@@ -987,14 +1025,17 @@ from .tunein import (
     TuneInTopicIE,
     TuneInShortenerIE,
 )
+from .tunepk import TunePkIE
 from .turbo import TurboIE
 from .tutv import TutvIE
 from .tv2 import (
     TV2IE,
     TV2ArticleIE,
 )
+from .tv2hu import TV2HuIE
 from .tv3 import TV3IE
 from .tv4 import TV4IE
+from .tv5mondeplus import TV5MondePlusIE
 from .tva import TVAIE
 from .tvanouvelles import (
     TVANouvellesIE,
@@ -1006,6 +1047,7 @@ from .tvc import (
 )
 from .tvigle import TvigleIE
 from .tvland import TVLandIE
+from .tvn24 import TVN24IE
 from .tvnoe import TVNoeIE
 from .tvp import (
     TVPEmbedIE,
@@ -1016,6 +1058,7 @@ from .tvplay import (
     TVPlayIE,
     ViafreeIE,
 )
+from .tvplayer import TVPlayerIE
 from .tweakers import TweakersIE
 from .twentyfourvideo import TwentyFourVideoIE
 from .twentymin import TwentyMinutenIE
@@ -1095,6 +1138,7 @@ from .videomore import (
     VideomoreSeasonIE,
 )
 from .videopremium import VideoPremiumIE
+from .videopress import VideoPressIE
 from .vidio import VidioIE
 from .vidme import (
     VidmeIE,
@@ -1144,11 +1188,19 @@ from .vlive import (
     VLiveChannelIE
 )
 from .vodlocker import VodlockerIE
+from .vodpl import VODPlIE
 from .vodplatform import VODPlatformIE
 from .voicerepublic import VoiceRepublicIE
 from .voxmedia import VoxMediaIE
 from .vporn import VpornIE
 from .vrt import VRTIE
+from .vrak import VrakIE
+from .vrv import (
+    VRVIE,
+    VRVSeriesIE,
+)
+from .vshare import VShareIE
+from .medialaan import MedialaanIE
 from .vube import VubeIE
 from .vuclip import VuClipIE
 from .vvvvid import VVVVIDIE
@@ -1181,7 +1233,10 @@ from .wrzuta import (
     WrzutaIE,
     WrzutaPlaylistIE,
 )
-from .wsj import WSJIE
+from .wsj import (
+    WSJIE,
+    WSJArticleIE,
+)
 from .xbef import XBefIE
 from .xboxclips import XboxClipsIE
 from .xfileshare import XFileShareIE

youtube_dl/extractor/eyedotv.py

@@ -54,7 +54,7 @@ class EyedoTVIE(InfoExtractor):
             'id': video_id,
             'title': title,
             'formats': self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', 'm3u8' if is_live else 'm3u8_native'),
+                m3u8_url, video_id, 'mp4', 'm3u8_native'),
             'description': xpath_text(video_data, _add_ns('Description')),
             'duration': parse_duration(xpath_text(video_data, _add_ns('Duration'))),
             'uploader': xpath_text(video_data, _add_ns('Createur')),

youtube_dl/extractor/facebook.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -134,6 +135,46 @@ class FacebookIE(InfoExtractor):
             'upload_date': '20161030',
             'uploader': 'CNN',
         },
+    }, {
+        # bigPipe.onPageletArrive ... onPageletArrive pagelet_group_mall
+        'url': 'https://www.facebook.com/yaroslav.korpan/videos/1417995061575415/',
+        'info_dict': {
+            'id': '1417995061575415',
+            'ext': 'mp4',
+            'title': 'md5:a7b86ca673f51800cd54687b7f4012fe',
+            'timestamp': 1486648217,
+            'upload_date': '20170209',
+            'uploader': 'Yaroslav Korpan',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.facebook.com/LaGuiaDelVaron/posts/1072691702860471',
+        'info_dict': {
+            'id': '1072691702860471',
+            'ext': 'mp4',
+            'title': 'md5:ae2d22a93fbb12dad20dc393a869739d',
+            'timestamp': 1477305000,
+            'upload_date': '20161024',
+            'uploader': 'La Guía Del Varón',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.facebook.com/groups/1024490957622648/permalink/1396382447100162/',
+        'info_dict': {
+            'id': '1396382447100162',
+            'ext': 'mp4',
+            'title': 'md5:e2d2700afdf84e121f5d0f999bad13a3',
+            'timestamp': 1486035494,
+            'upload_date': '20170202',
+            'uploader': 'Elisabeth Ahtn',
+        },
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
         'only_matching': True,
@@ -155,6 +196,10 @@ class FacebookIE(InfoExtractor):
     }, {
         'url': 'https://www.facebookcorewwwi.onion/video.php?v=274175099429670',
         'only_matching': True,
+    }, {
+        # no title
+        'url': 'https://www.facebook.com/onlycleverentertainment/videos/1947995502095005/',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -249,7 +294,7 @@ class FacebookIE(InfoExtractor):
             for item in instances:
                 if item[1][0] == 'VideoConfig':
                     video_item = item[2][0]
-                    if video_item.get('video_id') == video_id:
+                    if video_item.get('video_id'):
                         return video_item['videoData']
 
         server_js_data = self._parse_json(self._search_regex(
@@ -262,7 +307,7 @@ class FacebookIE(InfoExtractor):
         if not video_data:
             server_js_data = self._parse_json(
                 self._search_regex(
-                    r'bigPipe\.onPageletArrive\(({.+?})\)\s*;\s*}\s*\)\s*,\s*["\']onPageletArrive\s+stream_pagelet',
+                    r'bigPipe\.onPageletArrive\(({.+?})\)\s*;\s*}\s*\)\s*,\s*["\']onPageletArrive\s+(?:stream_pagelet|pagelet_group_mall|permalink_video_pagelet)',
                     webpage, 'js data', default='{}'),
                 video_id, transform_source=js_to_json, fatal=False)
             if server_js_data:
@@ -312,15 +357,15 @@ class FacebookIE(InfoExtractor):
         self._sort_formats(formats)
 
         video_title = self._html_search_regex(
-            r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title',
-            default=None)
+            r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage,
+            'title', default=None)
         if not video_title:
             video_title = self._html_search_regex(
                 r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
                 webpage, 'alternative title', default=None)
         if not video_title:
             video_title = self._html_search_meta(
-                'description', webpage, 'title')
+                'description', webpage, 'title', default=None)
         if video_title:
             video_title = limit_length(video_title, 80)
         else:

youtube_dl/extractor/fox.py

@@ -47,9 +47,12 @@ class FOXIE(AdobePassIE):
             resource = self._get_mvpd_resource('fbc-fox', None, ap_p['videoGUID'], rating)
             query['auth'] = self._extract_mvpd_auth(url, video_id, 'fbc-fox', resource)
 
-        return {
+        info = self._search_json_ld(webpage, video_id, fatal=False)
+        info.update({
             '_type': 'url_transparent',
             'ie_key': 'ThePlatform',
             'url': smuggle_url(update_url_query(release_url, query), {'force_smil_url': True}),
             'id': video_id,
-        }
+        })
+
+        return info

youtube_dl/extractor/franceculture.py

@@ -4,7 +4,8 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
-    unified_strdate,
+    extract_attributes,
+    int_or_none,
 )
 
 
@@ -19,6 +20,7 @@ class FranceCultureIE(InfoExtractor):
             'title': 'Rendez-vous au pays des geeks',
             'thumbnail': r're:^https?://.*\.jpg$',
             'upload_date': '20140301',
+            'timestamp': 1393642916,
             'vcodec': 'none',
         }
     }
@@ -28,30 +30,34 @@ class FranceCultureIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        video_url = self._search_regex(
-            r'(?s)<div[^>]+class="[^"]*?title-zone-diffusion[^"]*?"[^>]*>.*?<button[^>]+data-asset-source="([^"]+)"',
-            webpage, 'video path')
+        video_data = extract_attributes(self._search_regex(
+            r'(?s)<div[^>]+class="[^"]*?(?:title-zone-diffusion|heading-zone-(?:wrapper|player-button))[^"]*?"[^>]*>.*?(<button[^>]+data-asset-source="[^"]+"[^>]+>)',
+            webpage, 'video data'))
 
