diff --git a/.github/ISSUE_TEMPLATE/1_broken_site.md b/.github/ISSUE_TEMPLATE/1_broken_site.md index 7a2b16827..d82ff9111 100644 --- a/.github/ISSUE_TEMPLATE/1_broken_site.md +++ b/.github/ISSUE_TEMPLATE/1_broken_site.md @@ -18,7 +18,7 @@ title: '' - [ ] I'm reporting a broken site support -- [ ] I've verified that I'm running youtube-dl version **2019.06.21** +- [ ] I've verified that I'm running youtube-dl version **2020.03.08** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped - [ ] I've searched the bugtracker for similar issues including closed ones @@ -41,7 +41,7 @@ Add the `-v` flag to your command line you run youtube-dl with (`youtube-dl -v < [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 2019.06.21 + [debug] youtube-dl version 2020.03.08 [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: {} diff --git a/.github/ISSUE_TEMPLATE/2_site_support_request.md b/.github/ISSUE_TEMPLATE/2_site_support_request.md index d6180e672..04b350f76 100644 --- a/.github/ISSUE_TEMPLATE/2_site_support_request.md +++ b/.github/ISSUE_TEMPLATE/2_site_support_request.md @@ -19,7 +19,7 @@ labels: 'site-support-request' - [ ] I'm reporting a new site support request -- [ ] I've verified that I'm running youtube-dl version **2019.06.21** +- [ ] I've verified that I'm running youtube-dl version **2020.03.08** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that none of provided URLs violate any copyrights - [ ] I've searched the bugtracker for similar site support requests including closed ones diff --git a/.github/ISSUE_TEMPLATE/3_site_feature_request.md b/.github/ISSUE_TEMPLATE/3_site_feature_request.md index 7cb981abf..6f17ad7bc 100644 --- a/.github/ISSUE_TEMPLATE/3_site_feature_request.md +++ b/.github/ISSUE_TEMPLATE/3_site_feature_request.md @@ -18,13 +18,13 @@ title: '' - [ ] I'm reporting a site feature request -- [ ] I've verified that I'm running youtube-dl version **2019.06.21** +- [ ] I've verified that I'm running youtube-dl version **2020.03.08** - [ ] I've searched the bugtracker for similar site feature requests including closed ones diff --git a/.github/ISSUE_TEMPLATE/4_bug_report.md b/.github/ISSUE_TEMPLATE/4_bug_report.md index 802fa2313..efb179ea5 100644 --- a/.github/ISSUE_TEMPLATE/4_bug_report.md +++ b/.github/ISSUE_TEMPLATE/4_bug_report.md @@ -18,7 +18,7 @@ title: '' - [ ] I'm reporting a broken site support issue -- [ ] I've verified that I'm running youtube-dl version **2019.06.21** +- [ ] I've verified that I'm running youtube-dl version **2020.03.08** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped - [ ] I've searched the bugtracker for similar bug reports including closed ones @@ -43,7 +43,7 @@ Add the `-v` flag to your command line you run youtube-dl with (`youtube-dl -v < [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 2019.06.21 + [debug] youtube-dl version 2020.03.08 [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: {} diff --git a/.github/ISSUE_TEMPLATE/5_feature_request.md b/.github/ISSUE_TEMPLATE/5_feature_request.md index 5153864a1..cf4874bcc 100644 --- a/.github/ISSUE_TEMPLATE/5_feature_request.md +++ b/.github/ISSUE_TEMPLATE/5_feature_request.md @@ -19,13 +19,13 @@ labels: 'request' - [ ] I'm reporting a feature request -- [ ] I've verified that I'm running youtube-dl version **2019.06.21** +- [ ] I've verified that I'm running youtube-dl version **2020.03.08** - [ ] I've searched the bugtracker for similar feature requests including closed ones diff --git a/.travis.yml b/.travis.yml index 6d16c2955..51afd469a 100644 --- a/.travis.yml +++ b/.travis.yml @@ -13,7 +13,7 @@ dist: trusty env: - YTDL_TEST_SET=core - YTDL_TEST_SET=download -matrix: +jobs: include: - python: 3.7 dist: xenial @@ -21,6 +21,12 @@ matrix: - python: 3.7 dist: xenial env: YTDL_TEST_SET=download + - python: 3.8 + dist: xenial + env: YTDL_TEST_SET=core + - python: 3.8 + dist: xenial + env: YTDL_TEST_SET=download - python: 3.8-dev dist: xenial env: YTDL_TEST_SET=core @@ -29,6 +35,11 @@ matrix: env: YTDL_TEST_SET=download - env: JYTHON=true; YTDL_TEST_SET=core - env: JYTHON=true; YTDL_TEST_SET=download + - name: flake8 + python: 3.8 + dist: xenial + install: pip install flake8 + script: flake8 . fast_finish: true allow_failures: - env: YTDL_TEST_SET=download diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index cd9ccbe96..ac759ddc4 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -339,6 +339,72 @@ Incorrect: 'PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4' ``` +### Inline values + +Extracting variables is acceptable for reducing code duplication and improving readability of complex expressions. However, you should avoid extracting variables used only once and moving them to opposite parts of the extractor file, which makes reading the linear flow difficult. + +#### Example + +Correct: + +```python +title = self._html_search_regex(r'([^<]+)', webpage, 'title') +``` + +Incorrect: + +```python +TITLE_RE = r'([^<]+)' +# ...some lines of code... +title = self._html_search_regex(TITLE_RE, webpage, 'title') +``` + +### Collapse fallbacks + +Multiple fallback values can quickly become unwieldy. Collapse multiple fallback values into a single expression via a list of patterns. + +#### Example + +Good: + +```python +description = self._html_search_meta( + ['og:description', 'description', 'twitter:description'], + webpage, 'description', default=None) +``` + +Unwieldy: + +```python +description = ( + self._og_search_description(webpage, default=None) + or self._html_search_meta('description', webpage, default=None) + or self._html_search_meta('twitter:description', webpage, default=None)) +``` + +Methods supporting list of patterns are: `_search_regex`, `_html_search_regex`, `_og_search_property`, `_html_search_meta`. + +### Trailing parentheses + +Always move trailing parentheses after the last argument. + +#### Example + +Correct: + +```python + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list) +``` + +Incorrect: + +```python + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list, +) +``` + ### Use convenience conversion and parsing functions Wrap all extracted numeric data into safe functions from [`youtube_dl/utils.py`](https://github.com/ytdl-org/youtube-dl/blob/master/youtube_dl/utils.py): `int_or_none`, `float_or_none`. Use them for string to number conversions as well. diff --git a/ChangeLog b/ChangeLog index 2d9988da3..84b43c642 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,655 @@ +version 2020.03.08 + +Core ++ [utils] Add support for cookie files with spaces + +Extractors ++ [pornhub] Add support for pornhubpremium.com (#24288) +- [youtube] Remove outdated code and unnecessary requests +* [youtube] Improve extraction in 429 HTTP error conditions (#24283) +* [nhk] Update API version (#24270) + + +version 2020.03.06 + +Extractors +* [youtube] Fix age-gated videos support without login (#24248) +* [vimeo] Fix showcase password protected video extraction (#24224) +* [pornhub] Improve title extraction (#24184) +* [peertube] Improve extraction (#23657) ++ [servus] Add support for new URL schema (#23475, #23583, #24142) +* [vimeo] Fix subtitles URLs (#24209) + + +version 2020.03.01 + +Core +* [YoutubeDL] Force redirect URL to unicode on python 2 +- [options] Remove duplicate short option -v for --version (#24162) + +Extractors +* [xhamster] Fix extraction (#24205) +* [franceculture] Fix extraction (#24204) ++ [telecinco] Add support for article opening videos +* [telecinco] Fix extraction (#24195) +* [xtube] Fix metadata extraction (#21073, #22455) +* [youjizz] Fix extraction (#24181) +- Remove no longer needed compat_str around geturl +* [pornhd] Fix extraction (#24128) ++ [teachable] Add support for multiple videos per lecture (#24101) ++ [wistia] Add support for multiple generic embeds (#8347, 11385) +* [imdb] Fix extraction (#23443) +* [tv2dk:bornholm:play] Fix extraction (#24076) + + +version 2020.02.16 + +Core +* [YoutubeDL] Fix playlist entry indexing with --playlist-items (#10591, + #10622) +* [update] Fix updating via symlinks (#23991) ++ [compat] Introduce compat_realpath (#23991) + +Extractors ++ [npr] Add support for streams (#24042) ++ [24video] Add support for porn.24video.net (#23779, #23784) +- [jpopsuki] Remove extractor (#23858) +* [nova] Improve extraction (#23690) +* [nova:embed] Improve (#23690) +* [nova:embed] Fix extraction (#23672) ++ [abc:iview] Add support for 720p (#22907, #22921) +* [nytimes] Improve format sorting (#24010) ++ [toggle] Add support for mewatch.sg (#23895, #23930) +* [thisoldhouse] Fix extraction (#23951) ++ [popcorntimes] Add support for popcorntimes.tv (#23949) +* [sportdeutschland] Update to new API +* [twitch:stream] Lowercase channel id for stream request (#23917) +* [tv5mondeplus] Fix extraction (#23907, #23911) +* [tva] Relax URL regular expression (#23903) +* [vimeo] Fix album extraction (#23864) +* [viewlift] Improve extraction + * Fix extraction (#23851) + + Add support for authentication + + Add support for more domains +* [svt] Fix series extraction (#22297) +* [svt] Fix article extraction (#22897, #22919) +* [soundcloud] Imporve private playlist/set tracks extraction (#3707) + + +version 2020.01.24 + +Extractors +* [youtube] Fix sigfunc name extraction (#23819) +* [stretchinternet] Fix extraction (#4319) +* [voicerepublic] Fix extraction +* [azmedien] Fix extraction (#23783) +* [businessinsider] Fix jwplatform id extraction (#22929, #22954) ++ [24video] Add support for 24video.vip (#23753) +* [ivi:compilation] Fix entries extraction (#23770) +* [ard] Improve extraction (#23761) + * Simplify extraction + + Extract age limit and series + * Bypass geo-restriction ++ [nbc] Add support for nbc multi network URLs (#23049) +* [americastestkitchen] Fix extraction +* [zype] Improve extraction + + Extract subtitles (#21258) + + Support URLs with alternative keys/tokens (#21258) + + Extract more metadata +* [orf:tvthek] Improve geo restricted videos detection (#23741) +* [soundcloud] Restore previews extraction (#23739) + + +version 2020.01.15 + +Extractors +* [yourporn] Fix extraction (#21645, #22255, #23459) ++ [canvas] Add support for new API endpoint (#17680, #18629) +* [ndr:base:embed] Improve thumbnails extraction (#23731) ++ [vodplatform] Add support for embed.kwikmotion.com domain ++ [twitter] Add support for promo_video_website cards (#23711) +* [orf:radio] Clean description and improve extraction +* [orf:fm4] Fix extraction (#23599) +* [safari] Fix kaltura session extraction (#23679, #23670) +* [lego] Fix extraction and extract subtitle (#23687) +* [cloudflarestream] Improve extraction + + Add support for bytehighway.net domain + + Add support for signed URLs + + Extract thumbnail +* [naver] Improve extraction + * Improve geo-restriction handling + + Extract automatic captions + + Extract uploader metadata + + Extract VLive HLS formats + * Improve metadata extraction +- [pandatv] Remove extractor (#23630) +* [dctp] Fix format extraction (#23656) ++ [scrippsnetworks] Add support for www.discovery.com videos +* [discovery] Fix anonymous token extraction (#23650) +* [nrktv:seriebase] Fix extraction (#23625, #23537) +* [wistia] Improve format extraction and extract subtitles (#22590) +* [vice] Improve extraction (#23631) +* [redtube] Detect private videos (#23518) + + +version 2020.01.01 + +Extractors +* [brightcove] Invalidate policy key cache on failing requests +* [pornhub] Improve locked videos detection (#22449, #22780) ++ [pornhub] Add support for m3u8 formats +* [pornhub] Fix extraction (#22749, #23082) +* [brightcove] Update policy key on failing requests +* [spankbang] Improve removed video detection (#23423) +* [spankbang] Fix extraction (#23307, #23423, #23444) +* [soundcloud] Automatically update client id on failing requests +* [prosiebensat1] Improve geo restriction handling (#23571) +* [brightcove] Cache brightcove player policy keys +* [teachable] Fail with error message if no video URL found +* [teachable] Improve locked lessons detection (#23528) ++ [scrippsnetworks] Add support for Scripps Networks sites (#19857, #22981) +* [mitele] Fix extraction (#21354, #23456) +* [soundcloud] Update client id (#23516) +* [mailru] Relax URL regular expressions (#23509) + + +version 2019.12.25 + +Core +* [utils] Improve str_to_int ++ [downloader/hls] Add ability to override AES decryption key URL (#17521) + +Extractors +* [mediaset] Fix parse formats (#23508) ++ [tv2dk:bornholm:play] Add support for play.tv2bornholm.dk (#23291) ++ [slideslive] Add support for url and vimeo service names (#23414) +* [slideslive] Fix extraction (#23413) +* [twitch:clips] Fix extraction (#23375) ++ [soundcloud] Add support for token protected embeds (#18954) +* [vk] Improve extraction + * Fix User Videos extraction (#23356) + * Extract all videos for lists with more than 1000 videos (#23356) + + Add support for video albums (#14327, #14492) +- [kontrtube] Remove extractor +- [videopremium] Remove extractor +- [musicplayon] Remove extractor (#9225) ++ [ufctv] Add support for ufcfightpass.imgdge.com and + ufcfightpass.imggaming.com (#23343) ++ [twitch] Extract m3u8 formats frame rate (#23333) ++ [imggaming] Add support for playlists and extract subtitles ++ [ufcarabia] Add support for UFC Arabia (#23312) +* [ufctv] Fix extraction +* [yahoo] Fix gyao brightcove player id (#23303) +* [vzaar] Override AES decryption key URL (#17521) ++ [vzaar] Add support for AES HLS manifests (#17521, #23299) +* [nrl] Fix extraction +* [teachingchannel] Fix extraction +* [nintendo] Fix extraction and partially add support for Nintendo Direct + videos (#4592) ++ [ooyala] Add better fallback values for domain and streams variables ++ [youtube] Add support youtubekids.com (#23272) +* [tv2] Detect DRM protection ++ [tv2] Add support for katsomo.fi and mtv.fi (#10543) +* [tv2] Fix tv2.no article extraction +* [msn] Improve extraction + + Add support for YouTube and NBCSports embeds + + Add support for articles with multiple videos + * Improve AOL embed support + * Improve format extraction +* [abcotvs] Relax URL regular expression and improve metadata extraction + (#18014) +* [channel9] Reduce response size +* [adobetv] Improve extaction + * Use OnDemandPagedList for list extractors + * Reduce show extraction requests + * Extract original video format and subtitles + + Add support for adobe tv embeds + + +version 2019.11.28 + +Core ++ [utils] Add generic caesar cipher and rot47 +* [utils] Handle rd-suffixed day parts in unified_strdate (#23199) + +Extractors +* [vimeo] Improve extraction + * Fix review extraction + * Fix ondemand extraction + * Make password protected player case as an expected error (#22896) + * Simplify channel based extractors code +- [openload] Remove extractor (#11999) +- [verystream] Remove extractor +- [streamango] Remove extractor (#15406) +* [dailymotion] Improve extraction + * Extract http formats included in m3u8 manifest + * Fix user extraction (#3553, #21415) + + Add suport for User Authentication (#11491) + * Fix password protected videos extraction (#23176) + * Respect age limit option and family filter cookie value (#18437) + * Handle video url playlist query param + * Report allowed countries for geo-restricted videos +* [corus] Improve extraction + + Add support for Series Plus, W Network, YTV, ABC Spark, disneychannel.com + and disneylachaine.ca (#20861) + + Add support for self hosted videos (#22075) + * Detect DRM protection (#14910, #9164) +* [vivo] Fix extraction (#22328, #22279) ++ [bitchute] Extract upload date (#22990, #23193) +* [soundcloud] Update client id (#23214) + + +version 2019.11.22 + +Core ++ [extractor/common] Clean jwplayer description HTML tags ++ [extractor/common] Add data, headers and query to all major extract formats + methods + +Extractors +* [chaturbate] Fix extraction (#23010, #23012) ++ [ntvru] Add support for non relative file URLs (#23140) +* [vk] Fix wall audio thumbnails extraction (#23135) +* [ivi] Fix format extraction (#21991) +- [comcarcoff] Remove extractor ++ [drtv] Add support for new URL schema (#23059) ++ [nexx] Add support for Multi Player JS Setup (#23052) ++ [teamcoco] Add support for new videos (#23054) +* [soundcloud] Check if the soundtrack has downloads left (#23045) +* [facebook] Fix posts video data extraction (#22473) +- [addanime] Remove extractor +- [minhateca] Remove extractor +- [daisuki] Remove extractor +* [seeker] Fix extraction +- [revision3] Remove extractors +* [twitch] Fix video comments URL (#18593, #15828) +* [twitter] Improve extraction + + Add support for generic embeds (#22168) + * Always extract http formats for native videos (#14934) + + Add support for Twitter Broadcasts (#21369) + + Extract more metadata + * Improve VMap format extraction + * Unify extraction code for both twitter statuses and cards ++ [twitch] Add support for Clip embed URLs +* [lnkgo] Fix extraction (#16834) +* [mixcloud] Improve extraction + * Improve metadata extraction (#11721) + * Fix playlist extraction (#22378) + * Fix user mixes extraction (#15197, #17865) ++ [kinja] Add support for Kinja embeds (#5756, #11282, #22237, #22384) +* [onionstudios] Fix extraction ++ [hotstar] Pass Referer header to format requests (#22836) +* [dplay] Minimize response size ++ [patreon] Extract uploader_id and filesize +* [patreon] Minimize response size +* [roosterteeth] Fix login request (#16094, #22689) + + +version 2019.11.05 + +Extractors ++ [scte] Add support for learning.scte.org (#22975) ++ [msn] Add support for Vidible and AOL embeds (#22195, #22227) +* [myspass] Fix video URL extraction and improve metadata extraction (#22448) +* [jamendo] Improve extraction + * Fix album extraction (#18564) + * Improve metadata extraction (#18565, #21379) +* [mediaset] Relax URL guid matching (#18352) ++ [mediaset] Extract unprotected M3U and MPD manifests (#17204) +* [telegraaf] Fix extraction ++ [bellmedia] Add support for marilyn.ca videos (#22193) +* [stv] Fix extraction (#22928) +- [iconosquare] Remove extractor +- [keek] Remove extractor +- [gameone] Remove extractor (#21778) +- [flipagram] Remove extractor +- [bambuser] Remove extractor +* [wistia] Reduce embed extraction false positives ++ [wistia] Add support for inline embeds (#22931) +- [go90] Remove extractor +* [kakao] Remove raw request ++ [kakao] Extract format total bitrate +* [daum] Fix VOD and Clip extracton (#15015) +* [kakao] Improve extraction + + Add support for embed URLs + + Add support for Kakao Legacy vid based embed URLs + * Only extract fields used for extraction + * Strip description and extract tags +* [mixcloud] Fix cloudcast data extraction (#22821) +* [yahoo] Improve extraction + + Add support for live streams (#3597, #3779, #22178) + * Bypass cookie consent page for european domains (#16948, #22576) + + Add generic support for embeds (#20332) +* [tv2] Fix and improve extraction (#22787) ++ [tv2dk] Add support for TV2 DK sites +* [onet] Improve extraction … + + Add support for onet100.vod.pl + + Extract m3u8 formats + * Correct audio only format info +* [fox9] Fix extraction + + +version 2019.10.29 + +Core +* [utils] Actualize major IPv4 address blocks per country + +Extractors ++ [go] Add support for abc.com and freeform.com (#22823, #22864) ++ [mtv] Add support for mtvjapan.com +* [mtv] Fix extraction for mtv.de (#22113) +* [videodetective] Fix extraction +* [internetvideoarchive] Fix extraction +* [nbcnews] Fix extraction (#12569, #12576, #21703, #21923) +- [hark] Remove extractor +- [tutv] Remove extractor +- [learnr] Remove extractor +- [macgamestore] Remove extractor +* [la7] Update Kaltura service URL (#22358) +* [thesun] Fix extraction (#16966) +- [makertv] Remove extractor ++ [tenplay] Add support for 10play.com.au (#21446) +* [soundcloud] Improve extraction + * Improve format extraction (#22123) + + Extract uploader_id and uploader_url (#21916) + + Extract all known thumbnails (#19071, #20659) + * Fix extration for private playlists (#20976) + + Add support for playlist embeds (#20976) + * Skip preview formats (#22806) +* [dplay] Improve extraction + + Add support for dplay.fi, dplay.jp and es.dplay.com (#16969) + * Fix it.dplay.com extraction (#22826) + + Extract creator, tags and thumbnails + * Handle playback API call errors ++ [discoverynetworks] Add support for dplay.co.uk +* [vk] Improve extraction + + Add support for Odnoklassniki embeds + + Extract more videos from user lists (#4470) + + Fix wall post audio extraction (#18332) + * Improve error detection (#22568) ++ [odnoklassniki] Add support for embeds +* [puhutv] Improve extraction + * Fix subtitles extraction + * Transform HLS URLs to HTTP URLs + * Improve metadata extraction +* [ceskatelevize] Skip DRM media ++ [facebook] Extract subtitles (#22777) +* [globo] Handle alternative hash signing method + + +version 2019.10.22 + +Core +* [utils] Improve subtitles_filename (#22753) + +Extractors +* [facebook] Bypass download rate limits (#21018) ++ [contv] Add support for contv.com +- [viewster] Remove extractor +* [xfileshare] Improve extractor (#17032, #17906, #18237, #18239) + * Update the list of domains + + Add support for aa-encoded video data + * Improve jwplayer format extraction + + Add support for Clappr sources +* [mangomolo] Fix video format extraction and add support for player URLs +* [audioboom] Improve metadata extraction +* [twitch] Update VOD URL matching (#22395, #22727) +- [mit] Remove support for video.mit.edu (#22403) +- [servingsys] Remove extractor (#22639) +* [dumpert] Fix extraction (#22428, #22564) +* [atresplayer] Fix extraction (#16277, #16716) + + +version 2019.10.16 + +Core +* [extractor/common] Make _is_valid_url more relaxed + +Extractors +* [vimeo] Improve album videos id extraction (#22599) ++ [globo] Extract subtitles (#22713) +* [bokecc] Improve player params extraction (#22638) +* [nexx] Handle result list (#22666) +* [vimeo] Fix VHX embed extraction +* [nbc] Switch to graphql API (#18581, #22693, #22701) +- [vessel] Remove extractor +- [promptfile] Remove extractor (#6239) +* [kaltura] Fix service URL extraction (#22658) +* [kaltura] Fix embed info strip (#22658) +* [globo] Fix format extraction (#20319) +* [redtube] Improve metadata extraction (#22492, #22615) +* [pornhub:uservideos:upload] Fix extraction (#22619) ++ [telequebec:squat] Add support for squat.telequebec.tv (#18503) +- [wimp] Remove extractor (#22088, #22091) ++ [gfycat] Extend URL regular expression (#22225) ++ [chaturbate] Extend URL regular expression (#22309) +* [peertube] Update instances (#22414) ++ [telequebec] Add support for coucou.telequebec.tv (#22482) ++ [xvideos] Extend URL regular expression (#22471) +- [youtube] Remove support for invidious.enkirton.net (#22543) ++ [openload] Add support for oload.monster (#22592) +* [nrktv:seriebase] Fix extraction (#22596) ++ [youtube] Add support for yt.lelux.fi (#22597) +* [orf:tvthek] Make manifest requests non fatal (#22578) +* [teachable] Skip login when already logged in (#22572) +* [viewlift] Improve extraction (#22545) +* [nonktube] Fix extraction (#22544) + + +version 2019.09.28 + +Core +* [YoutubeDL] Honour all --get-* options with --flat-playlist (#22493) + +Extractors +* [vk] Fix extraction (#22522) +* [heise] Fix kaltura embeds extraction (#22514) +* [ted] Check for resources validity and extract subtitled downloads (#22513) ++ [youtube] Add support for + owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya.b32.i2p (#22292) ++ [nhk] Add support for clips +* [nhk] Fix video extraction (#22249, #22353) +* [byutv] Fix extraction (#22070) ++ [openload] Add support for oload.online (#22304) ++ [youtube] Add support for invidious.drycat.fr (#22451) +* [jwplatfom] Do not match video URLs (#20596, #22148) +* [youtube:playlist] Unescape playlist uploader (#22483) ++ [bilibili] Add support audio albums and songs (#21094) ++ [instagram] Add support for tv URLs ++ [mixcloud] Allow uppercase letters in format URLs (#19280) +* [brightcove] Delegate all supported legacy URLs to new extractor (#11523, + #12842, #13912, #15669, #16303) +* [hotstar] Use native HLS downloader by default ++ [hotstar] Extract more formats (#22323) +* [9now] Fix extraction (#22361) +* [zdf] Bypass geo restriction ++ [tv4] Extract series metadata +* [tv4] Fix extraction (#22443) + + +version 2019.09.12.1 + +Extractors +* [youtube] Remove quality and tbr for itag 43 (#22372) + + +version 2019.09.12 + +Extractors +* [youtube] Quick extraction tempfix (#22367, #22163) + + +version 2019.09.01 + +Core ++ [extractor/generic] Add support for squarespace embeds (#21294, #21802, + #21859) ++ [downloader/external] Respect mtime option for aria2c (#22242) + +Extractors ++ [xhamster:user] Add support for user pages (#16330, #18454) ++ [xhamster] Add support for more domains ++ [verystream] Add support for woof.tube (#22217) ++ [dailymotion] Add support for lequipe.fr (#21328, #22152) ++ [openload] Add support for oload.vip (#22205) ++ [bbccouk] Extend URL regular expression (#19200) ++ [youtube] Add support for invidious.nixnet.xyz and yt.elukerio.org (#22223) +* [safari] Fix authentication (#22161, #22184) +* [usanetwork] Fix extraction (#22105) ++ [einthusan] Add support for einthusan.ca (#22171) +* [youtube] Improve unavailable message extraction (#22117) ++ [piksel] Extract subtitles (#20506) + + +version 2019.08.13 + +Core +* [downloader/fragment] Fix ETA calculation of resumed download (#21992) +* [YoutubeDL] Check annotations availability (#18582) + +Extractors +* [youtube:playlist] Improve flat extraction (#21927) +* [youtube] Fix annotations extraction (#22045) ++ [discovery] Extract series meta field (#21808) +* [youtube] Improve error detection (#16445) +* [vimeo] Fix album extraction (#1933, #15704, #15855, #18967, #21986) ++ [roosterteeth] Add support for watch URLs +* [discovery] Limit video data by show slug (#21980) + + +version 2019.08.02 + +Extractors ++ [tvigle] Add support for HLS and DASH formats (#21967) +* [tvigle] Fix extraction (#21967) ++ [yandexvideo] Add support for DASH formats (#21971) +* [discovery] Use API call for video data extraction (#21808) ++ [mgtv] Extract format_note (#21881) +* [tvn24] Fix metadata extraction (#21833, #21834) +* [dlive] Relax URL regular expression (#21909) ++ [openload] Add support for oload.best (#21913) +* [youtube] Improve metadata extraction for age gate content (#21943) + + +version 2019.07.30 + +Extractors +* [youtube] Fix and improve title and description extraction (#21934) + + +version 2019.07.27 + +Extractors ++ [yahoo:japannews] Add support for yahoo.co.jp (#21698, #21265) ++ [discovery] Add support go.discovery.com URLs +* [youtube:playlist] Relax video regular expression (#21844) +* [generic] Restrict --default-search schemeless URLs detection pattern + (#21842) +* [vrv] Fix CMS signing query extraction (#21809) + + +version 2019.07.16 + +Extractors ++ [asiancrush] Add support for yuyutv.com, midnightpulp.com and cocoro.tv + (#21281, #21290) +* [kaltura] Check source format URL (#21290) +* [ctsnews] Fix YouTube embeds extraction (#21678) ++ [einthusan] Add support for einthusan.com (#21748, #21775) ++ [youtube] Add support for invidious.mastodon.host (#21777) ++ [gfycat] Extend URL regular expression (#21779, #21780) +* [youtube] Restrict is_live extraction (#21782) + + +version 2019.07.14 + +Extractors +* [porn91] Fix extraction (#21312) ++ [yandexmusic] Extract track number and disk number (#21421) ++ [yandexmusic] Add support for multi disk albums (#21420, #21421) +* [lynda] Handle missing subtitles (#20490, #20513) ++ [youtube] Add more invidious instances to URL regular expression (#21694) +* [twitter] Improve uploader id extraction (#21705) +* [spankbang] Fix and improve metadata extraction +* [spankbang] Fix extraction (#21763, #21764) ++ [dlive] Add support for dlive.tv (#18080) ++ [livejournal] Add support for livejournal.com (#21526) +* [roosterteeth] Fix free episode extraction (#16094) +* [dbtv] Fix extraction +* [bellator] Fix extraction +- [rudo] Remove extractor (#18430, #18474) +* [facebook] Fallback to twitter:image meta for thumbnail extraction (#21224) +* [bleacherreport] Fix Bleacher Report CMS extraction +* [espn] Fix fivethirtyeight.com extraction +* [5tv] Relax video URL regular expression and support https URLs +* [youtube] Fix is_live extraction (#21734) +* [youtube] Fix authentication (#11270) + + +version 2019.07.12 + +Core ++ [adobepass] Add support for AT&T U-verse (mso ATT) (#13938, #21016) + +Extractors ++ [mgtv] Pass Referer HTTP header for format URLs (#21726) ++ [beeg] Add support for api/v6 v2 URLs without t argument (#21701) +* [voxmedia:volume] Improvevox embed extraction (#16846) +* [funnyordie] Move extraction to VoxMedia extractor (#16846) +* [gameinformer] Fix extraction (#8895, #15363, #17206) +* [funk] Fix extraction (#17915) +* [packtpub] Relax lesson URL regular expression (#21695) +* [packtpub] Fix extraction (#21268) +* [philharmoniedeparis] Relax URL regular expression (#21672) +* [peertube] Detect embed URLs in generic extraction (#21666) +* [mixer:vod] Relax URL regular expression (#21657, #21658) ++ [lecturio] Add support id based URLs (#21630) ++ [go] Add site info for disneynow (#21613) +* [ted] Restrict info regular expression (#21631) +* [twitch:vod] Actualize m3u8 URL (#21538, #21607) +* [vzaar] Fix videos with empty title (#21606) +* [tvland] Fix extraction (#21384) +* [arte] Clean extractor (#15583, #21614) + + +version 2019.07.02 + +Core ++ [utils] Introduce random_user_agent and use as default User-Agent (#21546) + +Extractors ++ [vevo] Add support for embed.vevo.com URLs (#21565) ++ [openload] Add support for oload.biz (#21574) +* [xiami] Update API base URL (#21575) +* [yourporn] Fix extraction (#21585) ++ [acast] Add support for URLs with episode id (#21444) ++ [dailymotion] Add support for DM.player embeds +* [soundcloud] Update client id + + +version 2019.06.27 + +Extractors ++ [go] Add support for disneynow.com (#21528) +* [mixer:vod] Relax URL regular expression (#21531, #21536) +* [drtv] Relax URL regular expression +* [fusion] Fix extraction (#17775, #21269) +- [nfb] Remove extractor (#21518) ++ [beeg] Add support for api/v6 v2 URLs (#21511) ++ [brightcove:new] Add support for playlists (#21331) ++ [openload] Add support for oload.life (#21495) +* [vimeo:channel,group] Make title extraction non fatal +* [vimeo:likes] Implement extrator in terms of channel extractor (#21493) ++ [pornhub] Add support for more paged video sources ++ [pornhub] Add support for downloading single pages and search pages (#15570) +* [pornhub] Rework extractors (#11922, #16078, #17454, #17936) ++ [youtube] Add another signature function pattern +* [tf1] Fix extraction (#21365, #21372) +* [crunchyroll] Move Accept-Language workaround to video extractor since + it causes playlists not to list any videos +* [crunchyroll:playlist] Fix and relax title extraction (#21291, #21443) + + version 2019.06.21 Core @@ -139,7 +791,7 @@ Extractors version 2019.04.17 Extractors -* [openload] Randomize User-Agent (closes #20688) +* [openload] Randomize User-Agent (#20688) + [openload] Add support for oladblock domains (#20471) * [adn] Fix subtitle extraction (#12724) + [aol] Add support for localized websites @@ -704,7 +1356,7 @@ Extractors + [youtube] Extract channel meta fields (#9676, #12939) * [porntube] Fix extraction (#17541) * [asiancrush] Fix extraction (#15630) -+ [twitch:clips] Extend URL regular expression (closes #17559) ++ [twitch:clips] Extend URL regular expression (#17559) + [vzaar] Add support for HLS * [tube8] Fix metadata extraction (#17520) * [eporner] Extract JSON-LD (#17519) diff --git a/README.md b/README.md index 8c48a3012..4f54a5240 100644 --- a/README.md +++ b/README.md @@ -752,8 +752,8 @@ As a last resort, you can also uninstall the version installed by your package m Afterwards, simply follow [our manual installation instructions](https://ytdl-org.github.io/youtube-dl/download.html): ``` -sudo wget https://yt-dl.org/latest/youtube-dl -O /usr/local/bin/youtube-dl -sudo chmod a+x /usr/local/bin/youtube-dl +sudo wget https://yt-dl.org/downloads/latest/youtube-dl -O /usr/local/bin/youtube-dl +sudo chmod a+rx /usr/local/bin/youtube-dl hash -r ``` @@ -835,7 +835,9 @@ In February 2015, the new YouTube player contained a character sequence in a str ### HTTP Error 429: Too Many Requests or 402: Payment Required -These two error codes indicate that the service is blocking your IP address because of overuse. Contact the service and ask them to unblock your IP address, or - if you have acquired a whitelisted IP address already - use the [`--proxy` or `--source-address` options](#network-options) to select another IP address. +These two error codes indicate that the service is blocking your IP address because of overuse. Usually this is a soft block meaning that you can gain access again after solving CAPTCHA. Just open a browser and solve a CAPTCHA the service suggests you and after that [pass cookies](#how-do-i-pass-cookies-to-youtube-dl) to youtube-dl. Note that if your machine has multiple external IPs then you should also pass exactly the same IP you've used for solving CAPTCHA with [`--source-address`](#network-options). Also you may need to pass a `User-Agent` HTTP header of your browser with [`--user-agent`](#workarounds). + +If this is not the case (no CAPTCHA suggested to solve by the service) then you can contact the service and ask them to unblock your IP address, or - if you have acquired a whitelisted IP address already - use the [`--proxy` or `--source-address` options](#network-options) to select another IP address. ### SyntaxError: Non-ASCII character @@ -1216,6 +1218,72 @@ Incorrect: 'PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4' ``` +### Inline values + +Extracting variables is acceptable for reducing code duplication and improving readability of complex expressions. However, you should avoid extracting variables used only once and moving them to opposite parts of the extractor file, which makes reading the linear flow difficult. + +#### Example + +Correct: + +```python +title = self._html_search_regex(r'([^<]+)', webpage, 'title') +``` + +Incorrect: + +```python +TITLE_RE = r'([^<]+)' +# ...some lines of code... +title = self._html_search_regex(TITLE_RE, webpage, 'title') +``` + +### Collapse fallbacks + +Multiple fallback values can quickly become unwieldy. Collapse multiple fallback values into a single expression via a list of patterns. + +#### Example + +Good: + +```python +description = self._html_search_meta( + ['og:description', 'description', 'twitter:description'], + webpage, 'description', default=None) +``` + +Unwieldy: + +```python +description = ( + self._og_search_description(webpage, default=None) + or self._html_search_meta('description', webpage, default=None) + or self._html_search_meta('twitter:description', webpage, default=None)) +``` + +Methods supporting list of patterns are: `_search_regex`, `_html_search_regex`, `_og_search_property`, `_html_search_meta`. + +### Trailing parentheses + +Always move trailing parentheses after the last argument. + +#### Example + +Correct: + +```python + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list) +``` + +Incorrect: + +```python + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list, +) +``` + ### Use convenience conversion and parsing functions Wrap all extracted numeric data into safe functions from [`youtube_dl/utils.py`](https://github.com/ytdl-org/youtube-dl/blob/master/youtube_dl/utils.py): `int_or_none`, `float_or_none`. Use them for string to number conversions as well. diff --git a/devscripts/create-github-release.py b/devscripts/create-github-release.py index 428111b3f..2ddfa1096 100644 --- a/devscripts/create-github-release.py +++ b/devscripts/create-github-release.py @@ -1,7 +1,6 @@ #!/usr/bin/env python from __future__ import unicode_literals -import base64 import io import json import mimetypes @@ -15,7 +14,6 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from youtube_dl.compat import ( compat_basestring, - compat_input, compat_getpass, compat_print, compat_urllib_request, @@ -40,28 +38,20 @@ class GitHubReleaser(object): try: info = netrc.netrc().authenticators(self._NETRC_MACHINE) if info is not None: - self._username = info[0] - self._password = info[2] + self._token = info[2] compat_print('Using GitHub credentials found in .netrc...') return else: compat_print('No GitHub credentials found in .netrc') except (IOError, netrc.NetrcParseError): compat_print('Unable to parse .netrc') - self._username = compat_input( - 'Type your GitHub username or email address and press [Return]: ') - self._password = compat_getpass( - 'Type your GitHub password and press [Return]: ') + self._token = compat_getpass( + 'Type your GitHub PAT (personal access token) and press [Return]: ') def _call(self, req): if isinstance(req, compat_basestring): req = sanitized_Request(req) - # Authorizing manually since GitHub does not response with 401 with - # WWW-Authenticate header set (see - # https://developer.github.com/v3/#basic-authentication) - b64 = base64.b64encode( - ('%s:%s' % (self._username, self._password)).encode('utf-8')).decode('ascii') - req.add_header('Authorization', 'Basic %s' % b64) + req.add_header('Authorization', 'token %s' % self._token) response = self._opener.open(req).read().decode('utf-8') return json.loads(response) diff --git a/docs/supportedsites.md b/docs/supportedsites.md index bfd15b4dc..02bc088ab 100644 --- a/docs/supportedsites.md +++ b/docs/supportedsites.md @@ -26,13 +26,13 @@ - **AcademicEarth:Course** - **acast** - **acast:channel** - - **AddAnime** - **ADN**: Anime Digital Network - **AdobeConnect** - - **AdobeTV** - - **AdobeTVChannel** - - **AdobeTVShow** - - **AdobeTVVideo** + - **adobetv** + - **adobetv:channel** + - **adobetv:embed** + - **adobetv:show** + - **adobetv:video** - **AdultSwim** - **aenetworks**: A+E Networks: A&E, Lifetime, History.com, FYI Network and History Vault - **afreecatv**: afreecatv.com @@ -58,16 +58,8 @@ - **ARD:mediathek** - **ARDBetaMediathek** - **Arkena** - - **arte.tv** - **arte.tv:+7** - - **arte.tv:cinema** - - **arte.tv:concert** - - **arte.tv:creative** - - **arte.tv:ddc** - **arte.tv:embed** - - **arte.tv:future** - - **arte.tv:info** - - **arte.tv:magazine** - **arte.tv:playlist** - **AsianCrush** - **AsianCrushPlaylist** @@ -84,8 +76,6 @@ - **awaan:video** - **AZMedien**: AZ Medien videos - **BaiduVideo**: 百度视频 - - **bambuser** - - **bambuser:channel** - **Bandcamp** - **Bandcamp:album** - **Bandcamp:weekly** @@ -106,6 +96,8 @@ - **Bigflix** - **Bild**: Bild.de - **BiliBili** + - **BilibiliAudio** + - **BilibiliAudioAlbum** - **BioBioChileTV** - **BIQLE** - **BitChute** @@ -183,12 +175,12 @@ - **CNN** - **CNNArticle** - **CNNBlogs** - - **ComCarCoff** - **ComedyCentral** - **ComedyCentralFullEpisodes** - **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 + - **CONtv** - **Corus** - **Coub** - **Cracked** @@ -210,8 +202,6 @@ - **dailymotion** - **dailymotion:playlist** - **dailymotion:user** - - **DaisukiMotto** - - **DaisukiMottoPlaylist** - **daum.net** - **daum.net:clip** - **daum.net:playlist** @@ -231,11 +221,12 @@ - **DiscoveryNetworksDe** - **DiscoveryVR** - **Disney** + - **dlive:stream** + - **dlive:vod** - **Dotsub** - **DouyuShow** - **DouyuTV**: 斗鱼 - **DPlay** - - **DPlayIt** - **DRBonanza** - **Dropbox** - **DrTuber** @@ -288,12 +279,12 @@ - **FiveThirtyEight** - **FiveTV** - **Flickr** - - **Flipagram** - **Folketinget**: Folketinget (ft.dk; Danish parliament) - **FootyRoom** - **Formula1** - **FOX** - **FOX9** + - **FOX9News** - **Foxgay** - **foxnews**: Fox News and Fox Business Video - **foxnews:article** @@ -313,16 +304,12 @@ - **FrontendMastersCourse** - **FrontendMastersLesson** - **Funimation** - - **FunkChannel** - - **FunkMix** - - **FunnyOrDie** + - **Funk** - **Fusion** - **Fux** - **FXNetworks** - **Gaia** - **GameInformer** - - **GameOne** - - **gameone:playlist** - **GameSpot** - **GameStar** - **Gaskrank** @@ -337,14 +324,12 @@ - **Globo** - **GloboArticle** - **Go** - - **Go90** - **GodTube** - **Golem** - **GoogleDrive** - **Goshgay** - **GPUTechConf** - **Groupon** - - **Hark** - **hbo** - **HearThisAt** - **Heise** @@ -373,7 +358,6 @@ - **Hungama** - **HungamaSong** - **Hypem** - - **Iconosquare** - **ign.com** - **imdb**: Internet Movie Database trailers - **imdb:list**: Internet Movie Database lists @@ -405,7 +389,6 @@ - **JeuxVideo** - **Joj** - **Jove** - - **jpopsuki.tv** - **JWPlatform** - **Kakao** - **Kaltura** @@ -413,14 +396,14 @@ - **Kankan** - **Karaoketv** - **KarriereVideos** - - **keek** + - **Katsomo** - **KeezMovies** - **Ketnet** - **KhanAcademy** - **KickStarter** + - **KinjaEmbed** - **KinoPoisk** - **KonserthusetPlay** - - **kontrtube**: KontrTube.ru - Труба зовёт - **KrasView**: Красвью - **Ku6** - **KUSI** @@ -437,7 +420,6 @@ - **Lcp** - **LcpPlay** - **Le**: 乐视网 - - **Learnr** - **Lecture2Go** - **Lecturio** - **LecturioCourse** @@ -458,6 +440,7 @@ - **linkedin:learning:course** - **LinuxAcademy** - **LiTV** + - **LiveJournal** - **LiveLeak** - **LiveLeakEmbed** - **livestream** @@ -470,11 +453,9 @@ - **lynda**: lynda.com videos - **lynda:course**: lynda.com online courses - **m6** - - **macgamestore**: MacGameStore trailers - **mailru**: Видео@Mail.Ru - **mailru:music**: Музыка@Mail.Ru - **mailru:music:search**: Музыка@Mail.Ru - - **MakerTV** - **MallTV** - **mangomolo:live** - **mangomolo:video** @@ -501,14 +482,12 @@ - **Mgoon** - **MGTV**: 芒果TV - **MiaoPai** - - **Minhateca** - **MinistryGrid** - **Minoto** - **miomio.tv** - **MiTele**: mitele.es - **mixcloud** - **mixcloud:playlist** - - **mixcloud:stream** - **mixcloud:user** - **Mixer:live** - **Mixer:vod** @@ -530,11 +509,10 @@ - **mtg**: MTG services - **mtv** - **mtv.de** - - **mtv81** - **mtv:video** + - **mtvjapan** - **mtvservices:embedded** - **MuenchenTV**: münchen.tv - - **MusicPlayOn** - **mva**: Microsoft Virtual Academy videos - **mva:course**: Microsoft Virtual Academy courses - **Mwave** @@ -581,7 +559,6 @@ - **NextTV**: 壹電視 - **Nexx** - **NexxEmbed** - - **nfb**: National Film Board of Canada - **nfl.com** - **NhkVod** - **nhl.com** @@ -640,7 +617,6 @@ - **OnionStudios** - **Ooyala** - **OoyalaExternal** - - **Openload** - **OraTV** - **orf:fm4**: radio FM4 - **orf:fm4:story**: fm4.orf.at stories @@ -651,7 +627,6 @@ - **OutsideTV** - **PacktPub** - **PacktPubCourse** - - **PandaTV**: 熊猫TV - **pandora.tv**: 판도라TV - **ParamountNetwork** - **parliamentlive.tv**: UK parliament videos @@ -687,19 +662,20 @@ - **Pokemon** - **PolskieRadio** - **PolskieRadioCategory** + - **Popcorntimes** - **PopcornTV** - **PornCom** - **PornerBros** - **PornHd** - **PornHub**: PornHub and Thumbzilla - - **PornHubPlaylist** - - **PornHubUserVideos** + - **PornHubPagedVideoList** + - **PornHubUser** + - **PornHubUserVideosUpload** - **Pornotube** - **PornoVoisines** - **PornoXO** - **PornTube** - **PressTV** - - **PromptFile** - **prosiebensat1**: ProSiebenSat.1 Digital - **puhutv** - **puhutv:serie** @@ -740,8 +716,6 @@ - **Restudy** - **Reuters** - **ReverbNation** - - **revision** - - **revision3:embed** - **RICE** - **RMCDecouverte** - **RockstarGames** @@ -764,7 +738,6 @@ - **rtve.es:television** - **RTVNH** - **RTVS** - - **Rudo** - **RUHD** - **rutube**: Rutube videos - **rutube:channel**: Rutube channels @@ -787,11 +760,13 @@ - **screen.yahoo:search**: Yahoo screen search - **Screencast** - **ScreencastOMatic** + - **ScrippsNetworks** - **scrippsnetworks:watch** + - **SCTE** + - **SCTECourse** - **Seeker** - **SenateISVP** - **SendtoNews** - - **ServingSys** - **Servus** - **Sexu** - **SeznamZpravy** @@ -822,6 +797,7 @@ - **soundcloud:set** - **soundcloud:trackstation** - **soundcloud:user** + - **SoundcloudEmbed** - **soundgasm** - **soundgasm:profile** - **southpark.cc.com** @@ -848,7 +824,6 @@ - **Steam** - **Stitcher** - **Streamable** - - **Streamango** - **streamcloud.eu** - **StreamCZ** - **StreetVoice** @@ -890,13 +865,14 @@ - **TeleQuebec** - **TeleQuebecEmission** - **TeleQuebecLive** + - **TeleQuebecSquat** - **TeleTask** - **Telewebion** - **TennisTV** + - **TenPlay** - **TF1** - **TFO** - **TheIntercept** - - **theoperaplatform** - **ThePlatform** - **ThePlatformFeed** - **TheScene** @@ -932,11 +908,12 @@ - **tunein:topic** - **TunePk** - **Turbo** - - **Tutv** - **tv.dfb.de** - **TV2** - **tv2.hu** - **TV2Article** + - **TV2DK** + - **TV2DKBornholmPlay** - **TV4**: tv4.se and tv4play.se - **TV5MondePlus**: TV5MONDE+ - **TVA** @@ -973,10 +950,12 @@ - **twitch:vod** - **twitter** - **twitter:amplify** + - **twitter:broadcast** - **twitter:card** - **udemy** - **udemy:course** - **UDNEmbed**: 聯合影音 + - **UFCArabia** - **UFCTV** - **UKTVPlay** - **umg:de**: Universal Music Deutschland @@ -997,8 +976,6 @@ - **Vbox7** - **VeeHD** - **Veoh** - - **verystream** - - **Vessel** - **Vesti**: Вести.Ru - **Vevo** - **VevoPlaylist** @@ -1013,13 +990,11 @@ - **Viddler** - **Videa** - **video.google:search**: Google Video search - - **video.mit.edu** - **VideoDetective** - **videofy.me** - **videomore** - **videomore:season** - **videomore:video** - - **VideoPremium** - **VideoPress** - **Vidio** - **VidLii** @@ -1029,9 +1004,8 @@ - **Vidzi** - **vier**: vier.be and vijf.be - **vier:videos** - - **ViewLift** - - **ViewLiftEmbed** - - **Viewster** + - **viewlift** + - **viewlift:embed** - **Viidea** - **viki** - **viki:channel** @@ -1097,7 +1071,6 @@ - **Weibo** - **WeiboMobile** - **WeiqiTV**: WQTV - - **Wimp** - **Wistia** - **wnl**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl - **WorldStarHipHop** @@ -1106,9 +1079,10 @@ - **WWE** - **XBef** - **XboxClips** - - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To, XVIDSTAGE, Vid ABC, VidBom, vidlo, RapidVideo.TV, FastVideo.me + - **XFileShare**: XFileShare based sites: ClipWatching, GoUnlimited, GoVid, HolaVid, Streamty, TheVideoBee, Uqload, VidBom, vidlo, VidLocker, VidShare, VUp, XVideoSharing - **XHamster** - **XHamsterEmbed** + - **XHamsterUser** - **xiami:album**: 虾米音乐 - 专辑 - **xiami:artist**: 虾米音乐 - 歌手 - **xiami:collection**: 虾米音乐 - 精选集 @@ -1126,6 +1100,7 @@ - **Yahoo**: Yahoo screen and movies - **yahoo:gyao** - **yahoo:gyao:player** + - **yahoo:japannews**: Yahoo! Japan News - **YandexDisk** - **yandexmusic:album**: Яндекс.Музыка - Альбом - **yandexmusic:playlist**: Яндекс.Музыка - Плейлист diff --git a/test/test_YoutubeDL.py b/test/test_YoutubeDL.py index ce9666171..1e204e551 100644 --- a/test/test_YoutubeDL.py +++ b/test/test_YoutubeDL.py @@ -816,11 +816,15 @@ class TestYoutubeDL(unittest.TestCase): 'webpage_url': 'http://example.com', } - def get_ids(params): + def get_downloaded_info_dicts(params): ydl = YDL(params) - # make a copy because the dictionary can be modified - ydl.process_ie_result(playlist.copy()) - return [int(v['id']) for v in ydl.downloaded_info_dicts] + # make a deep copy because the dictionary and nested entries + # can be modified + ydl.process_ie_result(copy.deepcopy(playlist)) + return ydl.downloaded_info_dicts + + def get_ids(params): + return [int(v['id']) for v in get_downloaded_info_dicts(params)] result = get_ids({}) self.assertEqual(result, [1, 2, 3, 4]) @@ -852,6 +856,22 @@ class TestYoutubeDL(unittest.TestCase): result = get_ids({'playlist_items': '2-4,3-4,3'}) self.assertEqual(result, [2, 3, 4]) + # Tests for https://github.com/ytdl-org/youtube-dl/issues/10591 + # @{ + result = get_downloaded_info_dicts({'playlist_items': '2-4,3-4,3'}) + self.assertEqual(result[0]['playlist_index'], 2) + self.assertEqual(result[1]['playlist_index'], 3) + + result = get_downloaded_info_dicts({'playlist_items': '2-4,3-4,3'}) + self.assertEqual(result[0]['playlist_index'], 2) + self.assertEqual(result[1]['playlist_index'], 3) + self.assertEqual(result[2]['playlist_index'], 4) + + result = get_downloaded_info_dicts({'playlist_items': '4,2'}) + self.assertEqual(result[0]['playlist_index'], 4) + self.assertEqual(result[1]['playlist_index'], 2) + # @} + def test_urlopen_no_file_protocol(self): # see https://github.com/ytdl-org/youtube-dl/issues/8227 ydl = YDL() diff --git a/test/test_YoutubeDLCookieJar.py b/test/test_YoutubeDLCookieJar.py index f959798de..f833efac5 100644 --- a/test/test_YoutubeDLCookieJar.py +++ b/test/test_YoutubeDLCookieJar.py @@ -14,6 +14,9 @@ from youtube_dl.utils import YoutubeDLCookieJar class TestYoutubeDLCookieJar(unittest.TestCase): + def __assert_cookie_has_value(self, cookiejar, key): + self.assertEqual(cookiejar._cookies['www.foobar.foobar']['/'][key].value, key + '_VALUE') + def test_keep_session_cookies(self): cookiejar = YoutubeDLCookieJar('./test/testdata/cookies/session_cookies.txt') cookiejar.load(ignore_discard=True, ignore_expires=True) @@ -32,12 +35,13 @@ class TestYoutubeDLCookieJar(unittest.TestCase): def test_strip_httponly_prefix(self): cookiejar = YoutubeDLCookieJar('./test/testdata/cookies/httponly_cookies.txt') cookiejar.load(ignore_discard=True, ignore_expires=True) + self.__assert_cookie_has_value(cookiejar, 'HTTPONLY_COOKIE') + self.__assert_cookie_has_value(cookiejar, 'JS_ACCESSIBLE_COOKIE') - def assert_cookie_has_value(key): - self.assertEqual(cookiejar._cookies['www.foobar.foobar']['/'][key].value, key + '_VALUE') - - assert_cookie_has_value('HTTPONLY_COOKIE') - assert_cookie_has_value('JS_ACCESSIBLE_COOKIE') + def test_convert_spaces_to_tabs(self): + cookiejar = YoutubeDLCookieJar('./test/testdata/cookies/cookie_file_with_spaces.txt') + cookiejar.load(ignore_discard=True, ignore_expires=True) + self.