-        title = self._og_search_title(webpage)
+        video_url = video_data['data-asset-source']
+        title = video_data.get('data-asset-title') or self._og_search_title(webpage)
 
-        upload_date = unified_strdate(self._search_regex(
-            '(?s)<div[^>]+class="date"[^>]*>.*?<span[^>]+class="inner"[^>]*>([^<]+)<',
-            webpage, 'upload date', fatal=False))
+        description = self._html_search_regex(
+            r'(?s)<div[^>]+class="intro"[^>]*>.*?<h2>(.+?)</h2>',
+            webpage, 'description', default=None)
         thumbnail = self._search_regex(
-            r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+data-dejavu-src="([^"]+)"',
+            r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+(?:data-dejavu-)?src="([^"]+)"',
             webpage, 'thumbnail', fatal=False)
         uploader = self._html_search_regex(
-            r'(?s)<div id="emission".*?<span class="author">(.*?)</span>',
+            r'(?s)<span class="author">(.*?)</span>',
             webpage, 'uploader', default=None)
-        vcodec = 'none' if determine_ext(video_url.lower()) == 'mp3' else None
+        ext = determine_ext(video_url.lower())
 
         return {
             'id': display_id,
             'display_id': display_id,
             'url': video_url,
             'title': title,
+            'description': description,
             'thumbnail': thumbnail,
-            'vcodec': vcodec,
+            'ext': ext,
+            'vcodec': 'none' if ext == 'mp3' else None,
             'uploader': uploader,
-            'upload_date': upload_date,
+            'timestamp': int_or_none(video_data.get('data-asset-created-date')),
+            'duration': int_or_none(video_data.get('data-duration')),
         }

youtube_dl/extractor/freshlive.py

@@ -0,0 +1,83 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    try_get,
+    unified_timestamp,
+)
+
+
+class FreshLiveIE(InfoExtractor):
+    _VALID_URL = r'https?://freshlive\.tv/[^/]+/(?P<id>\d+)'
+    _TEST = {
+        'url': 'https://freshlive.tv/satotv/74712',
+        'md5': '9f0cf5516979c4454ce982df3d97f352',
+        'info_dict': {
+            'id': '74712',
+            'ext': 'mp4',
+            'title': 'テスト',
+            'description': 'テスト',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1511,
+            'timestamp': 1483619655,
+            'upload_date': '20170105',
+            'uploader': 'サトTV',
+            'uploader_id': 'satotv',
+            'view_count': int,
+            'comment_count': int,
+            'is_live': False,
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        options = self._parse_json(
+            self._search_regex(
+                r'window\.__CONTEXT__\s*=\s*({.+?});\s*</script>',
+                webpage, 'initial context'),
+            video_id)
+
+        info = options['context']['dispatcher']['stores']['ProgramStore']['programs'][video_id]
+
+        title = info['title']
+
+        if info.get('status') == 'upcoming':
+            raise ExtractorError('Stream %s is upcoming' % video_id, expected=True)
+
+        stream_url = info.get('liveStreamUrl') or info['archiveStreamUrl']
+
+        is_live = info.get('liveStreamUrl') is not None
+
+        formats = self._extract_m3u8_formats(
+            stream_url, video_id, 'mp4',
+            'm3u8_native', m3u8_id='hls')
+
+        if is_live:
+            title = self._live_title(title)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': title,
+            'description': info.get('description'),
+            'thumbnail': info.get('thumbnailUrl'),
+            'duration': int_or_none(info.get('airTime')),
+            'timestamp': unified_timestamp(info.get('createdAt')),
+            'uploader': try_get(
+                info, lambda x: x['channel']['title'], compat_str),
+            'uploader_id': try_get(
+                info, lambda x: x['channel']['code'], compat_str),
+            'uploader_url': try_get(
+                info, lambda x: x['channel']['permalink'], compat_str),
+            'view_count': int_or_none(info.get('viewCount')),
+            'comment_count': int_or_none(info.get('commentCount')),
+            'tags': info.get('tags', []),
+            'is_live': is_live,
+        }

youtube_dl/extractor/funimation.py

@@ -7,9 +7,9 @@ from ..compat import (
     compat_urllib_parse_unquote_plus,
 )
 from ..utils import (
-    clean_html,
     determine_ext,
     int_or_none,
+    js_to_json,
     sanitized_Request,
     ExtractorError,
     urlencode_postdata
@@ -17,34 +17,26 @@ from ..utils import (
 
 
 class FunimationIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?funimation\.com/shows/[^/]+/videos/(?:official|promotional)/(?P<id>[^/?#&]+)'
+    _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/shows/[^/]+/(?P<id>[^/?#&]+)'
 
     _NETRC_MACHINE = 'funimation'
 
     _TESTS = [{
-        'url': 'http://www.funimation.com/shows/air/videos/official/breeze',
+        'url': 'https://www.funimation.com/shows/hacksign/role-play/',
         'info_dict': {
-            'id': '658',
-            'display_id': 'breeze',
-            'ext': 'mp4',
-            'title': 'Air - 1 - Breeze',
-            'description': 'md5:1769f43cd5fc130ace8fd87232207892',
-            'thumbnail': r'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': {
-            'id': '31128',
+            'id': '91144',
             'display_id': 'role-play',
             'ext': 'mp4',
-            'title': '.hack//SIGN - 1 - Role Play',
+            'title': '.hack//SIGN - Role Play',
             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
             'thumbnail': r're:https?://.*\.jpg',
         },
-        'skip': 'Access without user interaction is forbidden by CloudFlare',
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }, {
-        'url': 'http://www.funimation.com/shows/attack-on-titan-junior-high/videos/promotional/broadcast-dub-preview',
+        'url': 'https://www.funimation.com/shows/attack-on-titan-junior-high/broadcast-dub-preview/',
         'info_dict': {
             'id': '9635',
             'display_id': 'broadcast-dub-preview',
@@ -54,25 +46,13 @@ class FunimationIE(InfoExtractor):
             'thumbnail': r're:https?://.*\.(?:jpg|png)',
         },
         'skip': 'Access without user interaction is forbidden by CloudFlare',
+    }, {
+        'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
+        'only_matching': True,
     }]
 
     _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:
@@ -114,119 +94,74 @@ class FunimationIE(InfoExtractor):
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        def _search_kane(name):
+            return self._search_regex(
+                r"KANE_customdimensions\.%s\s*=\s*'([^']+)';" % name,
+                webpage, name, default=None)
+
+        title_data = self._parse_json(self._search_regex(
+            r'TITLE_DATA\s*=\s*({[^}]+})',
+            webpage, 'title data', default=''),
+            display_id, js_to_json, fatal=False) or {}
+
+        video_id = title_data.get('id') or self._search_regex([
+            r"KANE_customdimensions.videoID\s*=\s*'(\d+)';",
+            r'<iframe[^>]+src="/player/(\d+)"',
+        ], webpage, 'video_id', default=None)
+        if not video_id:
+            player_url = self._html_search_meta([
+                'al:web:url',
+                'og:video:url',
+                'og:video:secure_url',
+            ], webpage, fatal=True)
+            video_id = self._search_regex(r'/player/(\d+)', player_url, 'video id')
+
+        title = episode = title_data.get('title') or _search_kane('videoTitle') or self._og_search_title(webpage)
+        series = _search_kane('showName')
+        if series:
+            title = '%s - %s' % (series, title)
+        description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
+
+        try:
+            sources = self._download_json(
+                'https://prod-api-funimationnow.dadcdigital.com/api/source/catalog/video/%s/signed/' % video_id,
+                video_id)['items']
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                error = self._parse_json(e.cause.read(), video_id)['errors'][0]
+                raise ExtractorError('%s said: %s' % (
+                    self.IE_NAME, error.get('detail') or error.get('title')), expected=True)
+            raise
 
-        errors = []
         formats = []
-
-        ERRORS_MAP = {
-            'ERROR_MATURE_CONTENT_LOGGED_IN': 'matureContentLoggedIn',
-            'ERROR_MATURE_CONTENT_LOGGED_OUT': 'matureContentLoggedOut',
-            'ERROR_SUBSCRIPTION_LOGGED_OUT': 'subscriptionLoggedOut',
-            'ERROR_VIDEO_EXPIRED': 'videoExpired',
-            'ERROR_TERRITORY_UNAVAILABLE': 'territoryUnavailable',
-            'SVODBASIC_SUBSCRIPTION_IN_PLAYER': 'basicSubscription',
-            'SVODNON_SUBSCRIPTION_IN_PLAYER': 'nonSubscription',
-            'ERROR_PLAYER_NOT_RESPONDING': 'playerNotResponding',
-            'ERROR_UNABLE_TO_CONNECT_TO_CDN': 'unableToConnectToCDN',
-            'ERROR_STREAM_NOT_FOUND': 'streamNotFound',
-        }
-
-        USER_AGENTS = (
-            # PC UA is served with m3u8 that provides some bonus lower quality formats
-            ('pc', 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'),
-            # Mobile UA allows to extract direct links and also does not fail when
-            # PC UA fails with hulu error (e.g.
-            # http://www.funimation.com/shows/hacksign/videos/official/role-play)
-            ('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 if kind else 'Downloading webpage')
-
-            playlist = self._parse_json(
-                self._search_regex(
-                    r'var\s+playersData\s*=\s*(\[.+?\]);\n',
-                    webpage, 'players data'),
-                display_id)[0]['playlist']
-
-            items = next(item['items'] for item in playlist if item.get('items'))
-            item = next(item for item in items if item.get('itemAK') == display_id)
-
-            error_messages = {}
-            video_error_messages = self._search_regex(
-                r'var\s+videoErrorMessages\s*=\s*({.+?});\n',
-                webpage, 'error messages', default=None)
-            if video_error_messages:
-                error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False)
-                if error_messages_json:
-                    for _, error in error_messages_json.items():
-                        type_ = error.get('type')
-                        description = error.get('description')
-                        content = error.get('content')
-                        if type_ == 'text' and description and content:
-                            error_message = ERRORS_MAP.get(description)
-                            if error_message:
-                                error_messages[error_message] = content
-
-            for video in item.get('videoSet', []):
-                auth_token = video.get('authToken')
-                if not auth_token:
-                    continue
-                funimation_id = video.get('FUNImationID') or video.get('videoId')
-                preference = 1 if video.get('languageMode') == 'dub' else 0
-                if not auth_token.startswith('?'):
-                    auth_token = '?%s' % auth_token
-                for quality, height in (('sd', 480), ('hd', 720), ('hd1080', 1080)):
-                    format_url = video.get('%sUrl' % quality)
-                    if not format_url:
-                        continue
-                    if not format_url.startswith(('http', '//')):
-                        errors.append(format_url)
-                        continue
-                    if determine_ext(format_url) == 'm3u8':
-                        formats.extend(self._extract_m3u8_formats(
-                            format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native',
-                            preference=preference, m3u8_id='%s-hls' % funimation_id, fatal=False))
-                    else:
-                        tbr = int_or_none(self._search_regex(
-                            r'-(\d+)[Kk]', format_url, 'tbr', default=None))
-                        formats.append({
-                            'url': format_url + auth_token,
-                            'format_id': '%s-http-%dp' % (funimation_id, height),
-                            'height': height,
-                            'tbr': tbr,
-                            'preference': preference,
-                        })
-
-        if not formats and errors:
-            raise ExtractorError(
-                '%s returned error: %s'
-                % (self.IE_NAME, clean_html(error_messages.get(errors[0], errors[0]))),
-                expected=True)
-
+        for source in sources:
+            source_url = source.get('src')
+            if not source_url:
+                continue
+            source_type = source.get('videoType') or determine_ext(source_url)
+            if source_type == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    source_url, video_id, 'mp4',
+                    m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'format_id': source_type,
+                    'url': source_url,
+                })
         self._sort_formats(formats)
 
-        title = item['title']
-        artist = item.get('artist')
-        if artist:
-            title = '%s - %s' % (artist, title)
-        description = self._og_search_description(webpage) or item.get('description')
-        thumbnail = self._og_search_thumbnail(webpage) or item.get('posterUrl')
-        video_id = item.get('itemId') or display_id
-
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,
             'description': description,
-            'thumbnail': thumbnail,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'series': series,
+            'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
+            'episode_number': int_or_none(title_data.get('episodeNum')),
+            'episode': episode,
+            'season_id': title_data.get('seriesId'),
             'formats': formats,
         }

youtube_dl/extractor/gaskrank.py

@@ -0,0 +1,123 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    int_or_none,
+    js_to_json,
+    unified_strdate,
+)
+
+
+class GaskrankIE(InfoExtractor):
+    """InfoExtractor for gaskrank.tv"""
+    _VALID_URL = r'https?://(?:www\.)?gaskrank\.tv/tv/(?P<categories>[^/]+)/(?P<id>[^/]+)\.html?'
+    _TESTS = [
+        {
+            'url': 'http://www.gaskrank.tv/tv/motorrad-fun/strike-einparken-durch-anfaenger-crash-mit-groesserem-flurschaden.htm',
+            'md5': '1ae88dbac97887d85ebd1157a95fc4f9',
+            'info_dict': {
+                'id': '201601/26955',
+                'ext': 'mp4',
+                'title': 'Strike! Einparken können nur Männer - Flurschaden hält sich in Grenzen *lol*',
+                'thumbnail': r're:^https?://.*\.jpg$',
+                'categories': ['motorrad-fun'],
+                'display_id': 'strike-einparken-durch-anfaenger-crash-mit-groesserem-flurschaden',
+                'uploader_id': 'Bikefun',
+                'upload_date': '20170110',
+                'uploader_url': None,
+            }
+        },
+        {
+            'url': 'http://www.gaskrank.tv/tv/racing/isle-of-man-tt-2011-michael-du-15920.htm',
+            'md5': 'c33ee32c711bc6c8224bfcbe62b23095',
+            'info_dict': {
+                'id': '201106/15920',
+                'ext': 'mp4',
+                'title': 'Isle of Man - Michael Dunlop vs Guy Martin - schwindelig kucken',
+                'thumbnail': r're:^https?://.*\.jpg$',
+                'categories': ['racing'],
+                'display_id': 'isle-of-man-tt-2011-michael-du-15920',
+                'uploader_id': 'IOM',
+                'upload_date': '20160506',
+                'uploader_url': 'www.iomtt.com',
+            }
+        }
+    ]
+
+    def _real_extract(self, url):
+        """extract information from gaskrank.tv"""
+        def fix_json(code):
+            """Removes trailing comma in json: {{},} --> {{}}"""
+            return re.sub(r',\s*}', r'}', js_to_json(code))
+
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        categories = [re.match(self._VALID_URL, url).group('categories')]
+        title = self._search_regex(
+            r'movieName\s*:\s*\'([^\']*)\'',
+            webpage, 'title')
+        thumbnail = self._search_regex(
+            r'poster\s*:\s*\'([^\']*)\'',
+            webpage, 'thumbnail', default=None)
+
+        mobj = re.search(
+            r'Video von:\s*(?P<uploader_id>[^|]*?)\s*\|\s*vom:\s*(?P<upload_date>[0-9][0-9]\.[0-9][0-9]\.[0-9][0-9][0-9][0-9])',
+            webpage)
+        if mobj is not None:
+            uploader_id = mobj.groupdict().get('uploader_id')
+            upload_date = unified_strdate(mobj.groupdict().get('upload_date'))
+
+        uploader_url = self._search_regex(
+            r'Homepage:\s*<[^>]*>(?P<uploader_url>[^<]*)',
+            webpage, 'uploader_url', default=None)
+        tags = re.findall(
+            r'/tv/tags/[^/]+/"\s*>(?P<tag>[^<]*?)<',
+            webpage)
+
+        view_count = self._search_regex(
+            r'class\s*=\s*"gkRight"(?:[^>]*>\s*<[^>]*)*icon-eye-open(?:[^>]*>\s*<[^>]*)*>\s*(?P<view_count>[0-9\.]*)',
+            webpage, 'view_count', default=None)
+        if view_count:
+            view_count = int_or_none(view_count.replace('.', ''))
+
+        average_rating = self._search_regex(
+            r'itemprop\s*=\s*"ratingValue"[^>]*>\s*(?P<average_rating>[0-9,]+)',
+            webpage, 'average_rating')
+        if average_rating:
+            average_rating = float_or_none(average_rating.replace(',', '.'))
+
+        playlist = self._parse_json(
+            self._search_regex(
+                r'playlist\s*:\s*\[([^\]]*)\]',
+                webpage, 'playlist', default='{}'),
+            display_id, transform_source=fix_json, fatal=False)
+
+        video_id = self._search_regex(
+            r'https?://movies\.gaskrank\.tv/([^-]*?)(-[^\.]*)?\.mp4',
+            playlist.get('0').get('src'), 'video id')
+
+        formats = []
+        for key in playlist:
+            formats.append({
+                'url': playlist[key]['src'],
+                'format_id': key,
+                'quality': playlist[key].get('quality')})
+        self._sort_formats(formats, field_preference=['format_id'])
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'thumbnail': thumbnail,
+            'categories': categories,
+            'display_id': display_id,
+            'uploader_id': uploader_id,
+            'upload_date': upload_date,
+            'uploader_url': uploader_url,
+            'tags': tags,
+            'view_count': view_count,
+            'average_rating': average_rating,
+        }