__assert_cookie_has_value(cookiejar, 'COOKIE') if __name__ == '__main__': diff --git a/test/test_all_urls.py b/test/test_all_urls.py index 465ce0050..81056a999 100644 --- a/test/test_all_urls.py +++ b/test/test_all_urls.py @@ -123,12 +123,6 @@ class TestAllURLsMatching(unittest.TestCase): self.assertMatch('http://video.pbs.org/viralplayer/2365173446/', ['pbs']) self.assertMatch('http://video.pbs.org/widget/partnerplayer/980042464/', ['pbs']) - def test_yahoo_https(self): - # https://github.com/ytdl-org/youtube-dl/issues/2701 - self.assertMatch( - 'https://screen.yahoo.com/smartwatches-latest-wearable-gadgets-163745379-cbs.html', - ['Yahoo']) - def test_no_duplicated_ie_names(self): name_accu = collections.defaultdict(list) for ie in self.ies: diff --git a/test/test_subtitles.py b/test/test_subtitles.py index 7d57a628e..17aaaf20d 100644 --- a/test/test_subtitles.py +++ b/test/test_subtitles.py @@ -26,7 +26,6 @@ from youtube_dl.extractor import ( ThePlatformIE, ThePlatformFeedIE, RTVEALaCartaIE, - FunnyOrDieIE, DemocracynowIE, ) @@ -322,18 +321,6 @@ class TestRtveSubtitles(BaseTestSubtitles): self.assertEqual(md5(subtitles['es']), '69e70cae2d40574fb7316f31d6eb7fca') -class TestFunnyOrDieSubtitles(BaseTestSubtitles): - url = 'http://www.funnyordie.com/videos/224829ff6d/judd-apatow-will-direct-your-vine' - IE = FunnyOrDieIE - - def test_allsubtitles(self): - self.DL.params['writesubtitles'] = True - self.DL.params['allsubtitles'] = True - subtitles = self.getSubtitles() - self.assertEqual(set(subtitles.keys()), set(['en'])) - self.assertEqual(md5(subtitles['en']), 'c5593c193eacd353596c11c2d4f9ecc4') - - class TestDemocracynowSubtitles(BaseTestSubtitles): url = 'http://www.democracynow.org/shows/2015/7/3' IE = DemocracynowIE diff --git a/test/test_utils.py b/test/test_utils.py index 659c6ece5..0896f4150 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -19,6 +19,7 @@ from youtube_dl.utils import ( age_restricted, args_to_str, encode_base_n, + caesar, clean_html, date_from_str, DateRange, @@ -69,11 +70,13 @@ from youtube_dl.utils import ( remove_start, remove_end, remove_quotes, + rot47, shell_quote, smuggle_url, str_to_int, strip_jsonp, strip_or_none, + subtitles_filename, timeconvert, unescapeHTML, unified_strdate, @@ -261,6 +264,11 @@ class TestUtil(unittest.TestCase): self.assertEqual(replace_extension('.abc', 'temp'), '.abc.temp') self.assertEqual(replace_extension('.abc.ext', 'temp'), '.abc.temp') + def test_subtitles_filename(self): + self.assertEqual(subtitles_filename('abc.ext', 'en', 'vtt'), 'abc.en.vtt') + self.assertEqual(subtitles_filename('abc.ext', 'en', 'vtt', 'ext'), 'abc.en.vtt') + self.assertEqual(subtitles_filename('abc.unexpected_ext', 'en', 'vtt', 'ext'), 'abc.unexpected_ext.en.vtt') + def test_remove_start(self): self.assertEqual(remove_start(None, 'A - '), None) self.assertEqual(remove_start('A - B', 'A - '), 'B') @@ -334,6 +342,8 @@ class TestUtil(unittest.TestCase): self.assertEqual(unified_strdate('July 15th, 2013'), '20130715') self.assertEqual(unified_strdate('September 1st, 2013'), '20130901') self.assertEqual(unified_strdate('Sep 2nd, 2013'), '20130902') + self.assertEqual(unified_strdate('November 3rd, 2019'), '20191103') + self.assertEqual(unified_strdate('October 23rd, 2005'), '20051023') def test_unified_timestamps(self): self.assertEqual(unified_timestamp('December 21, 2010'), 1292889600) @@ -489,6 +499,12 @@ class TestUtil(unittest.TestCase): def test_str_to_int(self): self.assertEqual(str_to_int('123,456'), 123456) self.assertEqual(str_to_int('123.456'), 123456) + self.assertEqual(str_to_int(523), 523) + # Python 3 has no long + if sys.version_info < (3, 0): + eval('self.assertEqual(str_to_int(123456L), 123456)') + self.assertEqual(str_to_int('noninteger'), None) + self.assertEqual(str_to_int([]), None) def test_url_basename(self): self.assertEqual(url_basename('http://foo.de/'), '') @@ -1361,6 +1377,20 @@ Line 1 self.assertRaises(ValueError, encode_base_n, 0, 70) self.assertRaises(ValueError, encode_base_n, 0, 60, custom_table) + def test_caesar(self): + self.assertEqual(caesar('ace', 'abcdef', 2), 'cea') + self.assertEqual(caesar('cea', 'abcdef', -2), 'ace') + self.assertEqual(caesar('ace', 'abcdef', -2), 'eac') + self.assertEqual(caesar('eac', 'abcdef', 2), 'ace') + self.assertEqual(caesar('ace', 'abcdef', 0), 'ace') + self.assertEqual(caesar('xyz', 'abcdef', 2), 'xyz') + self.assertEqual(caesar('abc', 'acegik', 2), 'ebg') + self.assertEqual(caesar('ebg', 'acegik', -2), 'abc') + + def test_rot47(self): + self.assertEqual(rot47('youtube-dl'), r'J@FEF36\5=') + self.assertEqual(rot47('YOUTUBE-DL'), r'*~&%&qt\s{') + def test_urshift(self): self.assertEqual(urshift(3, 1), 1) self.assertEqual(urshift(-3, 1), 2147483646) diff --git a/test/testdata/cookies/cookie_file_with_spaces.txt b/test/testdata/cookies/cookie_file_with_spaces.txt new file mode 100644 index 000000000..6fda35fa0 --- /dev/null +++ b/test/testdata/cookies/cookie_file_with_spaces.txt @@ -0,0 +1,5 @@ +# Netscape HTTP Cookie File +# http://curl.haxx.se/rfc/cookie_spec.html +# This is a generated file! Do not edit. + +www.foobar.foobar FALSE / TRUE 2147483647 COOKIE COOKIE_VALUE diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 3e832fec2..19370f62b 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -92,6 +92,7 @@ from .utils import ( YoutubeDLCookieJar, YoutubeDLCookieProcessor, YoutubeDLHandler, + YoutubeDLRedirectHandler, ) from .cache import Cache from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER @@ -852,8 +853,9 @@ class YoutubeDL(object): extract_flat = self.params.get('extract_flat', False) if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or extract_flat is True): - if self.params.get('forcejson', False): - self.to_stdout(json.dumps(ie_result)) + self.__forced_printings( + ie_result, self.prepare_filename(ie_result), + incomplete=True) return ie_result if result_type == 'video': @@ -989,7 +991,7 @@ class YoutubeDL(object): 'playlist_title': ie_result.get('title'), 'playlist_uploader': ie_result.get('uploader'), 'playlist_uploader_id': ie_result.get('uploader_id'), - 'playlist_index': i + playliststart, + 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart, 'extractor': ie_result['extractor'], 'webpage_url': ie_result['webpage_url'], 'webpage_url_basename': url_basename(ie_result['webpage_url']), @@ -1693,6 +1695,36 @@ class YoutubeDL(object): subs[lang] = f return subs + def __forced_printings(self, info_dict, filename, incomplete): + def print_mandatory(field): + if (self.params.get('force%s' % field, False) + and (not incomplete or info_dict.get(field) is not None)): + self.to_stdout(info_dict[field]) + + def print_optional(field): + if (self.params.get('force%s' % field, False) + and info_dict.get(field) is not None): + self.to_stdout(info_dict[field]) + + print_mandatory('title') + print_mandatory('id') + if self.params.get('forceurl', False) and not incomplete: + if info_dict.get('requested_formats') is not None: + for f in info_dict['requested_formats']: + self.to_stdout(f['url'] + f.get('play_path', '')) + else: + # For RTMP URLs, also include the playpath + self.to_stdout(info_dict['url'] + info_dict.get('play_path', '')) + print_optional('thumbnail') + print_optional('description') + if self.params.get('forcefilename', False) and filename is not None: + self.to_stdout(filename) + if self.params.get('forceduration', False) and info_dict.get('duration') is not None: + self.to_stdout(formatSeconds(info_dict['duration'])) + print_mandatory('format') + if self.params.get('forcejson', False): + self.to_stdout(json.dumps(info_dict)) + def process_info(self, info_dict): """Process a single resolved IE result.""" @@ -1703,9 +1735,8 @@ class YoutubeDL(object): if self._num_downloads >= int(max_downloads): raise MaxDownloadsReached() + # TODO: backward compatibility, to be removed info_dict['fulltitle'] = info_dict['title'] - if len(info_dict['title']) > 200: - info_dict['title'] = info_dict['title'][:197] + '...' if 'format' not in info_dict: info_dict['format'] = info_dict['ext'] @@ -1720,29 +1751,7 @@ class YoutubeDL(object): info_dict['_filename'] = filename = self.prepare_filename(info_dict) # Forced printings - if self.params.get('forcetitle', False): - self.to_stdout(info_dict['fulltitle']) - if self.params.get('forceid', False): - self.to_stdout(info_dict['id']) - if self.params.get('forceurl', False): - if info_dict.get('requested_formats') is not None: - for f in info_dict['requested_formats']: - self.to_stdout(f['url'] + f.get('play_path', '')) - else: - # For RTMP URLs, also include the playpath - self.to_stdout(info_dict['url'] + info_dict.get('play_path', '')) - if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None: - self.to_stdout(info_dict['thumbnail']) - if self.params.get('forcedescription', False) and info_dict.get('description') is not None: - self.to_stdout(info_dict['description']) - if self.params.get('forcefilename', False) and filename is not None: - self.to_stdout(filename) - if self.params.get('forceduration', False) and info_dict.get('duration') is not None: - self.to_stdout(formatSeconds(info_dict['duration'])) - if self.params.get('forceformat', False): - self.to_stdout(info_dict['format']) - if self.params.get('forcejson', False): - self.to_stdout(json.dumps(info_dict)) + self.__forced_printings(info_dict, filename, incomplete=False) # Do nothing else if in simulate mode if self.params.get('simulate', False): @@ -1783,6 +1792,8 @@ class YoutubeDL(object): annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext')) if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)): self.to_screen('[info] Video annotations are already present') + elif not info_dict.get('annotations'): + self.report_warning('There are no annotations to write.') else: try: self.to_screen('[info] Writing video annotations to: ' + annofn) @@ -1804,7 +1815,7 @@ class YoutubeDL(object): ie = self.get_info_extractor(info_dict['extractor_key']) for sub_lang, sub_info in subtitles.items(): sub_format = sub_info['ext'] - sub_filename = subtitles_filename(filename, sub_lang, sub_format) + sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext')) if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)): self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format)) else: @@ -2333,6 +2344,7 @@ class YoutubeDL(object): debuglevel = 1 if self.params.get('debug_printtraffic') else 0 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) + redirect_handler = YoutubeDLRedirectHandler() data_handler = compat_urllib_request_DataHandler() # When passing our own FileHandler instance, build_opener won't add the @@ -2346,7 +2358,7 @@ class YoutubeDL(object): file_handler.file_open = file_open opener = compat_urllib_request.build_opener( - proxy_handler, https_handler, cookie_processor, ydlh, data_handler, file_handler) + proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler) # Delete the default user-agent header, which would otherwise apply in # cases where our custom HTTP handler doesn't come into play diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 165c975dd..9a659fc65 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -94,7 +94,7 @@ def _real_main(argv=None): if opts.verbose: write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n') except IOError: - sys.exit('ERROR: batch file could not be read') + sys.exit('ERROR: batch file %s could not be read' % opts.batchfile) all_urls = batch_urls + [url.strip() for url in args] # batch_urls are already striped in read_batch_urls _enc = preferredencoding() all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls] diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py index c75ab131b..d1b86bd13 100644 --- a/youtube_dl/compat.py +++ b/youtube_dl/compat.py @@ -2754,6 +2754,17 @@ else: compat_expanduser = os.path.expanduser +if compat_os_name == 'nt' and sys.version_info < (3, 8): + # os.path.realpath on Windows does not follow symbolic links + # prior to Python 3.8 (see https://bugs.python.org/issue9949) + def compat_realpath(path): + while os.path.islink(path): + path = os.path.abspath(os.readlink(path)) + return path +else: + compat_realpath = os.path.realpath + + if sys.version_info < (3, 0): def compat_print(s): from .utils import preferredencoding @@ -2998,6 +3009,7 @@ __all__ = [ 'compat_os_name', 'compat_parse_qs', 'compat_print', + 'compat_realpath', 'compat_setenv', 'compat_shlex_quote', 'compat_shlex_split', diff --git a/youtube_dl/downloader/dash.py b/youtube_dl/downloader/dash.py index eaa7adf7c..c6d674bc6 100644 --- a/youtube_dl/downloader/dash.py +++ b/youtube_dl/downloader/dash.py @@ -53,7 +53,7 @@ class DashSegmentsFD(FragmentFD): except compat_urllib_error.HTTPError as err: # YouTube may often return 404 HTTP error for a fragment causing the # whole download to fail. However if the same fragment is immediately - # retried with the same request data this usually succeeds (1-2 attemps + # retried with the same request data this usually succeeds (1-2 attempts # is usually enough) thus allowing to download the whole file successfully. # To be future-proof we will retry all fragments that fail with any # HTTP error. diff --git a/youtube_dl/downloader/external.py b/youtube_dl/downloader/external.py index acdb27712..c31f8910a 100644 --- a/youtube_dl/downloader/external.py +++ b/youtube_dl/downloader/external.py @@ -194,6 +194,7 @@ class Aria2cFD(ExternalFD): cmd += self._option('--interface', 'source_address') cmd += self._option('--all-proxy', 'proxy') cmd += self._bool_option('--check-certificate', 'nocheckcertificate', 'false', 'true', '=') + cmd += self._bool_option('--remote-time', 'updatetime', 'true', 'false', '=') cmd += ['--', info_dict['url']] return cmd diff --git a/youtube_dl/downloader/fragment.py b/youtube_dl/downloader/fragment.py index f2e5733b6..02f35459e 100644 --- a/youtube_dl/downloader/fragment.py +++ b/youtube_dl/downloader/fragment.py @@ -190,12 +190,13 @@ class FragmentFD(FileDownloader): }) def _start_frag_download(self, ctx): + resume_len = ctx['complete_frags_downloaded_bytes'] total_frags = ctx['total_frags'] # This dict stores the download progress, it's updated by the progress # hook state = { 'status': 'downloading', - 'downloaded_bytes': ctx['complete_frags_downloaded_bytes'], + 'downloaded_bytes': resume_len, 'fragment_index': ctx['fragment_index'], 'fragment_count': total_frags, 'filename': ctx['filename'], @@ -234,8 +235,8 @@ class FragmentFD(FileDownloader): state['downloaded_bytes'] += frag_downloaded_bytes - ctx['prev_frag_downloaded_bytes'] if not ctx['live']: state['eta'] = self.calc_eta( - start, time_now, estimated_size, - state['downloaded_bytes']) + start, time_now, estimated_size - resume_len, + state['downloaded_bytes'] - resume_len) state['speed'] = s.get('speed') or ctx.get('speed') ctx['speed'] = state['speed'] ctx['prev_frag_downloaded_bytes'] = frag_downloaded_bytes diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py index b59aad73f..84bc34928 100644 --- a/youtube_dl/downloader/hls.py +++ b/youtube_dl/downloader/hls.py @@ -64,7 +64,7 @@ class HlsFD(FragmentFD): s = urlh.read().decode('utf-8', 'ignore') if not self.can_download(s, info_dict): - if info_dict.get('extra_param_to_segment_url'): + if info_dict.get('extra_param_to_segment_url') or info_dict.get('_decryption_key_url'): self.report_error('pycrypto not found. Please install it.') return False self.report_warning( @@ -169,7 +169,7 @@ class HlsFD(FragmentFD): 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( - self._prepare_url(info_dict, decrypt_info['URI'])).read() + self._prepare_url(info_dict, info_dict.get('_decryption_key_url') or decrypt_info['URI'])).read() frag_content = AES.new( decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content) self._append_fragment(ctx, frag_content) diff --git a/youtube_dl/downloader/ism.py b/youtube_dl/downloader/ism.py index 063fcf444..1ca666b4a 100644 --- a/youtube_dl/downloader/ism.py +++ b/youtube_dl/downloader/ism.py @@ -146,7 +146,7 @@ def write_piff_header(stream, params): sps, pps = codec_private_data.split(u32.pack(1))[1:] avcc_payload = u8.pack(1) # configuration version avcc_payload += sps[1:4] # avc profile indication + profile compatibility + avc level indication - avcc_payload += u8.pack(0xfc | (params.get('nal_unit_length_field', 4) - 1)) # complete represenation (1) + reserved (11111) + length size minus one + avcc_payload += u8.pack(0xfc | (params.get('nal_unit_length_field', 4) - 1)) # complete representation (1) + reserved (11111) + length size minus one avcc_payload += u8.pack(1) # reserved (0) + number of sps (0000001) avcc_payload += u16.pack(len(sps)) avcc_payload += sps diff --git a/youtube_dl/extractor/abc.py b/youtube_dl/extractor/abc.py index 4ac323bf6..6637f4f35 100644 --- a/youtube_dl/extractor/abc.py +++ b/youtube_dl/extractor/abc.py @@ -110,17 +110,17 @@ class ABCIViewIE(InfoExtractor): # ABC iview programs are normally available for 14 days only. _TESTS = [{ - 'url': 'https://iview.abc.net.au/show/ben-and-hollys-little-kingdom/series/0/video/ZX9371A050S00', - 'md5': 'cde42d728b3b7c2b32b1b94b4a548afc', + 'url': 'https://iview.abc.net.au/show/gruen/series/11/video/LE1927H001S00', + 'md5': '67715ce3c78426b11ba167d875ac6abf', 'info_dict': { - 'id': 'ZX9371A050S00', + 'id': 'LE1927H001S00', 'ext': 'mp4', - 'title': "Gaston's Birthday", - 'series': "Ben And Holly's Little Kingdom", - 'description': 'md5:f9de914d02f226968f598ac76f105bcf', - 'upload_date': '20180604', - 'uploader_id': 'abc4kids', - 'timestamp': 1528140219, + 'title': "Series 11 Ep 1", + 'series': "Gruen", + 'description': 'md5:52cc744ad35045baf6aded2ce7287f67', + 'upload_date': '20190925', + 'uploader_id': 'abc1', + 'timestamp': 1569445289, }, 'params': { 'skip_download': True, @@ -148,7 +148,7 @@ class ABCIViewIE(InfoExtractor): 'hdnea': token, }) - for sd in ('sd', 'sd-low'): + for sd in ('720', 'sd', 'sd-low'): sd_url = try_get( stream, lambda x: x['streams']['hls'][sd], compat_str) if not sd_url: diff --git a/youtube_dl/extractor/abcnews.py b/youtube_dl/extractor/abcnews.py index cd29aca77..8b407bf9c 100644 --- a/youtube_dl/extractor/abcnews.py +++ b/youtube_dl/extractor/abcnews.py @@ -15,10 +15,13 @@ class AbcNewsVideoIE(AMPIE): IE_NAME = 'abcnews:video' _VALID_URL = r'''(?x) https?:// - abcnews\.go\.com/ (?: - [^/]+/video/(?P[0-9a-z-]+)-| - video/embed\?.