youtube_dl/extractor/generic.py

@@ -20,6 +20,7 @@ from ..utils import (
     float_or_none,
     HEADRequest,
     is_html,
+    js_to_json,
     orderedSet,
     sanitized_Request,
     smuggle_url,
@@ -29,6 +30,7 @@ from ..utils import (
     UnsupportedError,
     xpath_text,
 )
+from .commonprotocols import RtmpIE
 from .brightcove import (
     BrightcoveLegacyIE,
     BrightcoveNewIE,
@@ -81,6 +83,9 @@ from .videa import VideaIE
 from .twentymin import TwentyMinutenIE
 from .ustream import UstreamIE
 from .openload import OpenloadIE
+from .videopress import VideoPressIE
+from .rutube import RutubeIE
+from .limelight import LimelightBaseIE
 
 
 class GenericIE(InfoExtractor):
@@ -425,6 +430,22 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,  # m3u8 download
             },
         },
+        {
+            # Brightcove video in <iframe>
+            'url': 'http://www.un.org/chinese/News/story.asp?NewsID=27724',
+            'md5': '36d74ef5e37c8b4a2ce92880d208b968',
+            'info_dict': {
+                'id': '5360463607001',
+                'ext': 'mp4',
+                'title': '叙利亚失明儿童在废墟上演唱《心跳》  呼吁获得正常童年生活',
+                'description': '联合国儿童基金会中东和北非区域大使、作曲家扎德·迪拉尼（Zade Dirani）在3月15日叙利亚冲突爆发7周年纪念日之际发布了为叙利亚谱写的歌曲《心跳》（HEARTBEAT），为受到六年冲突影响的叙利亚儿童发出强烈呐喊，呼吁世界做出共同努力，使叙利亚儿童重新获得享有正常童年生活的权利。',
+                'uploader': 'United Nations',
+                'uploader_id': '1362235914001',
+                'timestamp': 1489593889,
+                'upload_date': '20170315',
+            },
+            'add_ie': ['BrightcoveLegacy'],
+        },
         {
             # Brightcove with alternative playerID key
             'url': 'http://www.nature.com/nmeth/journal/v9/n7/fig_tab/nmeth.2062_SV1.html',
@@ -445,6 +466,76 @@ class GenericIE(InfoExtractor):
                 },
             }],
         },
+        {
+            # Brightcove with UUID in videoPlayer
+            'url': 'http://www8.hp.com/cn/zh/home.html',
+            'info_dict': {
+                'id': '5255815316001',
+                'ext': 'mp4',
+                'title': 'Sprocket Video - China',
+                'description': 'Sprocket Video - China',
+                'uploader': 'HP-Video Gallery',
+                'timestamp': 1482263210,
+                'upload_date': '20161220',
+                'uploader_id': '1107601872001',
+            },
+            'params': {
+                'skip_download': True,  # m3u8 download
+            },
+            'skip': 'video rotates...weekly?',
+        },
+        {
+            # Brightcove:new type [2].
+            'url': 'http://www.delawaresportszone.com/video-st-thomas-more-earns-first-trip-to-basketball-semis',
+            'md5': '2b35148fcf48da41c9fb4591650784f3',
+            'info_dict': {
+                'id': '5348741021001',
+                'ext': 'mp4',
+                'upload_date': '20170306',
+                'uploader_id': '4191638492001',
+                'timestamp': 1488769918,
+                'title': 'VIDEO:  St. Thomas More earns first trip to basketball semis',
+
+            },
+        },
+        {
+            # Alternative brightcove <video> attributes
+            'url': 'http://www.programme-tv.net/videos/extraits/81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche/',
+            'info_dict': {
+                'id': '81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche',
+                'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche, Extraits : toutes les vidéos avec Télé-Loisirs",
+            },
+            'playlist': [{
+                'md5': '732d22ba3d33f2f3fc253c39f8f36523',
+                'info_dict': {
+                    'id': '5311302538001',
+                    'ext': 'mp4',
+                    'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche",
+                    'description': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche (France 2, 5 février 2017)",
+                    'timestamp': 1486321708,
+                    'upload_date': '20170205',
+                    'uploader_id': '800000640001',
+                },
+                'only_matching': True,
+            }],
+        },
+        {
+            # Brightcove with UUID in videoPlayer
+            'url': 'http://www8.hp.com/cn/zh/home.html',
+            'info_dict': {
+                'id': '5255815316001',
+                'ext': 'mp4',
+                'title': 'Sprocket Video - China',
+                'description': 'Sprocket Video - China',
+                'uploader': 'HP-Video Gallery',
+                'timestamp': 1482263210,
+                'upload_date': '20161220',
+                'uploader_id': '1107601872001',
+            },
+            'params': {
+                'skip_download': True,  # m3u8 download
+            },
+        },
         # ooyala video
         {
             'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
@@ -709,6 +800,21 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             }
         },
+        # YouTube <object> embed
+        {
+            'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
+            'md5': '516718101ec834f74318df76259fb3cc',
+            'info_dict': {
+                'id': 'msN87y-iEx0',
+                'ext': 'webm',
+                'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
+                'upload_date': '20080526',
+                'description': 'md5:0ffc78ea3f01b2e2c247d5f8d1d3c18d',
+                'uploader': 'Christopher Sykes',
+                'uploader_id': 'ChristopherJSykes',
+            },
+            'add_ie': ['Youtube'],
+        },
         # Camtasia studio
         {
             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
@@ -881,12 +987,13 @@ class GenericIE(InfoExtractor):
         },
         # LazyYT
         {
-            'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
+            'url': 'https://skiplagged.com/',
             'info_dict': {
-                'id': '1986',
-                'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
+                'id': 'skiplagged',
+                'title': 'Skiplagged: The smart way to find cheap flights',
             },
-            'playlist_mincount': 2,
+            'playlist_mincount': 1,
+            'add_ie': ['Youtube'],
         },
         # Cinchcast embed
         {
@@ -946,6 +1053,43 @@ class GenericIE(InfoExtractor):
                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
             },
         },
+        # jwplayer rtmp
+        {
+            'url': 'http://www.suffolk.edu/sjc/',
+            'info_dict': {
+                'id': 'sjclive',
+                'ext': 'flv',
+                'title': 'Massachusetts Supreme Judicial Court Oral Arguments',
+                'uploader': 'www.suffolk.edu',
+            },
+            'params': {
+                'skip_download': True,
+            }
+        },
+        # Complex jwplayer
+        {
+            'url': 'http://www.indiedb.com/games/king-machine/videos',
+            'info_dict': {
+                'id': 'videos',
+                'ext': 'mp4',
+                'title': 'king machine trailer 1',
+                'thumbnail': r're:^https?://.*\.jpg$',
+            },
+        },
+        {
+            # JWPlayer config passed as variable
+            'url': 'http://www.txxx.com/videos/3326530/ariele/',
+            'info_dict': {
+                'id': '3326530_hq',
+                'ext': 'mp4',
+                'title': 'ARIELE | Tube Cup',
+                'uploader': 'www.txxx.com',
+                'age_limit': 18,
+            },
+            'params': {
+                'skip_download': True,
+            }
+        },
         # rtl.nl embed
         {
             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
@@ -976,19 +1120,6 @@ class GenericIE(InfoExtractor):
                 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
             },
         },
-        # Kaltura embed protected with referrer
-        {
-            'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero',
-            'info_dict': {
-                'id': '1_g4fbemnq',
-                'ext': 'mp4',
-                'title': 'Violetta - Achter De Schermen - Ruggero',
-                'description': 'Achter de schermen met Ruggero',
-                'timestamp': 1435133761,
-                'upload_date': '20150624',
-                'uploader_id': 'echojecka',
-            },
-        },
         # Kaltura embed with single quotes
         {
             'url': 'http://fod.infobase.com/p_ViewPlaylist.aspx?AssignmentID=NUN8ZY',
@@ -1034,6 +1165,21 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Kaltura'],
         },
+        {
+            # Kaltura iframe embed
+            'url': 'http://www.gsd.harvard.edu/event/i-m-pei-a-centennial-celebration/',
+            'md5': 'ae5ace8eb09dc1a35d03b579a9c2cc44',
+            'info_dict': {
+                'id': '0_f2cfbpwy',
+                'ext': 'mp4',
+                'title': 'I. M. Pei: A Centennial Celebration',
+                'description': 'md5:1db8f40c69edc46ca180ba30c567f37c',
+                'upload_date': '20170403',
+                'uploader_id': 'batchUser',
+                'timestamp': 1491232186,
+            },
+            'add_ie': ['Kaltura'],
+        },
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
@@ -1473,7 +1619,64 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
             'add_ie': [TwentyMinutenIE.ie_key()],
-        }
+        },
+        {
+            # VideoPress embed
+            'url': 'https://en.support.wordpress.com/videopress/',
+            'info_dict': {
+                'id': 'OcobLTqC',
+                'ext': 'm4v',
+                'title': 'IMG_5786',
+                'timestamp': 1435711927,
+                'upload_date': '20150701',
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'add_ie': [VideoPressIE.ie_key()],
+        },
+        {
+            # Rutube embed
+            'url': 'http://magazzino.friday.ru/videos/vipuski/kazan-2',
+            'info_dict': {
+                'id': '9b3d5bee0a8740bf70dfd29d3ea43541',
+                'ext': 'flv',
+                'title': 'Магаззино: Казань 2',
+                'description': 'md5:99bccdfac2269f0e8fdbc4bbc9db184a',
+                'uploader': 'Магаззино',
+                'upload_date': '20170228',
+                'uploader_id': '996642',
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'add_ie': [RutubeIE.ie_key()],
+        },
+        {
+            # ThePlatform embedded with whitespaces in URLs
+            'url': 'http://www.golfchannel.com/topics/shows/golftalkcentral.htm',
+            'only_matching': True,
+        },
+        {
+            # Senate ISVP iframe https
+            'url': 'https://www.hsgac.senate.gov/hearings/canadas-fast-track-refugee-plan-unanswered-questions-and-implications-for-us-national-security',
+            'md5': 'fb8c70b0b515e5037981a2492099aab8',
+            'info_dict': {
+                'id': 'govtaff020316',
+                'ext': 'mp4',
+                'title': 'Integrated Senate Video Player',
+            },
+            'add_ie': [SenateISVPIE.ie_key()],
+        },
+        {
+            # Limelight embeds (1 channel embed + 4 media embeds)
+            'url': 'http://www.sedona.com/FacilitatorTraining2017',
+            'info_dict': {
+                'id': 'FacilitatorTraining2017',
+                'title': 'Facilitator Training 2017',
+            },
+            'playlist_mincount': 5,
+        },
         # {
         #     # TODO: find another test
         #     # http://schema.org/VideoObject
@@ -1516,7 +1719,7 @@ class GenericIE(InfoExtractor):
                 continue
 
             entries.append({
-                '_type': 'url',
+                '_type': 'url_transparent',
                 'url': next_url,
                 'title': it.find('title').text,
             })
@@ -1773,18 +1976,9 @@ class GenericIE(InfoExtractor):
         video_description = self._og_search_description(webpage, default=None)
         video_thumbnail = self._og_search_thumbnail(webpage, default=None)
 
-        # Helper method
-        def _playlist_from_matches(matches, getter=None, ie=None):
-            urlrs = orderedSet(
-                self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
-                for m in matches)
-            return self.playlist_result(
-                urlrs, playlist_id=video_id, playlist_title=video_title)
-
         # Look for Brightcove Legacy Studio embeds
         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
         if bc_urls:
-            self.to_screen('Brightcove video detected.')
             entries = [{
                 '_type': 'url',
                 'url': smuggle_url(bc_url, {'Referer': url}),
@@ -1799,30 +1993,30 @@ class GenericIE(InfoExtractor):
             }
 
         # Look for Brightcove New Studio embeds
-        bc_urls = BrightcoveNewIE._extract_urls(webpage)
+        bc_urls = BrightcoveNewIE._extract_urls(self, webpage)
         if bc_urls:
-            return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
+            return self.playlist_from_matches(bc_urls, video_id, video_title, ie='BrightcoveNew')
 
         # Look for ThePlatform embeds
         tp_urls = ThePlatformIE._extract_urls(webpage)
         if tp_urls:
-            return _playlist_from_matches(tp_urls, ie='ThePlatform')
+            return self.playlist_from_matches(tp_urls, video_id, video_title, ie='ThePlatform')
 
         # Look for Vessel embeds
         vessel_urls = VesselIE._extract_urls(webpage)
         if vessel_urls:
-            return _playlist_from_matches(vessel_urls, ie=VesselIE.ie_key())
+            return self.playlist_from_matches(vessel_urls, video_id, video_title, ie=VesselIE.ie_key())
 
         # Look for embedded rtl.nl player
         matches = re.findall(
             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
             webpage)
         if matches:
-            return _playlist_from_matches(matches, ie='RtlNl')
+            return self.playlist_from_matches(matches, video_id, video_title, ie='RtlNl')
 
         vimeo_urls = VimeoIE._extract_urls(url, webpage)
         if vimeo_urls:
-            return _playlist_from_matches(vimeo_urls, ie=VimeoIE.ie_key())
+            return self.playlist_from_matches(vimeo_urls, video_id, video_title, ie=VimeoIE.ie_key())
 
         vid_me_embed_url = self._search_regex(
             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
@@ -1837,6 +2031,7 @@ class GenericIE(InfoExtractor):
                 data-video-url=|
                 <embed[^>]+?src=|
                 embedSWF\(?:\s*|
+                <object[^>]+data=|
                 new\s+SWFObject\(
             )
             (["\'])
@@ -1844,25 +2039,25 @@ class GenericIE(InfoExtractor):
                 (?:embed|v|p)/.+?)
             \1''', webpage)
         if matches:
-            return _playlist_from_matches(
-                matches, lambda m: unescapeHTML(m[1]))
+            return self.playlist_from_matches(
+                matches, video_id, video_title, lambda m: unescapeHTML(m[1]))
 
         # Look for lazyYT YouTube embed
         matches = re.findall(
             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
         if matches:
-            return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
+            return self.playlist_from_matches(matches, video_id, video_title, lambda m: unescapeHTML(m))
 
         # Look for Wordpress "YouTube Video Importer" plugin
         matches = re.findall(r'''(?x)<div[^>]+
             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
         if matches:
-            return _playlist_from_matches(matches, lambda m: m[-1])
+            return self.playlist_from_matches(matches, video_id, video_title, lambda m: m[-1])
 
         matches = DailymotionIE._extract_urls(webpage)
         if matches:
-            return _playlist_from_matches(matches)
+            return self.playlist_from_matches(matches, video_id, video_title)
 
         # Look for embedded Dailymotion playlist player (#3822)
         m = re.search(
@@ -1871,8 +2066,8 @@ class GenericIE(InfoExtractor):
             playlists = re.findall(
                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
             if playlists:
-                return _playlist_from_matches(
-                    playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
+                return self.playlist_from_matches(
+                    playlists, video_id, video_title, lambda p: '//dailymotion.com/playlist/%s' % p)
 
         # Look for embedded Wistia player
         match = re.search(
@@ -1979,8 +2174,9 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
             if embeds:
-                return _playlist_from_matches(
-                    embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
+                return self.playlist_from_matches(
+                    embeds, video_id, video_title,
+                    getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
 