*?\bid= + abcnews\.go\.com/ + (?: + [^/]+/video/(?P[0-9a-z-]+)-| + video/embed\?.*?\bid= + )| + fivethirtyeight\.abcnews\.go\.com/video/embed/\d+/ ) (?P\d+) ''' diff --git a/youtube_dl/extractor/abcotvs.py b/youtube_dl/extractor/abcotvs.py index 03b92a39c..0bc69a64f 100644 --- a/youtube_dl/extractor/abcotvs.py +++ b/youtube_dl/extractor/abcotvs.py @@ -4,29 +4,30 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_str from ..utils import ( + dict_get, int_or_none, - parse_iso8601, + try_get, ) class ABCOTVSIE(InfoExtractor): IE_NAME = 'abcotvs' IE_DESC = 'ABC Owned Television Stations' - _VALID_URL = r'https?://(?:abc(?:7(?:news|ny|chicago)?|11|13|30)|6abc)\.com(?:/[^/]+/(?P[^/]+))?/(?P\d+)' + _VALID_URL = r'https?://(?Pabc(?:7(?:news|ny|chicago)?|11|13|30)|6abc)\.com(?:(?:/[^/]+)*/(?P[^/]+))?/(?P\d+)' _TESTS = [ { 'url': 'http://abc7news.com/entertainment/east-bay-museum-celebrates-vintage-synthesizers/472581/', 'info_dict': { - 'id': '472581', + 'id': '472548', 'display_id': 'east-bay-museum-celebrates-vintage-synthesizers', 'ext': 'mp4', - 'title': 'East Bay museum celebrates vintage synthesizers', + 'title': 'East Bay museum celebrates synthesized music', 'description': 'md5:24ed2bd527096ec2a5c67b9d5a9005f3', 'thumbnail': r're:^https?://.*\.jpg$', - 'timestamp': 1421123075, + 'timestamp': 1421118520, 'upload_date': '20150113', - 'uploader': 'Jonathan Bloom', }, 'params': { # m3u8 download @@ -37,39 +38,63 @@ class ABCOTVSIE(InfoExtractor): 'url': 'http://abc7news.com/472581', 'only_matching': True, }, + { + 'url': 'https://6abc.com/man-75-killed-after-being-struck-by-vehicle-in-chester/5725182/', + 'only_matching': True, + }, ] + _SITE_MAP = { + '6abc': 'wpvi', + 'abc11': 'wtvd', + 'abc13': 'ktrk', + 'abc30': 'kfsn', + 'abc7': 'kabc', + 'abc7chicago': 'wls', + 'abc7news': 'kgo', + 'abc7ny': 'wabc', + } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') or video_id + site, display_id, video_id = re.match(self._VALID_URL, url).groups() + display_id = display_id or video_id + station = self._SITE_MAP[site] - webpage = self._download_webpage(url, display_id) + data = self._download_json( + 'https://api.abcotvs.com/v2/content', display_id, query={ + 'id': video_id, + 'key': 'otv.web.%s.story' % station, + 'station': station, + })['data'] + video = try_get(data, lambda x: x['featuredMedia']['video'], dict) or data + video_id = compat_str(dict_get(video, ('id', 'publishedKey'), video_id)) + title = video.get('title') or video['linkText'] - m3u8 = self._html_search_meta( - 'contentURL', webpage, 'm3u8 url', fatal=True).split('?')[0] - - formats = self._extract_m3u8_formats(m3u8, display_id, 'mp4') + formats = [] + m3u8_url = video.get('m3u8') + if m3u8_url: + formats = self._extract_m3u8_formats( + video['m3u8'].split('?')[0], display_id, 'mp4', m3u8_id='hls', fatal=False) + mp4_url = video.get('mp4') + if mp4_url: + formats.append({ + 'abr': 128, + 'format_id': 'https', + 'height': 360, + 'url': mp4_url, + 'width': 640, + }) self._sort_formats(formats) - title = self._og_search_title(webpage).strip() - description = self._og_search_description(webpage).strip() - thumbnail = self._og_search_thumbnail(webpage) - timestamp = parse_iso8601(self._search_regex( - r'
\s*
', + webpage, 'description', default=None) if not video_description: video_description = self._html_search_meta( 'description', webpage, default=None) - if not video_description and mobj.group('pro'): + if not video_description and is_pro: orig_webpage = self._download_webpage( orig_url, video_id, note='Downloading webpage for description', @@ -682,7 +710,7 @@ class VimeoIE(VimeoBaseInfoExtractor): if orig_webpage: video_description = self._html_search_meta( 'description', orig_webpage, default=None) - if not video_description and not mobj.group('player'): + if not video_description and not is_player: self._downloader.report_warning('Cannot find video description') # Extract upload date @@ -722,7 +750,6 @@ class VimeoIE(VimeoBaseInfoExtractor): channel_url = 'https://vimeo.com/channels/%s' % channel_id if channel_id else None info_dict = { - 'id': video_id, 'formats': formats, 'timestamp': unified_timestamp(timestamp), 'description': video_description, @@ -740,9 +767,9 @@ class VimeoIE(VimeoBaseInfoExtractor): return info_dict -class VimeoOndemandIE(VimeoBaseInfoExtractor): +class VimeoOndemandIE(VimeoIE): IE_NAME = 'vimeo:ondemand' - _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P[^/?#&]+)' + _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/([^/]+/)?(?P[^/?#&]+)' _TESTS = [{ # ondemand video not available via https://vimeo.com/id 'url': 'https://vimeo.com/ondemand/20704', @@ -754,24 +781,32 @@ class VimeoOndemandIE(VimeoBaseInfoExtractor): 'uploader': 'גם סרטים', 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms', 'uploader_id': 'gumfilms', + 'description': 'md5:4c027c965e439de4baab621e48b60791', + 'upload_date': '20140906', + 'timestamp': 1410032453, }, 'params': { 'format': 'best[protocol=https]', }, + 'expected_warnings': ['Unable to download JSON metadata'], }, { # requires Referer to be passed along with og:video:url 'url': 'https://vimeo.com/ondemand/36938/126682985', 'info_dict': { - 'id': '126682985', + 'id': '126584684', 'ext': 'mp4', 'title': 'Rävlock, rätt läte på rätt plats', 'uploader': 'Lindroth & Norin', - 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847', - 'uploader_id': 'user14430847', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/lindrothnorin', + 'uploader_id': 'lindrothnorin', + 'description': 'md5:c3c46a90529612c8279fb6af803fc0df', + 'upload_date': '20150502', + 'timestamp': 1430586422, }, 'params': { 'skip_download': True, }, + 'expected_warnings': ['Unable to download JSON metadata'], }, { 'url': 'https://vimeo.com/ondemand/nazmaalik', 'only_matching': True, @@ -783,16 +818,6 @@ class VimeoOndemandIE(VimeoBaseInfoExtractor): 'only_matching': True, }] - def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - return self.url_result( - # Some videos require Referer to be passed along with og:video:url - # similarly to generic vimeo embeds (e.g. - # https://vimeo.com/ondemand/36938/126682985). - VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url), - VimeoIE.ie_key()) - class VimeoChannelIE(VimeoBaseInfoExtractor): IE_NAME = 'vimeo:channel' @@ -808,6 +833,7 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): }, 'playlist_mincount': 25, }] + _BASE_URL_TEMPL = 'https://vimeo.com/channels/%s' def _page_url(self, base_url, pagenum): return '%s/videos/page:%d/' % (base_url, pagenum) @@ -816,33 +842,6 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): return self._TITLE or self._html_search_regex( self._TITLE_RE, webpage, 'list title', fatal=False) - def _login_list_password(self, page_url, list_id, webpage): - login_form = self._search_regex( - r'(?s)]+?id="pw_form"(.*?)', - webpage, 'login form', default=None) - if not login_form: - return webpage - - password = self._downloader.params.get('videopassword') - if password is None: - raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True) - fields = self._hidden_inputs(login_form) - token, vuid = self._extract_xsrft_and_vuid(webpage) - fields['token'] = token - fields['password'] = password - post = urlencode_postdata(fields) - password_path = self._search_regex( - r'action="([^"]+)"', login_form, 'password URL') - password_url = compat_urlparse.urljoin(page_url, password_path) - password_request = sanitized_Request(password_url, post) - password_request.add_header('Content-type', 'application/x-www-form-urlencoded') - self._set_vimeo_cookie('vuid', vuid) - self._set_vimeo_cookie('xsrft', token) - - return self._download_webpage( - password_request, list_id, - 'Verifying the password', 'Wrong password') - def _title_and_entries(self, list_id, base_url): for pagenum in itertools.count(1): page_url = self._page_url(base_url, pagenum) @@ -851,7 +850,6 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): 'Downloading page %s' % pagenum) if pagenum == 1: - webpage = self._login_list_password(page_url, list_id, webpage) yield self._extract_list_title(webpage) # Try extracting href first since not all videos are available via @@ -879,14 +877,13 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): return self.playlist_result(title_and_entries, list_id, list_title) def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - channel_id = mobj.group('id') - return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id) + channel_id = self._match_id(url) + return self._extract_videos(channel_id, self._BASE_URL_TEMPL % channel_id) class VimeoUserIE(VimeoChannelIE): IE_NAME = 'vimeo:user' - _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P[^/]+)(?:/videos|[#?]|$)' + _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P[^/]+)(?:/videos|[#?]|$)' _TITLE_RE = r']+?class="user">([^<>]+?)' _TESTS = [{ 'url': 'https://vimeo.com/nkistudio/videos', @@ -896,16 +893,12 @@ class VimeoUserIE(VimeoChannelIE): }, 'playlist_mincount': 66, }] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - name = mobj.group('name') - return self._extract_videos(name, 'https://vimeo.com/%s' % name) + _BASE_URL_TEMPL = 'https://vimeo.com/%s' -class VimeoAlbumIE(VimeoChannelIE): +class VimeoAlbumIE(VimeoBaseInfoExtractor): IE_NAME = 'vimeo:album' - _VALID_URL = r'https://vimeo\.com/album/(?P\d+)(?:$|[?#]|/(?!video))' + _VALID_URL = r'https://vimeo\.com/(?:album|showcase)/(?P\d+)(?:$|[?#]|/(?!video))' _TITLE_RE = r'
\n\s*

(.*?)

' _TESTS = [{ 'url': 'https://vimeo.com/album/2632481', @@ -925,42 +918,81 @@ class VimeoAlbumIE(VimeoChannelIE): 'params': { 'videopassword': 'youtube-dl', } - }, { - 'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail', - 'only_matching': True, - }, { - # TODO: respect page number - 'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail', - 'only_matching': True, }] + _PAGE_SIZE = 100 - def _page_url(self, base_url, pagenum): - return '%s/page:%d/' % (base_url, pagenum) + def _fetch_page(self, album_id, authorizaion, hashed_pass, page): + api_page = page + 1 + query = { + 'fields': 'link,uri', + 'page': api_page, + 'per_page': self._PAGE_SIZE, + } + if hashed_pass: + query['_hashed_pass'] = hashed_pass + videos = self._download_json( + 'https://api.vimeo.com/albums/%s/videos' % album_id, + album_id, 'Downloading page %d' % api_page, query=query, headers={ + 'Authorization': 'jwt ' + authorizaion, + })['data'] + for video in videos: + link = video.get('link') + if not link: + continue + uri = video.get('uri') + video_id = self._search_regex(r'/videos/(\d+)', uri, 'video_id', default=None) if uri else None + yield self.url_result(link, VimeoIE.ie_key(), video_id) def _real_extract(self, url): album_id = self._match_id(url) - return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id) + webpage = self._download_webpage(url, album_id) + viewer = self._parse_json(self._search_regex( + r'bootstrap_data\s*=\s*({.+?})', + webpage, 'bootstrap data'), album_id)['viewer'] + jwt = viewer['jwt'] + album = self._download_json( + 'https://api.vimeo.com/albums/' + album_id, + album_id, headers={'Authorization': 'jwt ' + jwt}, + query={'fields': 'description,name,privacy'}) + hashed_pass = None + if try_get(album, lambda x: x['privacy']['view']) == 'password': + password = self._downloader.params.get('videopassword') + if not password: + raise ExtractorError( + 'This album is protected by a password, use the --video-password option', + expected=True) + self._set_vimeo_cookie('vuid', viewer['vuid']) + try: + hashed_pass = self._download_json( + 'https://vimeo.com/showcase/%s/auth' % album_id, + album_id, 'Verifying the password', data=urlencode_postdata({ + 'password': password, + 'token': viewer['xsrft'], + }), headers={ + 'X-Requested-With': 'XMLHttpRequest', + })['hashed_pass'] + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401: + raise ExtractorError('Wrong password', expected=True) + raise + entries = OnDemandPagedList(functools.partial( + self._fetch_page, album_id, jwt, hashed_pass), self._PAGE_SIZE) + return self.playlist_result( + entries, album_id, album.get('name'), album.get('description')) -class VimeoGroupsIE(VimeoAlbumIE): +class VimeoGroupsIE(VimeoChannelIE): IE_NAME = 'vimeo:group' - _VALID_URL = r'https://vimeo\.com/groups/(?P[^/]+)(?:/(?!videos?/\d+)|$)' + _VALID_URL = r'https://vimeo\.com/groups/(?P[^/]+)(?:/(?!videos?/\d+)|$)' _TESTS = [{ - 'url': 'https://vimeo.com/groups/rolexawards', + 'url': 'https://vimeo.com/groups/kattykay', 'info_dict': { - 'id': 'rolexawards', - 'title': 'Rolex Awards for Enterprise', + 'id': 'kattykay', + 'title': 'Katty Kay', }, - 'playlist_mincount': 73, + 'playlist_mincount': 27, }] - - def _extract_list_title(self, webpage): - return self._og_search_title(webpage, fatal=False) - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - name = mobj.group('name') - return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name) + _BASE_URL_TEMPL = 'https://vimeo.com/groups/%s' class VimeoReviewIE(VimeoBaseInfoExtractor): @@ -976,7 +1008,9 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): 'title': "DICK HARDWICK 'Comedian'", 'uploader': 'Richard Hardwick', 'uploader_id': 'user21297594', - } + 'description': "Comedian Dick Hardwick's five minute demo filmed in front of a live theater audience.\nEdit by Doug Mattocks", + }, + 'expected_warnings': ['Unable to download JSON metadata'], }, { 'note': 'video player needs Referer', 'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053', @@ -989,7 +1023,8 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): 'duration': 2773, 'thumbnail': r're:^https?://.*\.jpg$', 'uploader_id': 'user22258446', - } + }, + 'skip': 'video gone', }, { 'note': 'Password protected', 'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde', @@ -1009,33 +1044,20 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): def _real_initialize(self): self._login() - def _get_config_url(self, webpage_url, video_id, video_password_verified=False): - webpage = self._download_webpage(webpage_url, video_id) - config_url = self._html_search_regex( - r'data-config-url=(["\'])(?P(?:(?!\1).)