         # Look for Aparat videos
         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
@@ -2042,13 +2238,13 @@ class GenericIE(InfoExtractor):
         # Look for funnyordie embed
         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
         if matches:
-            return _playlist_from_matches(
-                matches, getter=unescapeHTML, ie='FunnyOrDie')
+            return self.playlist_from_matches(
+                matches, video_id, video_title, getter=unescapeHTML, ie='FunnyOrDie')
 
         # Look for BBC iPlayer embed
         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
         if matches:
-            return _playlist_from_matches(matches, ie='BBCCoUk')
+            return self.playlist_from_matches(matches, video_id, video_title, ie='BBCCoUk')
 
         # Look for embedded RUTV player
         rutv_url = RUTVIE._extract_url(webpage)
@@ -2063,32 +2259,32 @@ class GenericIE(InfoExtractor):
         # Look for embedded SportBox player
         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
         if sportbox_urls:
-            return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
+            return self.playlist_from_matches(sportbox_urls, video_id, video_title, ie='SportBoxEmbed')
 
         # Look for embedded XHamster player
         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
         if xhamster_urls:
-            return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
+            return self.playlist_from_matches(xhamster_urls, video_id, video_title, ie='XHamsterEmbed')
 
         # Look for embedded TNAFlixNetwork player
         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
         if tnaflix_urls:
-            return _playlist_from_matches(tnaflix_urls, ie=TNAFlixNetworkEmbedIE.ie_key())
+            return self.playlist_from_matches(tnaflix_urls, video_id, video_title, ie=TNAFlixNetworkEmbedIE.ie_key())
 
         # Look for embedded PornHub player
         pornhub_urls = PornHubIE._extract_urls(webpage)
         if pornhub_urls:
-            return _playlist_from_matches(pornhub_urls, ie=PornHubIE.ie_key())
+            return self.playlist_from_matches(pornhub_urls, video_id, video_title, ie=PornHubIE.ie_key())
 
         # Look for embedded DrTuber player
         drtuber_urls = DrTuberIE._extract_urls(webpage)
         if drtuber_urls:
-            return _playlist_from_matches(drtuber_urls, ie=DrTuberIE.ie_key())
+            return self.playlist_from_matches(drtuber_urls, video_id, video_title, ie=DrTuberIE.ie_key())
 
         # Look for embedded RedTube player
         redtube_urls = RedTubeIE._extract_urls(webpage)
         if redtube_urls:
-            return _playlist_from_matches(redtube_urls, ie=RedTubeIE.ie_key())
+            return self.playlist_from_matches(redtube_urls, video_id, video_title, ie=RedTubeIE.ie_key())
 
         # Look for embedded Tvigle player
         mobj = re.search(
@@ -2134,12 +2330,12 @@ class GenericIE(InfoExtractor):
         # Look for embedded soundcloud player
         soundcloud_urls = SoundcloudIE._extract_urls(webpage)
         if soundcloud_urls:
-            return _playlist_from_matches(soundcloud_urls, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
+            return self.playlist_from_matches(soundcloud_urls, video_id, video_title, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
 
         # Look for tunein player
         tunein_urls = TuneInBaseIE._extract_urls(webpage)
         if tunein_urls:
-            return _playlist_from_matches(tunein_urls)
+            return self.playlist_from_matches(tunein_urls, video_id, video_title)
 
         # Look for embedded mtvservices player
         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
@@ -2313,6 +2509,11 @@ class GenericIE(InfoExtractor):
             return self.url_result(piksel_url, PikselIE.ie_key())
 
         # Look for Limelight embeds
+        limelight_urls = LimelightBaseIE._extract_urls(webpage, url)
+        if limelight_urls:
+            return self.playlist_result(
+                limelight_urls, video_id, video_title, video_description)
+
         mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
         if mobj:
             lm = {
@@ -2320,8 +2521,9 @@ class GenericIE(InfoExtractor):
                 'Channel': 'channel',
                 'ChannelList': 'channel_list',
             }
-            return self.url_result('limelight:%s:%s' % (
-                lm[mobj.group(1)], mobj.group(2)), 'Limelight%s' % mobj.group(1), mobj.group(2))
+            return self.url_result(smuggle_url('limelight:%s:%s' % (
+                lm[mobj.group(1)], mobj.group(2)), {'source_url': url}),
+                'Limelight%s' % mobj.group(1), mobj.group(2))
 
         mobj = re.search(
             r'''(?sx)
@@ -2331,7 +2533,9 @@ class GenericIE(InfoExtractor):
                         value=(["\'])(?:(?!\3).)*mediaId=(?P<id>[a-z0-9]{32})
             ''', webpage)
         if mobj:
-            return self.url_result('limelight:media:%s' % mobj.group('id'))
+            return self.url_result(smuggle_url(
+                'limelight:media:%s' % mobj.group('id'),
+                {'source_url': url}), 'LimelightMedia', mobj.group('id'))
 
         # Look for AdobeTVVideo embeds
         mobj = re.search(
@@ -2419,24 +2623,36 @@ class GenericIE(InfoExtractor):
         # Look for DBTV embeds
         dbtv_urls = DBTVIE._extract_urls(webpage)
         if dbtv_urls:
-            return _playlist_from_matches(dbtv_urls, ie=DBTVIE.ie_key())
+            return self.playlist_from_matches(dbtv_urls, video_id, video_title, ie=DBTVIE.ie_key())
 
         # Look for Videa embeds
         videa_urls = VideaIE._extract_urls(webpage)
         if videa_urls:
-            return _playlist_from_matches(videa_urls, ie=VideaIE.ie_key())
+            return self.playlist_from_matches(videa_urls, video_id, video_title, ie=VideaIE.ie_key())
 
         # Look for 20 minuten embeds
         twentymin_urls = TwentyMinutenIE._extract_urls(webpage)
         if twentymin_urls:
-            return _playlist_from_matches(
-                twentymin_urls, ie=TwentyMinutenIE.ie_key())
+            return self.playlist_from_matches(
+                twentymin_urls, video_id, video_title, ie=TwentyMinutenIE.ie_key())
 
         # Look for Openload embeds
         openload_urls = OpenloadIE._extract_urls(webpage)
         if openload_urls:
-            return _playlist_from_matches(
-                openload_urls, ie=OpenloadIE.ie_key())
+            return self.playlist_from_matches(
+                openload_urls, video_id, video_title, ie=OpenloadIE.ie_key())
+
+        # Look for VideoPress embeds
+        videopress_urls = VideoPressIE._extract_urls(webpage)
+        if videopress_urls:
+            return self.playlist_from_matches(
+                videopress_urls, video_id, video_title, ie=VideoPressIE.ie_key())
+
+        # Look for Rutube embeds
+        rutube_urls = RutubeIE._extract_urls(webpage)
+        if rutube_urls:
+            return self.playlist_from_matches(
+                rutube_urls, ie=RutubeIE.ie_key())
 
         # Looking for http://schema.org/VideoObject
         json_ld = self._search_json_ld(
@@ -2462,12 +2678,23 @@ class GenericIE(InfoExtractor):
                 self._sort_formats(entry['formats'])
             return self.playlist_result(entries)
 
+        jwplayer_data = self._find_jwplayer_data(
+            webpage, video_id, transform_source=js_to_json)
+        if jwplayer_data:
+            info = self._parse_jwplayer_data(
+                jwplayer_data, video_id, require_title=False, base_url=url)
+            if not info.get('title'):
+                info['title'] = video_title
+            return info
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True
+            if RtmpIE.suitable(vurl):
+                return True
             vpath = compat_urlparse.urlparse(vurl).path
             vext = determine_ext(vpath)
-            return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml', 'js')
+            return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml', 'js', 'xml')
 
         def filter_video(urls):
             return list(filter(check_video, urls))
@@ -2533,11 +2760,14 @@ class GenericIE(InfoExtractor):
                     found = re.search(REDIRECT_REGEX, refresh_header)
             if found:
                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
-                self.report_following_redirect(new_url)
-                return {
-                    '_type': 'url',
-                    'url': new_url,
-                }
+                if new_url != url:
+                    self.report_following_redirect(new_url)
+                    return {
+                        '_type': 'url',
+                        'url': new_url,
+                    }
+                else:
+                    found = None
 
         if not found:
             # twitter:player is a https URL to iframe player that may or may not
@@ -2572,6 +2802,15 @@ class GenericIE(InfoExtractor):
                 'age_limit': age_limit,
             }
 
+            if RtmpIE.suitable(video_url):
+                entry_info_dict.update({
+                    '_type': 'url_transparent',
+                    'ie_key': RtmpIE.ie_key(),
+                    'url': video_url,
+                })
+                entries.append(entry_info_dict)
+                continue
+
             ext = determine_ext(video_url)
             if ext == 'smil':
                 entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)

youtube_dl/extractor/go.py

@@ -3,7 +3,7 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .adobepass import AdobePassIE
 from ..utils import (
     int_or_none,
     determine_ext,
@@ -13,15 +13,30 @@ from ..utils import (
 )
 
 
-class GoIE(InfoExtractor):
-    _BRANDS = {
-        'abc': '001',
-        'freeform': '002',
-        'watchdisneychannel': '004',
-        'watchdisneyjunior': '008',
-        'watchdisneyxd': '009',
+class GoIE(AdobePassIE):
+    _SITE_INFO = {
+        'abc': {
+            'brand': '001',
+            'requestor_id': 'ABC',
+        },
+        'freeform': {
+            'brand': '002',
+            'requestor_id': 'ABCFamily',
+        },
+        'watchdisneychannel': {
+            'brand': '004',
+            'requestor_id': 'Disney',
+        },
+        'watchdisneyjunior': {
+            'brand': '008',
+            'requestor_id': 'DisneyJunior',
+        },
+        'watchdisneyxd': {
+            'brand': '009',
+            'requestor_id': 'DisneyXD',
+        }
     }
-    _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/(?:[^/]+/)*(?:vdka(?P<id>\w+)|season-\d+/\d+-(?P<display_id>[^/?#]+))' % '|'.join(_BRANDS.keys())
+    _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/(?:[^/]+/)*(?:vdka(?P<id>\w+)|(?:[^/]+/)*(?P<display_id>[^/?#]+))' % '|'.join(_SITE_INFO.keys())
     _TESTS = [{
         'url': 'http://abc.go.com/shows/castle/video/most-recent/vdka0_g86w5onx',
         'info_dict': {
@@ -37,6 +52,12 @@ class GoIE(InfoExtractor):
     }, {
         'url': 'http://abc.go.com/shows/after-paradise/video/most-recent/vdka3335601',
         'only_matching': True,
+    }, {
+        'url': 'http://abc.go.com/shows/the-catch/episode-guide/season-01/10-the-wedding',
+        'only_matching': True,
+    }, {
+        'url': 'http://abc.go.com/shows/world-news-tonight/episode-guide/2017-02/17-021717-intense-stand-off-between-man-with-rifle-and-police-in-oakland',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -47,7 +68,8 @@ class GoIE(InfoExtractor):
                 # There may be inner quotes, e.g. data-video-id="'VDKA3609139'"
                 # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood
                 r'data-video-id=["\']*VDKA(\w+)', webpage, 'video id')
-        brand = self._BRANDS[sub_domain]
+        site_info = self._SITE_INFO[sub_domain]
+        brand = site_info['brand']
         video_data = self._download_json(
             'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (brand, video_id),
             video_id)['video'][0]
@@ -62,28 +84,60 @@ class GoIE(InfoExtractor):
             ext = determine_ext(asset_url)
             if ext == 'm3u8':
                 video_type = video_data.get('type')
-                if video_type == 'lf':
-                    entitlement = self._download_json(
-                        'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
-                        video_id, data=urlencode_postdata({
-                            'video_id': video_data['id'],
-                            'video_type': video_type,
-                            'brand': brand,
-                            'device': '001',
-                        }))
-                    errors = entitlement.get('errors', {}).get('errors', [])
-                    if errors:
-                        error_message = ', '.join([error['message'] for error in errors])
-                        raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
-                    asset_url += '?' + entitlement['uplynkData']['sessionKey']
+                data = {
+                    'video_id': video_data['id'],
+                    'video_type': video_type,
+                    'brand': brand,
+                    'device': '001',
+                }
+                if video_data.get('accesslevel') == '1':
+                    requestor_id = site_info['requestor_id']
+                    resource = self._get_mvpd_resource(
+                        requestor_id, title, video_id, None)
+                    auth = self._extract_mvpd_auth(
+                        url, video_id, requestor_id, resource)
+                    data.update({
+                        'token': auth,
+                        'token_type': 'ap',
+                        'adobe_requestor_id': requestor_id,
+                    })
+                else:
+                    self._initialize_geo_bypass(['US'])
+                entitlement = self._download_json(
+                    'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
+                    video_id, data=urlencode_postdata(data), headers=self.geo_verification_headers())
+                errors = entitlement.get('errors', {}).get('errors', [])
+                if errors:
+                    for error in errors:
+                        if error.get('code') == 1002:
+                            self.raise_geo_restricted(
+                                error['message'], countries=['US'])
+                    error_message = ', '.join([error['message'] for error in errors])
+                    raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
+                asset_url += '?' + entitlement['uplynkData']['sessionKey']
                 formats.extend(self._extract_m3u8_formats(
                     asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False))
             else:
-                formats.append({
+                f = {
                     'format_id': format_id,
                     'url': asset_url,
                     'ext': ext,
-                })
+                }
+                if re.search(r'(?:/mp4/source/|_source\.mp4)', asset_url):
+                    f.update({
+                        'format_id': ('%s-' % format_id if format_id else '') + 'SOURCE',
+                        'preference': 1,
+                    })
+                else:
+                    mobj = re.search(r'/(\d+)x(\d+)/', asset_url)
+                    if mobj:
+                        height = int(mobj.group(2))
+                        f.update({
+                            'format_id': ('%s-' % format_id if format_id else '') + '%dP' % height,
+                            'width': int(mobj.group(1)),
+                            'height': height,
+                        })
+                formats.append(f)
         self._sort_formats(formats)
 
         subtitles = {}

youtube_dl/extractor/go90.py

@@ -0,0 +1,126 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    parse_iso8601,
+)
+
+
+class Go90IE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?go90\.com/videos/(?P<id>[0-9a-zA-Z]+)'
+    _TEST = {
+        'url': 'https://www.go90.com/videos/84BUqjLpf9D',
+        'md5': 'efa7670dbbbf21a7b07b360652b24a32',
+        'info_dict': {
+            'id': '84BUqjLpf9D',
+            'ext': 'mp4',
+            'title': 'Daily VICE - Inside The Utah Coalition Against Pornography Convention',
+            'description': 'VICE\'s Karley Sciortino meets with activists who discuss the state\'s strong anti-porn stance. Then, VICE Sports explains NFL contracts.',
+            'timestamp': 1491868800,
+            'upload_date': '20170411',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        video_data = self._download_json(
+            'https://www.go90.com/api/view/items/' + video_id,
+            video_id, headers={
+                'Content-Type': 'application/json; charset=utf-8',
+            }, data=b'{"client":"web","device_type":"pc"}')
+        main_video_asset = video_data['main_video_asset']
+
+        episode_number = int_or_none(video_data.get('episode_number'))
+        series = None
+        season = None
+        season_id = None
+        season_number = None
+        for metadata in video_data.get('__children', {}).get('Item', {}).values():
+            if metadata.get('type') == 'show':
+                series = metadata.get('title')
+            elif metadata.get('type') == 'season':
+                season = metadata.get('title')
+                season_id = metadata.get('id')
+                season_number = int_or_none(metadata.get('season_number'))
+
+        title = episode = video_data.get('title') or series
+        if series and series != title:
+            title = '%s - %s' % (series, title)
+
+        thumbnails = []
+        formats = []
+        subtitles = {}
+        for asset in video_data.get('assets'):
+            if asset.get('id') == main_video_asset:
+                for source in asset.get('sources', []):
+                    source_location = source.get('location')
+                    if not source_location:
+                        continue
+                    source_type = source.get('type')
+                    if source_type == 'hls':
+                        m3u8_formats = self._extract_m3u8_formats(
+                            source_location, video_id, 'mp4',
+                            'm3u8_native', m3u8_id='hls', fatal=False)
+                        for f in m3u8_formats:
+                            mobj = re.search(r'/hls-(\d+)-(\d+)K', f['url'])
+                            if mobj:
+                                height, tbr = mobj.groups()
+                                height = int_or_none(height)
+                                f.update({
+                                    'height': f.get('height') or height,
+                                    'width': f.get('width') or int_or_none(height / 9.0 * 16.0 if height else None),
+                                    'tbr': f.get('tbr') or int_or_none(tbr),
+                                })
+                        formats.extend(m3u8_formats)
+                    elif source_type == 'dash':
+                        formats.extend(self._extract_mpd_formats(
+                            source_location, video_id, mpd_id='dash', fatal=False))
+                    else:
+                        formats.append({
+                            'format_id': source.get('name'),
+                            'url': source_location,
+                            'width': int_or_none(source.get('width')),
+                            'height': int_or_none(source.get('height')),
+                            'tbr': int_or_none(source.get('bitrate')),
+                        })
+
+                for caption in asset.get('caption_metadata', []):
+                    caption_url = caption.get('source_url')
+                    if not caption_url:
+                        continue
+                    subtitles.setdefault(caption.get('language', 'en'), []).append({
+                        'url': caption_url,
+                        'ext': determine_ext(caption_url, 'vtt'),
+                    })
+            elif asset.get('type') == 'image':
+                asset_location = asset.get('location')
+                if not asset_location:
+                    continue
+                thumbnails.append({
+                    'url': asset_location,
+                    'width': int_or_none(asset.get('width')),
+                    'height': int_or_none(asset.get('height')),
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'thumbnails': thumbnails,
+            'description': video_data.get('short_description'),
+            'like_count': int_or_none(video_data.get('like_count')),
+            'timestamp': parse_iso8601(video_data.get('released_at')),
+            'series': series,
+            'episode': episode,
+            'season': season,
+            'season_id': season_id,
+            'season_number': season_number,
+            'episode_number': episode_number,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/googledrive.py