+)\1', webpage, - 'config URL', default=None, group='url') - if not config_url: - data = self._parse_json(self._search_regex( - r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data', - default=NO_DEFAULT if video_password_verified else '{}'), video_id) - config = data.get('vimeo_esi', {}).get('config', {}) - config_url = config.get('configUrl') or try_get(config, lambda x: x['clipData']['configUrl']) - if config_url is None: - self._verify_video_password(webpage_url, video_id, webpage) - config_url = self._get_config_url( - webpage_url, video_id, video_password_verified=True) - return config_url - def _real_extract(self, url): page_url, video_id = re.match(self._VALID_URL, url).groups() - config_url = self._get_config_url(url, video_id) + clip_data = self._download_json( + page_url.replace('/review/', '/review/data/'), + video_id)['clipData'] + config_url = clip_data['configUrl'] config = self._download_json(config_url, video_id) info_dict = self._parse_config(config, video_id) - source_format = self._extract_original_format(page_url, video_id) + source_format = self._extract_original_format( + page_url + '/action', video_id) if source_format: info_dict['formats'].append(source_format) self._vimeo_sort_formats(info_dict['formats']) - info_dict['id'] = video_id + info_dict['description'] = clean_html(clip_data.get('description')) return info_dict @@ -1089,94 +1111,17 @@ class VimeoLikesIE(VimeoChannelIE): return self._extract_videos(user_id, 'https://vimeo.com/%s/likes' % user_id) -class VHXEmbedIE(InfoExtractor): +class VHXEmbedIE(VimeoBaseInfoExtractor): IE_NAME = 'vhx:embed' _VALID_URL = r'https?://embed\.vhx\.tv/videos/(?P\d+)' - def _call_api(self, video_id, access_token, path='', query=None): - return self._download_json( - 'https://api.vhx.tv/videos/' + video_id + path, video_id, headers={ - 'Authorization': 'Bearer ' + access_token, - }, query=query) - def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - credentials = self._parse_json(self._search_regex( - r'(?s)credentials\s*:\s*({.+?}),', webpage, - 'config'), video_id, js_to_json) - access_token = credentials['access_token'] - - query = {} - for k, v in credentials.items(): - if k in ('authorization', 'authUserToken', 'ticket') and v and v != 'undefined': - if k == 'authUserToken': - query['auth_user_token'] = v - else: - query[k] = v - files = self._call_api(video_id, access_token, '/files', query) - - formats = [] - for f in files: - href = try_get(f, lambda x: x['_links']['source']['href']) - if not href: - continue - method = f.get('method') - if method == 'hls': - formats.extend(self._extract_m3u8_formats( - href, video_id, 'mp4', 'm3u8_native', - m3u8_id='hls', fatal=False)) - elif method == 'dash': - formats.extend(self._extract_mpd_formats( - href, video_id, mpd_id='dash', fatal=False)) - else: - fmt = { - 'filesize': int_or_none(try_get(f, lambda x: x['size']['bytes'])), - 'format_id': 'http', - 'preference': 1, - 'url': href, - 'vcodec': f.get('codec'), - } - quality = f.get('quality') - if quality: - fmt.update({ - 'format_id': 'http-' + quality, - 'height': int_or_none(self._search_regex(r'(\d+)p', quality, 'height', default=None)), - }) - formats.append(fmt) - self._sort_formats(formats) - - video_data = self._call_api(video_id, access_token) - title = video_data.get('title') or video_data['name'] - - subtitles = {} - for subtitle in try_get(video_data, lambda x: x['tracks']['subtitles'], list) or []: - lang = subtitle.get('srclang') or subtitle.get('label') - for _link in subtitle.get('_links', {}).values(): - href = _link.get('href') - if not href: - continue - subtitles.setdefault(lang, []).append({ - 'url': href, - }) - - q = qualities(['small', 'medium', 'large', 'source']) - thumbnails = [] - for thumbnail_id, thumbnail_url in video_data.get('thumbnail', {}).items(): - thumbnails.append({ - 'id': thumbnail_id, - 'url': thumbnail_url, - 'preference': q(thumbnail_id), - }) - - return { - 'id': video_id, - 'title': title, - 'description': video_data.get('description'), - 'duration': int_or_none(try_get(video_data, lambda x: x['duration']['seconds'])), - 'formats': formats, - 'subtitles': subtitles, - 'thumbnails': thumbnails, - 'timestamp': unified_timestamp(video_data.get('created_at')), - 'view_count': int_or_none(video_data.get('plays_count')), - } + config_url = self._parse_json(self._search_regex( + r'window\.OTTData\s*=\s*({.+})', webpage, + 'ott data'), video_id, js_to_json)['config_url'] + config = self._download_json(config_url, video_id) + info = self._parse_config(config, video_id) + self._vimeo_sort_formats(info['formats']) + return info diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index f57ed2288..00ec006c4 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals import collections +import functools import re from .common import InfoExtractor @@ -11,8 +12,8 @@ from ..utils import ( ExtractorError, get_element_by_class, int_or_none, + OnDemandPagedList, orderedSet, - remove_start, str_or_none, str_to_int, unescapeHTML, @@ -21,6 +22,7 @@ from ..utils import ( urlencode_postdata, ) from .dailymotion import DailymotionIE +from .odnoklassniki import OdnoklassnikiIE from .pladform import PladformIE from .vimeo import VimeoIE from .youtube import YoutubeIE @@ -60,6 +62,18 @@ class VKBaseIE(InfoExtractor): def _real_initialize(self): self._login() + def _download_payload(self, path, video_id, data, fatal=True): + data['al'] = 1 + code, payload = self._download_json( + 'https://vk.com/%s.php' % path, video_id, + data=urlencode_postdata(data), fatal=fatal, + headers={'X-Requested-With': 'XMLHttpRequest'})['payload'] + if code == '3': + self.raise_login_required() + elif code == '8': + raise ExtractorError(clean_html(payload[0][1:-1]), expected=True) + return payload + class VKIE(VKBaseIE): IE_NAME = 'vk' @@ -96,7 +110,6 @@ class VKIE(VKBaseIE): }, { 'url': 'http://vk.com/video205387401_165548505', - 'md5': '6c0aeb2e90396ba97035b9cbde548700', 'info_dict': { 'id': '205387401_165548505', 'ext': 'mp4', @@ -110,18 +123,18 @@ class VKIE(VKBaseIE): }, { 'note': 'Embedded video', - 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1', - 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a', + 'url': 'https://vk.com/video_ext.php?oid=-77521&id=162222515&hash=87b046504ccd8bfa', + 'md5': '7babad3b85ea2e91948005b1b8b0cb84', 'info_dict': { - 'id': '32194266_162925554', + 'id': '-77521_162222515', 'ext': 'mp4', - 'uploader': 'Vladimir Gavrin', - 'title': 'Lin Dan', - 'duration': 101, - 'upload_date': '20120730', - 'view_count': int, + 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*', + 'title': 'ProtivoGunz - Хуёвая песня', + 'duration': 195, + 'upload_date': '20120212', + 'timestamp': 1329049880, + 'uploader_id': '-77521', }, - 'skip': 'This video has been removed from public access.', }, { # VIDEO NOW REMOVED @@ -138,18 +151,19 @@ class VKIE(VKBaseIE): 'upload_date': '20121218', 'view_count': int, }, - 'skip': 'Requires vk account credentials', + 'skip': 'Removed', }, { 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d', - 'md5': '4d7a5ef8cf114dfa09577e57b2993202', 'info_dict': { 'id': '-43215063_168067957', 'ext': 'mp4', - 'uploader': 'Киномания - лучшее из мира кино', + 'uploader': 'Bro Mazter', 'title': ' ', 'duration': 7291, 'upload_date': '20140328', + 'uploader_id': '223413403', + 'timestamp': 1396018030, }, 'skip': 'Requires vk account credentials', }, @@ -165,7 +179,7 @@ class VKIE(VKBaseIE): 'upload_date': '20140626', 'view_count': int, }, - 'skip': 'Only works from Russia', + 'skip': 'Removed', }, { # video (removed?) only available with list id @@ -204,8 +218,7 @@ class VKIE(VKBaseIE): 'id': 'k3lz2cmXyRuJQSjGHUv', 'ext': 'mp4', 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f', - # TODO: fix test by fixing dailymotion description extraction - 'description': 'md5:c651358f03c56f1150b555c26d90a0fd', + 'description': 'md5:424b8e88cc873217f520e582ba28bb36', 'uploader': 'AniLibria.Tv', 'upload_date': '20160914', 'uploader_id': 'x1p5vl5', @@ -247,6 +260,9 @@ class VKIE(VKBaseIE): 'uploader_id': '-387766', 'timestamp': 1475137527, }, + 'params': { + 'skip_download': True, + }, }, { # live stream, hls and rtmp links, most likely already finished live @@ -288,80 +304,94 @@ class VKIE(VKBaseIE): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('videoid') + mv_data = {} if video_id: - info_url = 'https://vk.com/al_video.php?act=show_inline&al=1&video=' + video_id + data = { + 'act': 'show_inline', + 'video': video_id, + } # Some videos (removed?) can only be downloaded with list id specified list_id = mobj.group('list_id') if list_id: - info_url += '&list=%s' % list_id + data['list'] = list_id + + payload = self._download_payload('al_video', video_id, data) + info_page = payload[1] + opts = payload[-1] + mv_data = opts.get('mvData') or {} + player = opts.get('player') or {} else: - info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query') video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id')) - info_page = self._download_webpage(info_url, video_id) + info_page = self._download_webpage( + 'http://vk.com/video_ext.php?' + mobj.group('embed_query'), video_id) - error_message = self._html_search_regex( - [r'(?s)]+class="video_layer_message"[^>]*>(.+?)', - r'(?s)]+id="video_ext_msg"[^>]*>(.+?)'], - info_page, 'error message', default=None) - if error_message: - raise ExtractorError(error_message, expected=True) + error_message = self._html_search_regex( + [r'(?s)]+class="video_layer_message"[^>]*>(.+?)', + r'(?s)]+id="video_ext_msg"[^>]*>(.+?)'], + info_page, 'error message', default=None) + if error_message: + raise ExtractorError(error_message, expected=True) - if re.search(r'/login\.php\?.*\bact=security_check', info_page): - raise ExtractorError( - 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.', - expected=True) + if re.search(r'/login\.php\?.*\bact=security_check', info_page): + raise ExtractorError( + 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.', + expected=True) - ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.' + ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.' - ERRORS = { - r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<': - ERROR_COPYRIGHT, + ERRORS = { + r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<': + ERROR_COPYRIGHT, - r'>The video .*? was removed from public access by request of the copyright holder.<': - ERROR_COPYRIGHT, + r'>The video .*? was removed from public access by request of the copyright holder.<': + ERROR_COPYRIGHT, - r'Please log in or <': - 'Video %s is only available for registered users, ' - 'use --username and --password options to provide account credentials.', + r'Please log in or <': + 'Video %s is only available for registered users, ' + 'use --username and --password options to provide account credentials.', - r'Unknown error': - 'Video %s does not exist.', + r'Unknown error': + 'Video %s does not exist.', - r'Видео временно недоступно': - 'Video %s is temporarily unavailable.', + r'Видео временно недоступно': + 'Video %s is temporarily unavailable.', - r'Access denied': - 'Access denied to video %s.', + r'Access denied': + 'Access denied to video %s.', - r'Видеозапись недоступна, так как её автор был заблокирован.': - 'Video %s is no longer available, because its author has been blocked.', + r'Видеозапись недоступна, так как её автор был заблокирован.': + 'Video %s is no longer available, because its author has been blocked.', - r'This video is no longer available, because its author has been blocked.': - 'Video %s is no longer available, because its author has been blocked.', + r'This video is no longer available, because its author has been blocked.': + 'Video %s is no longer available, because its author has been blocked.', - r'This video is no longer available, because it has been deleted.': - 'Video %s is no longer available, because it has been deleted.', + r'This video is no longer available, because it has been deleted.': + 'Video %s is no longer available, because it has been deleted.', - r'The video .+? is not available in your region.': - 'Video %s is not available in your region.', - } + r'The video .+? is not available in your region.': + 'Video %s is not available in your region.', + } - for error_re, error_msg in ERRORS.items(): - if re.search(error_re, info_page): - raise ExtractorError(error_msg % video_id, expected=True) + for error_re, error_msg in ERRORS.items(): + if re.search(error_re, info_page): + raise ExtractorError(error_msg % video_id, expected=True) + + player = self._parse_json(self._search_regex( + r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', + info_page, 'player params'), video_id) youtube_url = YoutubeIE._extract_url(info_page) if youtube_url: - return self.url_result(youtube_url, ie=YoutubeIE.ie_key()) + return self.url_result(youtube_url, YoutubeIE.ie_key()) vimeo_url = VimeoIE._extract_url(url, info_page) if vimeo_url is not None: - return self.url_result(vimeo_url) + return self.url_result(vimeo_url, VimeoIE.ie_key()) pladform_url = PladformIE._extract_url(info_page) if pladform_url: - return self.url_result(pladform_url) + return self.url_result(pladform_url, PladformIE.ie_key()) m_rutube = re.search( r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page) @@ -374,6 +404,10 @@ class VKIE(VKBaseIE): if dailymotion_urls: return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key()) + odnoklassniki_url = OdnoklassnikiIE._extract_url(info_page) + if odnoklassniki_url: + return self.url_result(odnoklassniki_url, OdnoklassnikiIE.ie_key()) + m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page) if m_opts: m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1)) @@ -383,29 +417,7 @@ class VKIE(VKBaseIE): opts_url = 'http:' + opts_url return self.url_result(opts_url) - # vars does not look to be served anymore since 24.10.2016 - data = self._parse_json( - self._search_regex( - r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'), - video_id, fatal=False) - - # is served instead - if not data: - data = self._parse_json( - self._search_regex( - [r'\s*({.+?})\s*', r'\s*({.+})'], - info_page, 'json', default='{}'), - video_id) - if data: - data = data['player']['params'][0] - - if not data: - data = self._parse_json( - self._search_regex( - r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page, - 'player params'), - video_id)['params'][0] - + data = player['params'][0] title = unescapeHTML(data['md_title']) # 2 = live @@ -454,12 +466,12 @@ class VKIE(VKBaseIE): 'title': title, 'thumbnail': data.get('jpg'), 'uploader': data.get('md_author'), - 'uploader_id': str_or_none(data.get('author_id')), - 'duration': data.get('duration'), + 'uploader_id': str_or_none(data.get('author_id') or mv_data.get('authorId')), + 'duration': int_or_none(data.get('duration') or mv_data.get('duration')), 'timestamp': timestamp, 'view_count': view_count, - 'like_count': int_or_none(data.get('liked')), - 'dislike_count': int_or_none(data.get('nolikes')), + 'like_count': int_or_none(mv_data.get('likes')), + 'comment_count': int_or_none(mv_data.get('commcount')), 'is_live': is_live, } @@ -467,15 +479,23 @@ class VKIE(VKBaseIE): class VKUserVideosIE(VKBaseIE): IE_NAME = 'vk:uservideos' IE_DESC = "VK - User's Videos" - _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)' + _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&](?:.*?\bsection=(?P