@@ -6,6 +6,7 @@ from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     int_or_none,
+    lowercase_escape,
 )
 
 
@@ -13,12 +14,12 @@ class GoogleDriveIE(InfoExtractor):
     _VALID_URL = r'https?://(?:(?:docs|drive)\.google\.com/(?:uc\?.*?id=|file/d/)|video\.google\.com/get_player\?.*?docid=)(?P<id>[a-zA-Z0-9_-]{28,})'
     _TESTS = [{
         'url': 'https://drive.google.com/file/d/0ByeS4oOUV-49Zzh4R1J6R09zazQ/edit?pli=1',
-        'md5': '881f7700aec4f538571fa1e0eed4a7b6',
+        'md5': 'd109872761f7e7ecf353fa108c0dbe1e',
         'info_dict': {
             'id': '0ByeS4oOUV-49Zzh4R1J6R09zazQ',
             'ext': 'mp4',
             'title': 'Big Buck Bunny.mp4',
-            'duration': 46,
+            'duration': 45,
         }
     }, {
         # video id is longer than 28 characters
@@ -55,7 +56,7 @@ class GoogleDriveIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(
-            'http://docs.google.com/file/d/%s' % video_id, video_id, encoding='unicode_escape')
+            'http://docs.google.com/file/d/%s' % video_id, video_id)
 
         reason = self._search_regex(r'"reason"\s*,\s*"([^"]+)', webpage, 'reason', default=None)
         if reason:
@@ -74,7 +75,7 @@ class GoogleDriveIE(InfoExtractor):
             resolution = fmt.split('/')[1]
             width, height = resolution.split('x')
             formats.append({
-                'url': fmt_url,
+                'url': lowercase_escape(fmt_url),
                 'format_id': fmt_id,
                 'resolution': resolution,
                 'width': int_or_none(width),

youtube_dl/extractor/hbo.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     xpath_text,
     xpath_element,
@@ -14,14 +15,26 @@ from ..utils import (
 
 class HBOBaseIE(InfoExtractor):
     _FORMATS_INFO = {
+        'pro7': {
+            'width': 1280,
+            'height': 720,
+        },
         '1920': {
             'width': 1280,
             'height': 720,
         },
+        'pro6': {
+            'width': 768,
+            'height': 432,
+        },
         '640': {
             'width': 768,
             'height': 432,
         },
+        'pro5': {
+            'width': 640,
+            'height': 360,
+        },
         'highwifi': {
             'width': 640,
             'height': 360,
@@ -78,6 +91,19 @@ class HBOBaseIE(InfoExtractor):
                     formats.extend(self._extract_m3u8_formats(
                         video_url.replace('.tar', '/base_index_w8.m3u8'),
                         video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                elif source.tag == 'hls':
+                    m3u8_formats = self._extract_m3u8_formats(
+                        video_url.replace('.tar', '/base_index.m3u8'),
+                        video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
+                    for f in m3u8_formats:
+                        if f.get('vcodec') == 'none' and not f.get('tbr'):
+                            f['tbr'] = int_or_none(self._search_regex(
+                                r'-(\d+)k/', f['url'], 'tbr', default=None))
+                    formats.extend(m3u8_formats)
+                elif source.tag == 'dash':
+                    formats.extend(self._extract_mpd_formats(
+                        video_url.replace('.tar', '/manifest.mpd'),
+                        video_id, mpd_id='dash', fatal=False))
                 else:
                     format_info = self._FORMATS_INFO.get(source.tag, {})
                     formats.append({
@@ -86,7 +112,7 @@ class HBOBaseIE(InfoExtractor):
                         'width': format_info.get('width'),
                         'height': format_info.get('height'),
                     })
-        self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
+        self._sort_formats(formats)
 
         thumbnails = []
         card_sizes = xpath_element(video_data, 'titleCardSizes')
@@ -112,10 +138,11 @@ class HBOBaseIE(InfoExtractor):
 
 
 class HBOIE(HBOBaseIE):
+    IE_NAME = 'hbo'
     _VALID_URL = r'https?://(?:www\.)?hbo\.com/video/video\.html\?.*vid=(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.hbo.com/video/video.html?autoplay=true&g=u&vid=1437839',
-        'md5': '1c33253f0c7782142c993c0ba62a8753',
+        'md5': '2c6a6bc1222c7e91cb3334dad1746e5a',
         'info_dict': {
             'id': '1437839',
             'ext': 'mp4',
@@ -131,11 +158,12 @@ class HBOIE(HBOBaseIE):
 
 
 class HBOEpisodeIE(HBOBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?!video)([^/]+/)+video/(?P<id>[0-9a-z-]+)\.html'
+    IE_NAME = 'hbo:episode'
+    _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?P<path>(?!video)(?:(?:[^/]+/)+video|watch-free-episodes)/(?P<id>[0-9a-z-]+))(?:\.html)?'
 
     _TESTS = [{
         'url': 'http://www.hbo.com/girls/episodes/5/52-i-love-you-baby/video/ep-52-inside-the-episode.html?autoplay=true',
-        'md5': '689132b253cc0ab7434237fc3a293210',
+        'md5': '61ead79b9c0dfa8d3d4b07ef4ac556fb',
         'info_dict': {
             'id': '1439518',
             'display_id': 'ep-52-inside-the-episode',
@@ -147,16 +175,19 @@ class HBOEpisodeIE(HBOBaseIE):
     }, {
         'url': 'http://www.hbo.com/game-of-thrones/about/video/season-5-invitation-to-the-set.html?autoplay=true',
         'only_matching': True,
+    }, {
+        'url': 'http://www.hbo.com/watch-free-episodes/last-week-tonight-with-john-oliver',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
+        path, display_id = re.match(self._VALID_URL, url).groups()
 
-        webpage = self._download_webpage(url, display_id)
+        content = self._download_json(
+            'http://www.hbo.com/api/content/' + path, display_id)['content']
 
-        video_id = self._search_regex(
-            r'(?P<q1>[\'"])videoId(?P=q1)\s*:\s*(?P<q2>[\'"])(?P<video_id>\d+)(?P=q2)',
-            webpage, 'video ID', group='video_id')
+        video_id = compat_str((content.get('parsed', {}).get(
+            'common:FullBleedVideo', {}) or content['selectedEpisode'])['videoId'])
 
         info_dict = self._extract_from_id(video_id)
         info_dict['display_id'] = display_id