\w+))?|$)' _TEMPLATE_URL = 'https://vk.com/videos' _TESTS = [{ - 'url': 'http://vk.com/videos205387401', + 'url': 'https://vk.com/videos-767561', 'info_dict': { - 'id': '205387401', - 'title': "Tom Cruise's Videos", + 'id': '-767561_all', }, - 'playlist_mincount': 4, + 'playlist_mincount': 1150, + }, { + 'url': 'https://vk.com/videos-767561?section=uploaded', + 'info_dict': { + 'id': '-767561_uploaded', + }, + 'playlist_mincount': 425, + }, { + 'url': 'http://vk.com/videos205387401', + 'only_matching': True, }, { 'url': 'http://vk.com/videos-77521', 'only_matching': True, @@ -489,22 +509,33 @@ class VKUserVideosIE(VKBaseIE): 'url': 'http://new.vk.com/videos205387401', 'only_matching': True, }] + _PAGE_SIZE = 1000 + _VIDEO = collections.namedtuple('Video', ['owner_id', 'id']) + + def _fetch_page(self, page_id, section, page): + l = self._download_payload('al_video', page_id, { + 'act': 'load_videos_silent', + 'offset': page * self._PAGE_SIZE, + 'oid': page_id, + 'section': section, + })[0][section]['list'] + + for video in l: + v = self._VIDEO._make(video[:2]) + video_id = '%d_%d' % (v.owner_id, v.id) + yield self.url_result( + 'http://vk.com/video' + video_id, VKIE.ie_key(), video_id) def _real_extract(self, url): - page_id = self._match_id(url) + page_id, section = re.match(self._VALID_URL, url).groups() + if not section: + section = 'all' - webpage = self._download_webpage(url, page_id) + entries = OnDemandPagedList( + functools.partial(self._fetch_page, page_id, section), + self._PAGE_SIZE) - entries = [ - self.url_result( - 'http://vk.com/video' + video_id, 'VK', video_id=video_id) - for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))] - - title = unescapeHTML(self._search_regex( - r'\s*([^<]+?)\s+\|\s+\d+\s+videos', - webpage, 'title', default=page_id)) - - return self.playlist_result(entries, page_id, title) + return self.playlist_result(entries, '%s_%s' % (page_id, section)) class VKWallPostIE(VKBaseIE): @@ -514,15 +545,15 @@ class VKWallPostIE(VKBaseIE): # public page URL, audio playlist 'url': 'https://vk.com/bs.official?w=wall-23538238_35', 'info_dict': { - 'id': '23538238_35', - 'title': 'Black Shadow - Wall post 23538238_35', + 'id': '-23538238_35', + 'title': 'Black Shadow - Wall post -23538238_35', 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c', }, 'playlist': [{ 'md5': '5ba93864ec5b85f7ce19a9af4af080f6', 'info_dict': { 'id': '135220665_111806521', - 'ext': 'mp3', + 'ext': 'mp4', 'title': 'Black Shadow - Слепое Верование', 'duration': 370, 'uploader': 'Black Shadow', @@ -533,18 +564,16 @@ class VKWallPostIE(VKBaseIE): 'md5': '4cc7e804579122b17ea95af7834c9233', 'info_dict': { 'id': '135220665_111802303', - 'ext': 'mp3', + 'ext': 'mp4', 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!', 'duration': 423, 'uploader': 'Black Shadow', 'artist': 'Black Shadow', 'track': 'Война - Негасимое Бездны Пламя!', }, - 'params': { - 'skip_download': True, - }, }], 'params': { + 'skip_download': True, 'usenetrc': True, }, 'skip': 'Requires vk account credentials', @@ -553,7 +582,7 @@ class VKWallPostIE(VKBaseIE): 'url': 'https://vk.com/wall85155021_6319', 'info_dict': { 'id': '85155021_6319', - 'title': 'Sergey Gorbunov - Wall post 85155021_6319', + 'title': 'Сергей Горбунов - Wall post 85155021_6319', }, 'playlist_count': 1, 'params': { @@ -569,58 +598,72 @@ class VKWallPostIE(VKBaseIE): 'url': 'https://m.vk.com/wall-23538238_35', 'only_matching': True, }] + _BASE64_CHARS = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMN0PQRSTUVWXYZO123456789+/=' + _AUDIO = collections.namedtuple('Audio', ['id', 'owner_id', 'url', 'title', 'performer', 'duration', 'album_id', 'unk', 'author_link', 'lyrics', 'flags', 'context', 'extra', 'hashes', 'cover_url', 'ads']) + + def _decode(self, enc): + dec = '' + e = n = 0 + for c in enc: + r = self._BASE64_CHARS.index(c) + cond = n % 4 + e = 64 * e + r if cond else r + n += 1 + if cond: + dec += chr(255 & e >> (-2 * n & 6)) + return dec + + def _unmask_url(self, mask_url, vk_id): + if 'audio_api_unavailable' in mask_url: + extra = mask_url.split('?extra=')[1].split('#') + func, base = self._decode(extra[1]).split(chr(11)) + mask_url = list(self._decode(extra[0])) + url_len = len(mask_url) + indexes = [None] * url_len + index = int(base) ^ vk_id + for n in range(url_len - 1, -1, -1): + index = (url_len * (n + 1) ^ index + n) % url_len + indexes[n] = index + for n in range(1, url_len): + c = mask_url[n] + index = indexes[url_len - 1 - n] + mask_url[n] = mask_url[index] + mask_url[index] = c + mask_url = ''.join(mask_url) + return mask_url def _real_extract(self, url): post_id = self._match_id(url) - wall_url = 'https://vk.com/wall%s' % post_id - - post_id = remove_start(post_id, '-') - - webpage = self._download_webpage(wall_url, post_id) - - error = self._html_search_regex( - r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)', - webpage, 'error', default=None) - if error: - raise ExtractorError('VK said: %s' % error, expected=True) + webpage = self._download_payload('wkview', post_id, { + 'act': 'show', + 'w': 'wall' + post_id, + })[1] description = clean_html(get_element_by_class('wall_post_text', webpage)) uploader = clean_html(get_element_by_class('author', webpage)) - thumbnail = self._og_search_thumbnail(webpage) entries = [] - audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage) - if audio_ids: - al_audio = self._download_webpage( - 'https://vk.com/al_audio.php', post_id, - note='Downloading audio info', fatal=False, - data=urlencode_postdata({ - 'act': 'reload_audio', - 'al': '1', - 'ids': ','.join(audio_ids) - })) - if al_audio: - Audio = collections.namedtuple( - 'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration']) - audios = self._parse_json( - self._search_regex( - r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'), - post_id, fatal=False, transform_source=unescapeHTML) - if isinstance(audios, list): - for audio in audios: - a = Audio._make(audio[:6]) - entries.append({ - 'id': '%s_%s' % (a.user_id, a.id), - 'url': a.url, - 'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id, - 'thumbnail': thumbnail, - 'duration': a.duration, - 'uploader': uploader, - 'artist': a.artist, - 'track': a.track, - }) + for audio in re.findall(r'data-audio="([^"]+)', webpage): + audio = self._parse_json(unescapeHTML(audio), post_id) + a = self._AUDIO._make(audio[:16]) + if not a.url: + continue + title = unescapeHTML(a.title) + performer = unescapeHTML(a.performer) + entries.append({ + 'id': '%s_%s' % (a.owner_id, a.id), + 'url': self._unmask_url(a.url, a.ads['vk_id']), + 'title': '%s - %s' % (performer, title) if performer else title, + 'thumbnails': [{'url': c_url} for c_url in a.cover_url.split(',')] if a.cover_url else None, + 'duration': int_or_none(a.duration), + 'uploader': uploader, + 'artist': performer, + 'track': title, + 'ext': 'mp4', + 'protocol': 'm3u8', + }) for video in re.finditer( r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage): diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py index c3429f723..f79531e6f 100644 --- a/youtube_dl/extractor/vlive.py +++ b/youtube_dl/extractor/vlive.py @@ -6,22 +6,18 @@ import time import itertools from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlencode, - compat_str, -) +from .naver import NaverBaseIE +from ..compat import compat_str from ..utils import ( - dict_get, ExtractorError, - float_or_none, - int_or_none, + merge_dicts, remove_start, try_get, urlencode_postdata, ) -class VLiveIE(InfoExtractor): +class VLiveIE(NaverBaseIE): IE_NAME = 'vlive' _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)' _NETRC_MACHINE = 'vlive' @@ -34,6 +30,7 @@ class VLiveIE(InfoExtractor): 'title': "[V LIVE] Girl's Day's Broadcast", 'creator': "Girl's Day", 'view_count': int, + 'uploader_id': 'muploader_a', }, }, { 'url': 'http://www.vlive.tv/video/16937', @@ -44,6 +41,7 @@ class VLiveIE(InfoExtractor): 'creator': 'EXO', 'view_count': int, 'subtitles': 'mincount:12', + 'uploader_id': 'muploader_j', }, 'params': { 'skip_download': True, @@ -187,45 +185,9 @@ class VLiveIE(InfoExtractor): 'This video is only available for CH+ subscribers') long_video_id, key = video_info['vid'], video_info['inkey'] - playinfo = self._download_json( - 'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s' - % compat_urllib_parse_urlencode({ - 'videoId': long_video_id, - 'key': key, - 'ptc': 'http', - 'doct': 'json', # document type (xml or json) - 'cpt': 'vtt', # captions type (vtt or ttml) - }), video_id) - - formats = [{ - 'url': vid['source'], - 'format_id': vid.get('encodingOption', {}).get('name'), - 'abr': float_or_none(vid.get('bitrate', {}).get('audio')), - 'vbr': float_or_none(vid.get('bitrate', {}).get('video')), - 'width': int_or_none(vid.get('encodingOption', {}).get('width')), - 'height': int_or_none(vid.get('encodingOption', {}).get('height')), - 'filesize': int_or_none(vid.get('size')), - } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')] - self._sort_formats(formats) - - view_count = int_or_none(playinfo.get('meta', {}).get('count')) - - subtitles = {} - for caption in playinfo.get('captions', {}).get('list', []): - lang = dict_get(caption, ('locale', 'language', 'country', 'label')) - if lang and caption.get('source'): - subtitles[lang] = [{ - 'ext': 'vtt', - 'url': caption['source']}] - - info = self._get_common_fields(webpage) - info.update({ - 'id': video_id, - 'formats': formats, - 'view_count': view_count, - 'subtitles': subtitles, - }) - return info + return merge_dicts( + self._get_common_fields(webpage), + self._extract_video_info(video_id, long_video_id, key)) def _download_init_page(self, video_id): return self._download_webpage( diff --git a/youtube_dl/extractor/vodplatform.py b/youtube_dl/extractor/vodplatform.py index 239644340..74d2257e7 100644 --- a/youtube_dl/extractor/vodplatform.py +++ b/youtube_dl/extractor/vodplatform.py @@ -6,8 +6,8 @@ from ..utils import unescapeHTML class VODPlatformIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vod-platform\.net/[eE]mbed/(?P<id>[^/?#]+)' - _TEST = { + _VALID_URL = r'https?://(?:(?:www\.)?vod-platform\.net|embed\.kwikmotion\.com)/[eE]mbed/(?P<id>[^/?#]+)' + _TESTS = [{ # from http://www.lbcgroup.tv/watch/chapter/29143/52844/%D8%A7%D9%84%D9%86%D8%B5%D8%B1%D8%A9-%D9%81%D9%8A-%D8%B6%D9%8A%D8%A7%D9%81%D8%A9-%D8%A7%D9%84%D9%80-cnn/ar 'url': 'http://vod-platform.net/embed/RufMcytHDolTH1MuKHY9Fw', 'md5': '1db2b7249ce383d6be96499006e951fc', @@ -16,7 +16,10 @@ class VODPlatformIE(InfoExtractor): 'ext': 'mp4', 'title': 'LBCi News_ النصرة في ضيافة الـ "سي.أن.أن"', } - } + }, { + 'url': 'http://embed.kwikmotion.com/embed/RufMcytHDolTH1MuKHY9Fw', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) diff --git a/youtube_dl/extractor/voicerepublic.py b/youtube_dl/extractor/voicerepublic.py index 59e1359c4..a52e40afa 100644 --- a/youtube_dl/extractor/voicerepublic.py +++ b/youtube_dl/extractor/voicerepublic.py @@ -1,17 +1,12 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor -from ..compat import ( - compat_str, - compat_urlparse, -) +from ..compat import compat_str from ..utils import ( ExtractorError, determine_ext, int_or_none, - sanitized_Request, + urljoin, ) @@ -26,8 +21,7 @@ class VoiceRepublicIE(InfoExtractor): 'ext': 'm4a', 'title': 'Watching the Watchers: Building a Sousveillance State', 'description': 'Secret surveillance programs have metadata too. The people and companies that operate secret surveillance programs can be surveilled.', - 'thumbnail': r're:^https?://.*\.(?:png|jpg)$', - 'duration': 1800, + 'duration': 1556, 'view_count': int, } }, { @@ -38,63 +32,31 @@ class VoiceRepublicIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) - req = sanitized_Request( - compat_urlparse.urljoin(url, '/talks/%s' % display_id)) - # Older versions of Firefox get redirected to an "upgrade browser" page - req.add_header('User-Agent', 'youtube-dl') - webpage = self._download_webpage(req, display_id) + webpage = self._download_webpage(url, display_id) if '>Queued for processing, please stand by...<' in webpage: raise ExtractorError( 'Audio is still queued for processing', expected=True) - config = self._search_regex( - r'(?s)return ({.+?});\s*\n', webpage, - 'data', default=None) - data = self._parse_json(config, display_id, fatal=False) if config else None - if data: - title = data['title'] - description = data.get('teaser') - talk_id = compat_str(data.get('talk_id') or display_id) - talk = data['talk'] - duration = int_or_none(talk.get('duration')) - formats = [{ - 'url': compat_urlparse.urljoin(url, talk_url), - 'format_id': format_id, - 'ext': determine_ext(talk_url) or format_id, - 'vcodec': 'none', - } for format_id, talk_url in talk['links'].items()] - else: - title = self._og_search_title(webpage) - description = self._html_search_regex( - r"(?s)<div class='talk-teaser'[^>]*>(.+?)</div>", - webpage, 'description', fatal=False) - talk_id = self._search_regex( - [r"id='jc-(\d+)'", r"data-shareable-id='(\d+)'"], - webpage, 'talk id', default=None) or display_id - duration = None - player = self._search_regex( - r"class='vr-player jp-jplayer'([^>]+)>", webpage, 'player') - formats = [{ - 'url': compat_urlparse.urljoin(url, talk_url), - 'format_id': format_id, - 'ext': determine_ext(talk_url) or format_id, - 'vcodec': 'none', - } for format_id, talk_url in re.findall(r"data-([^=]+)='([^']+)'", player)] + talk = self._parse_json(self._search_regex( + r'initialSnapshot\s*=\s*({.+?});', + webpage, 'talk'), display_id)['talk'] + title = talk['title'] + formats = [{ + 'url': urljoin(url, talk_url), + 'format_id': format_id, + 'ext': determine_ext(talk_url) or format_id, + 'vcodec': 'none', + } for format_id, talk_url in talk['media_links'].items()] self._sort_formats(formats) - thumbnail = self._og_search_thumbnail(webpage) - view_count = int_or_none(self._search_regex( - r"class='play-count[^']*'>\s*(\d+) plays", - webpage, 'play count', fatal=False)) - return { - 'id': talk_id, + 'id': compat_str(talk.get('id') or display_id), 'display_id': display_id, 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'duration': duration, - 'view_count': view_count, + 'description': talk.get('teaser'), + 'thumbnail': talk.get('image_url'), + 'duration': int_or_none(talk.get('archived_duration')), + 'view_count': int_or_none(talk.get('play_count')), 'formats': formats, } diff --git a/youtube_dl/extractor/voxmedia.py b/youtube_dl/extractor/voxmedia.py index c7a0a88fe..b318e15d4 100644 --- a/youtube_dl/extractor/voxmedia.py +++ b/youtube_dl/extractor/voxmedia.py @@ -4,7 +4,10 @@ from __future__ import unicode_literals from .common import InfoExtractor from .once import OnceIE from ..compat import compat_urllib_parse_unquote -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + int_or_none, +) class VoxMediaVolumeIE(OnceIE): @@ -13,18 +16,43 @@ class VoxMediaVolumeIE(OnceIE): def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - video_data = self._parse_json(self._search_regex( - r'Volume\.createVideo\(({.+})\s*,\s*{.*}\s*,\s*\[.*\]\s*,\s*{.*}\);', webpage, 'video data'), video_id) + + setup = self._parse_json(self._search_regex( + r'setup\s*=\s*({.+});', webpage, 'setup'), video_id) + video_data = setup.get('video') or {} + info = { + 'id': video_id, + 'title': video_data.get('title_short'), + 'description': video_data.get('description_long') or video_data.get('description_short'), + 'thumbnail': video_data.get('brightcove_thumbnail') + } + asset = setup.get('asset') or setup.get('params') or {} + + formats = [] + hls_url = asset.get('hls_url') + if hls_url: + formats.extend(self._extract_m3u8_formats( + hls_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + mp4_url = asset.get('mp4_url') + if mp4_url: + tbr = self._search_regex(r'-(\d+)k\.', mp4_url, 'bitrate', default=None) + format_id = 'http' + if tbr: + format_id += '-' + tbr + formats.append({ + 'format_id': format_id, + 'url': mp4_url, + 'tbr': int_or_none(tbr), + }) + if formats: + self._sort_formats(formats) + info['formats'] = formats + return info + for provider_video_type in ('ooyala', 'youtube', 'brightcove'): provider_video_id = video_data.get('%s_id' % provider_video_type) if not provider_video_id: continue - info = { - 'id': video_id, - 'title': video_data.get('title_short'), - 'description': video_data.get('description_long') or video_data.get('description_short'), - 'thumbnail': video_data.get('brightcove_thumbnail') - } if provider_video_type == 'brightcove': info['formats'] = self._extract_once_formats(provider_video_id) self._sort_formats(info['formats']) @@ -39,46 +67,49 @@ class VoxMediaVolumeIE(OnceIE): class VoxMediaIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?:(?:theverge|vox|sbnation|eater|polygon|curbed|racked)\.com|recode\.net)/(?:[^/]+/)*(?P<id>[^/?]+)' + _VALID_URL = r'https?://(?:www\.)?(?:(?:theverge|vox|sbnation|eater|polygon|curbed|racked|funnyordie)\.com|recode\.net)/(?:[^/]+/)*(?P<id>[^/?]+)' _TESTS = [{ + # Volume embed, Youtube 'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of', 'info_dict': { - 'id': '11eXZobjrG8DCSTgrNjVinU-YmmdYjhe', + 'id': 'j4mLW6x17VM', 'ext': 'mp4', - 'title': 'Google\'s new material design direction', - 'description': 'md5:2f44f74c4d14a1f800ea73e1c6832ad2', + 'title': 'Material world: how Google discovered what software is made of', + 'description': 'md5:dfc17e7715e3b542d66e33a109861382', + 'upload_date': '20190710', + 'uploader_id': 'TheVerge', + 'uploader': 'The Verge', }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - 'add_ie': ['Ooyala'], + 'add_ie': ['Youtube'], }, { - # data-ooyala-id + # Volume embed, Youtube 'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet', - 'md5': 'd744484ff127884cd2ba09e3fa604e4b', + 'md5': '4c8f4a0937752b437c3ebc0ed24802b5', 'info_dict': { - 'id': 'RkZXU4cTphOCPDMZg5oEounJyoFI0g-B', + 'id': 'Gy8Md3Eky38', 'ext': 'mp4', 'title': 'The Nexus 6: hands-on with Google\'s phablet', - 'description': 'md5:87a51fe95ff8cea8b5bdb9ac7ae6a6af', + 'description': 'md5:d9f0216e5fb932dd2033d6db37ac3f1d', + 'uploader_id': 'TheVerge', + 'upload_date': '20141021', + 'uploader': 'The Verge', }, - 'add_ie': ['Ooyala'], - 'skip': 'Video Not Found', + 'add_ie': ['Youtube'], + 'skip': 'similar to the previous test', }, { - # volume embed + # Volume embed, Youtube 'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill', 'info_dict': { - 'id': 'wydzk3dDpmRz7PQoXRsTIX6XTkPjYL0b', + 'id': 'YCjDnX-Xzhg', 'ext': 'mp4', - 'title': 'The new frontier of LGBTQ civil rights, explained', - 'description': 'md5:0dc58e94a465cbe91d02950f770eb93f', + 'title': "Mississippi's laws are so bad that its anti-LGBTQ law isn't needed to allow discrimination", + 'description': 'md5:fc1317922057de31cd74bce91eb1c66c', + 'uploader_id': 'voxdotcom', + 'upload_date': '20150915', + 'uploader': 'Vox', }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - 'add_ie': ['Ooyala'], + 'add_ie': ['Youtube'], + 'skip': 'similar to the previous test', }, { # youtube embed 'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance', @@ -93,6 +124,7 @@ class VoxMediaIE(InfoExtractor): 'uploader': 'Vox', }, 'add_ie': ['Youtube'], + 'skip': 'Page no longer contain videos', }, { # SBN.VideoLinkset.entryGroup multiple ooyala embeds 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok', @@ -118,10 +150,11 @@ class VoxMediaIE(InfoExtractor): 'description': 'md5:e02d56b026d51aa32c010676765a690d', }, }], + 'skip': 'Page no longer contain videos', }, { # volume embed, Brightcove Once 'url': 'https://www.recode.net/2014/6/17/11628066/post-post-pc-ceo-the-full-code-conference-video-of-microsofts-satya', - 'md5': '01571a896281f77dc06e084138987ea2', + 'md5': '2dbc77b8b0bff1894c2fce16eded637d', 'info_dict': { 'id': '1231c973d', 'ext': 'mp4', diff --git a/youtube_dl/extractor/vrv.py b/youtube_dl/extractor/vrv.py index c814a8a4a..6e51469b0 100644 --- a/youtube_dl/extractor/vrv.py +++ b/youtube_dl/extractor/vrv.py @@ -64,7 +64,15 @@ class VRVBaseIE(InfoExtractor): def _call_cms(self, path, video_id, note): if not self._CMS_SIGNING: - self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing'] + index = self._call_api('index', video_id, 'CMS Signing') + self._CMS_SIGNING = index.get('cms_signing') or {} + if not self._CMS_SIGNING: + for signing_policy in index.get('signing_policies', []): + signing_path = signing_policy.get('path') + if signing_path and signing_path.startswith('/cms/'): + name, value = signing_policy.get('name'), signing_policy.get('value') + if name and value: + self._CMS_SIGNING[name] = value return self._download_json( self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING, note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers()) diff --git a/youtube_dl/extractor/vzaar.py b/youtube_dl/extractor/vzaar.py index 6000671c3..b7d02fca3 100644 --- a/youtube_dl/extractor/vzaar.py +++ b/youtube_dl/extractor/vzaar.py @@ -32,6 +32,22 @@ class VzaarIE(InfoExtractor): 'ext': 'mp3', 'title': 'MP3', }, + }, { + # hlsAes = true + 'url': 'https://view.vzaar.com/11379930/player', + 'info_dict': { + 'id': '11379930', + 'ext': 'mp4', + 'title': 'Videoaula', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + # with null videoTitle + 'url': 'https://view.vzaar.com/20313539/download', + 'only_matching': True, }] @staticmethod @@ -45,7 +61,7 @@ class VzaarIE(InfoExtractor): video_data = self._download_json( 'http://view.vzaar.com/v2/%s/video' % video_id, video_id) - title = video_data['videoTitle'] + title = video_data.get('videoTitle') or video_id formats = [] @@ -54,6 +70,7 @@ class VzaarIE(InfoExtractor): f = { 'url': source_url, 'format_id': 'http', + 'preference': 1, } if 'audio' in source_url: f.update({ @@ -71,13 +88,17 @@ class VzaarIE(InfoExtractor): video_guid = video_data.get('guid') usp = video_data.get('usp') - if isinstance(video_guid, compat_str) and isinstance(usp, dict): - m3u8_url = ('http://fable.vzaar.com/v4/usp/%s/%s.ism/.m3u8?' - % (video_guid, video_id)) + '&'.join( - '%s=%s' % (k, v) for k, v in usp.items()) - formats.extend(self._extract_m3u8_formats( - m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native', - m3u8_id='hls', fatal=False)) + if video_data.get('uspEnabled') and isinstance(video_guid, compat_str) and isinstance(usp, dict): + hls_aes = video_data.get('hlsAes') + qs = '&'.join('%s=%s' % (k, v) for k, v in usp.items()) + url_templ = 'http://%%s.vzaar.com/v5/usp%s/%s/%s.ism%%s?' % ('aes' if hls_aes else '', video_guid, video_id) + m3u8_formats = self._extract_m3u8_formats( + url_templ % ('fable', '/.m3u8') + qs, video_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False) + if hls_aes: + for f in m3u8_formats: + f['_decryption_key_url'] = url_templ % ('goose', '') + qs + formats.extend(m3u8_formats) self._sort_formats(formats) diff --git a/youtube_dl/extractor/wimp.py b/youtube_dl/extractor/wimp.py deleted file mode 100644 index ea234e3c5..000000000 --- a/youtube_dl/extractor/wimp.py +++ /dev/null @@ -1,54 +0,0 @@ -from __future__ import unicode_literals - -from .common import InfoExtractor -from .youtube import YoutubeIE - - -class WimpIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?wimp\.com/(?P<id>[^/]+)' - _TESTS = [{ - 'url': 'http://www.wimp.com/maru-is-exhausted/', - 'md5': 'ee21217ffd66d058e8b16be340b74883', - 'info_dict': { - 'id': 'maru-is-exhausted', - 'ext': 'mp4', - 'title': 'Maru is exhausted.', - 'description': 'md5:57e099e857c0a4ea312542b684a869b8', - } - }, { - 'url': 'http://www.wimp.com/clowncar/', - 'md5': '5c31ad862a90dc5b1f023956faec13fe', - 'info_dict': { - 'id': 'cG4CEr2aiSg', - 'ext': 'webm', - 'title': 'Basset hound clown car...incredible!', - 'description': '5 of my Bassets crawled in this dog loo! www.bellinghambassets.com\n\nFor licensing/usage please contact: licensing(at)jukinmediadotcom', - 'upload_date': '20140303', - 'uploader': 'Gretchen Hoey', - 'uploader_id': 'gretchenandjeff1', - }, - 'add_ie': ['Youtube'], - }] - - def _real_extract(self, url): - video_id = self._match_id(url) - - webpage = self._download_webpage(url, video_id) - - youtube_id = self._search_regex( - (r"videoId\s*:\s*[\"']([0-9A-Za-z_-]{11})[\"']", - r'data-id=["\']([0-9A-Za-z_-]{11})'), - webpage, 'video URL', default=None) - if youtube_id: - return self.url_result(youtube_id, YoutubeIE.ie_key()) - - info_dict = self._extract_jwplayer_data( - webpage, video_id, require_title=False) - - info_dict.update({ - 'id': video_id, - 'title': self._og_search_title(webpage), - 'description': self._og_search_description(webpage), - }) - - return info_dict diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py index fa142b974..168e5e901 100644 --- a/youtube_dl/extractor/wistia.py +++ b/youtube_dl/extractor/wistia.py @@ -12,9 +12,8 @@ from ..utils import ( class WistiaIE(InfoExtractor): - _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/)(?P<id>[a-z0-9]+)' - _API_URL = 'http://fast.wistia.com/embed/medias/%s.json' - _IFRAME_URL = 'http://fast.wistia.net/embed/iframe/%s' + _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/)(?P<id>[a-z0-9]{10})' + _EMBED_BASE_URL = 'http://fast.wistia.com/embed/' _TESTS = [{ 'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt', @@ -43,33 +42,35 @@ class WistiaIE(InfoExtractor): 'only_matching': True, }] + # https://wistia.com/support/embed-and-share/video-on-your-website @staticmethod def _extract_url(webpage): - match = re.search( - r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/iframe/.+?)\1', webpage) - if match: - return unescapeHTML(match.group('url')) + urls = WistiaIE._extract_urls(webpage) + return urls[0] if urls else None - match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage) - if match: - return 'wistia:%s' % match.group('id') - - match = re.search( - r'''(?sx) - <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*? - <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2 - ''', webpage) - if match: - return 'wistia:%s' % match.group('id') + @staticmethod + def _extract_urls(webpage): + urls = [] + for match in re.finditer( + r'<(?:meta[^>]+?content|(?:iframe|script)[^>]+?src)=["\'](?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/[a-z0-9]{10})', webpage): + urls.append(unescapeHTML(match.group('url'))) + for match in re.finditer( + r'''(?sx) + <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]{10})\b.*?\2 + ''', webpage): + urls.append('wistia:%s' % match.group('id')) + for match in re.finditer(r'(?:data-wistia-?id=["\']|Wistia\.embed\(["\']|id=["\']wistia_)(?P<id>[a-z0-9]{10})', webpage): + urls.append('wistia:%s' % match.group('id')) + return urls def _real_extract(self, url): video_id = self._match_id(url) data_json = self._download_json( - self._API_URL % video_id, video_id, + self._EMBED_BASE_URL + 'medias/%s.json' % video_id, video_id, # Some videos require this. headers={ - 'Referer': url if url.startswith('http') else self._IFRAME_URL % video_id, + 'Referer': url if url.startswith('http') else self._EMBED_BASE_URL + 'iframe/' + video_id, }) if data_json.get('error'): @@ -94,27 +95,61 @@ class WistiaIE(InfoExtractor): 'url': aurl, 'width': int_or_none(a.get('width')), 'height': int_or_none(a.get('height')), + 'filesize': int_or_none(a.get('size')), }) else: aext = a.get('ext') - is_m3u8 = a.get('container') == 'm3u8' or aext == 'm3u8' - formats.append({ - 'format_id': atype, + display_name = a.get('display_name') + format_id = atype + if atype and atype.endswith('_video') and display_name: + format_id = '%s-%s' % (atype[:-6], display_name) + f = { + 'format_id': format_id, 'url': aurl, - 'tbr': int_or_none(a.get('bitrate')), - 'vbr': int_or_none(a.get('opt_vbitrate')), - 'width': int_or_none(a.get('width')), - 'height': int_or_none(a.get('height')), - 'filesize': int_or_none(a.get('size')), - 'vcodec': a.get('codec'), - 'container': a.get('container'), - 'ext': 'mp4' if is_m3u8 else aext, - 'protocol': 'm3u8' if is_m3u8 else None, + 'tbr': int_or_none(a.get('bitrate')) or None, 'preference': 1 if atype == 'original' else None, - }) + } + if display_name == 'Audio': + f.update({ + 'vcodec': 'none', + }) + else: + f.update({ + 'width': int_or_none(a.get('width')), + 'height': int_or_none(a.get('height')), + 'vcodec': a.get('codec'), + }) + if a.get('container') == 'm3u8' or aext == 'm3u8': + ts_f = f.copy() + ts_f.update({ + 'ext': 'ts', + 'format_id': f['format_id'].replace('hls-', 'ts-'), + 'url': f['url'].replace('.bin', '.ts'), + }) + formats.append(ts_f) + f.update({ + 'ext': 'mp4', + 'protocol': 'm3u8_native', + }) + else: + f.update({ + 'container': a.get('container'), + 'ext': aext, + 'filesize': int_or_none(a.get('size')), + }) + formats.append(f) self._sort_formats(formats) + subtitles = {} + for caption in data.get('captions', []): + language = caption.get('language') + if not language: + continue + subtitles[language] = [{ + 'url': self._EMBED_BASE_URL + 'captions/' + video_id + '.vtt?language=' + language, + }] + return { 'id': video_id, 'title': title, @@ -123,4 +158,5 @@ class WistiaIE(InfoExtractor): 'thumbnails': thumbnails, 'duration': float_or_none(data.get('duration')), 'timestamp': int_or_none(data.get('createdAt')), + 'subtitles': subtitles, } diff --git a/youtube_dl/extractor/xfileshare.py b/youtube_dl/extractor/xfileshare.py index b38c7a7b3..48ef07ed1 100644 --- a/youtube_dl/extractor/xfileshare.py +++ b/youtube_dl/extractor/xfileshare.py @@ -4,37 +4,64 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_chr from ..utils import ( decode_packed_codes, determine_ext, ExtractorError, int_or_none, - NO_DEFAULT, + js_to_json, urlencode_postdata, ) +# based on openload_decode from 2bfeee69b976fe049761dd3012e30b637ee05a58 +def aa_decode(aa_code): + symbol_table = [ + ('7', '((ﾟｰﾟ) + (o^_^o))'), + ('6', '((o^_^o) +(o^_^o))'), + ('5', '((ﾟｰﾟ) + (ﾟΘﾟ))'), + ('2', '((o^_^o) - (ﾟΘﾟ))'), + ('4', '(ﾟｰﾟ)'), + ('3', '(o^_^o)'), + ('1', '(ﾟΘﾟ)'), + ('0', '(c^_^o)'), + ] + delim = '(ﾟДﾟ)[ﾟεﾟ]+' + ret = '' + for aa_char in aa_code.split(delim): + for val, pat in symbol_table: + aa_char = aa_char.replace(pat, val) + aa_char = aa_char.replace('+ ', '') + m = re.match(r'^\d+', aa_char) + if m: + ret += compat_chr(int(m.group(0), 8)) + else: + m = re.match(r'^u([\da-f]+)', aa_char) + if m: + ret += compat_chr(int(m.group(1), 16)) + return ret + + class XFileShareIE(InfoExtractor): _SITES = ( - (r'daclips\.(?:in|com)', 'DaClips'), - (r'filehoot\.com', 'FileHoot'), - (r'gorillavid\.(?:in|com)', 'GorillaVid'), - (r'movpod\.in', 'MovPod'), - (r'powerwatch\.pw', 'PowerWatch'), - (r'rapidvideo\.ws', 'Rapidvideo.ws'), + (r'clipwatching\.com', 'ClipWatching'), + (r'gounlimited\.to', 'GoUnlimited'), + (r'govid\.me', 'GoVid'), + (r'holavid\.com', 'HolaVid'), + (r'streamty\.com', 'Streamty'), (r'thevideobee\.to', 'TheVideoBee'), - (r'vidto\.(?:me|se)', 'Vidto'), - (r'streamin\.to', 'Streamin.To'), - (r'xvidstage\.com', 'XVIDSTAGE'), - (r'vidabc\.com', 'Vid ABC'), + (r'uqload\.com', 'Uqload'), (r'vidbom\.com', 'VidBom'), (r'vidlo\.us', 'vidlo'), - (r'rapidvideo\.(?:cool|org)', 'RapidVideo.TV'), - (r'fastvideo\.me', 'FastVideo.me'), + (r'vidlocker\.xyz', 'VidLocker'), + (r'vidshare\.tv', 'VidShare'), + (r'vup\.to', 'VUp'), + (r'xvideosharing\.com', 'XVideoSharing'), ) IE_DESC = 'XFileShare based sites: %s' % ', '.join(list(zip(*_SITES))[1]) - _VALID_URL = (r'https?://(?P<host>(?:www\.)?(?:%s))/(?:embed-)?(?P<id>[0-9a-zA-Z]+)' + _VALID_URL = (r'https?://(?:www\.)?(?P<host>%s)/(?:embed-)?(?P<id>[0-9a-zA-Z]+)' % '|'.join(site for site in list(zip(*_SITES))[0])) _FILE_NOT_FOUND_REGEXES = ( @@ -43,82 +70,14 @@ class XFileShareIE(InfoExtractor): ) _TESTS = [{ - 'url': 'http://gorillavid.in/06y9juieqpmi', - 'md5': '5ae4a3580620380619678ee4875893ba', + 'url': 'http://xvideosharing.com/fq65f94nd2ve', + 'md5': '4181f63957e8fe90ac836fa58dc3c8a6', 'info_dict': { - 'id': '06y9juieqpmi', + 'id': 'fq65f94nd2ve', 'ext': 'mp4', - 'title': 'Rebecca Black My Moment Official Music Video Reaction-6GK87Rc8bzQ', + 'title': 'sample', 'thumbnail': r're:http://.*\.jpg', }, - }, { - 'url': 'http://gorillavid.in/embed-z08zf8le23c6-960x480.html', - 'only_matching': True, - }, { - 'url': 'http://daclips.in/3rso4kdn6f9m', - 'md5': '1ad8fd39bb976eeb66004d3a4895f106', - 'info_dict': { - 'id': '3rso4kdn6f9m', - 'ext': 'mp4', - 'title': 'Micro Pig piglets ready on 16th July 2009-bG0PdrCdxUc', - 'thumbnail': r're:http://.*\.jpg', - } - }, { - 'url': 'http://movpod.in/0wguyyxi1yca', - 'only_matching': True, - }, { - 'url': 'http://filehoot.com/3ivfabn7573c.html', - 'info_dict': { - 'id': '3ivfabn7573c', - 'ext': 'mp4', - 'title': 'youtube-dl test video \'äBaW_jenozKc.mp4.mp4', - 'thumbnail': r're:http://.*\.jpg', - }, - 'skip': 'Video removed', - }, { - 'url': 'http://vidto.me/ku5glz52nqe1.html', - 'info_dict': { - 'id': 'ku5glz52nqe1', - 'ext': 'mp4', - 'title': 'test' - } - }, { - 'url': 'http://powerwatch.pw/duecjibvicbu', - 'info_dict': { - 'id': 'duecjibvicbu', - 'ext': 'mp4', - 'title': 'Big Buck Bunny trailer', - }, - }, { - 'url': 'http://xvidstage.com/e0qcnl03co6z', - 'info_dict': { - 'id': 'e0qcnl03co6z', - 'ext': 'mp4', - 'title': 'Chucky Prank 2015.mp4', - }, - }, { - # removed by administrator - 'url': 'http://xvidstage.com/amfy7atlkx25', - 'only_matching': True, - }, { - 'url': 'http://vidabc.com/i8ybqscrphfv', - 'info_dict': { - 'id': 'i8ybqscrphfv', - 'ext': 'mp4', - 'title': 're:Beauty and the Beast 2017', - }, - 'params': { - 'skip_download': True, - }, - }, { - 'url': 'http://www.rapidvideo.cool/b667kprndr8w', - 'only_matching': True, - }, { - 'url': 'http://www.fastvideo.me/k8604r8nk8sn/FAST_FURIOUS_8_-_Trailer_italiano_ufficiale.mp4.html', - 'only_matching': True, - }, { - 'url': 'http://vidto.se/1tx1pf6t12cg.html', - 'only_matching': True, }] @staticmethod @@ -131,10 +90,9 @@ class XFileShareIE(InfoExtractor): webpage)] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + host, video_id = re.match(self._VALID_URL, url).groups() - url = 'http://%s/%s' % (mobj.group('host'), video_id) + url = 'https://%s/' % host + ('embed-%s.html' % video_id if host in ('govid.me', 'vidlo.us') else video_id) webpage = self._download_webpage(url, video_id) if any(re.search(p, webpage) for p in self._FILE_NOT_FOUND_REGEXES): @@ -142,7 +100,7 @@ class XFileShareIE(InfoExtractor): fields = self._hidden_inputs(webpage) - if fields['op'] == 'download1': + if fields.get('op') == 'download1': countdown = int_or_none(self._search_regex( r'<span id="countdown_str">(?:[Ww]ait)?\s*<span id="cxc">(\d+)</span>\s*(?:seconds?)?</span>', webpage, 'countdown', default=None)) @@ -160,13 +118,37 @@ class XFileShareIE(InfoExtractor): (r'style="z-index: [0-9]+;">([^<]+)</span>', r'<td nowrap>([^<]+)</td>', r'h4-fine[^>]*>([^<]+)<', - r'>Watch (.+) ', + r'>Watch (.+)[ <]', r'<h2 class="video-page-head">([^<]+)</h2>', - r'<h2 style="[^"]*color:#403f3d[^"]*"[^>]*>([^<]+)<'), # streamin.to + r'<h2 style="[^"]*color:#403f3d[^"]*"[^>]*>([^<]+)<', # streamin.to + r'title\s*:\s*"([^"]+)"'), # govid.me webpage, 'title', default=None) or self._og_search_title( webpage, default=None) or video_id).strip() - def extract_formats(default=NO_DEFAULT): + for regex, func in ( + (r'(eval\(function\(p,a,c,k,e,d\){.+)', decode_packed_codes), + (r'(ﾟ.+)', aa_decode)): + obf_code = self._search_regex(regex, webpage, 'obfuscated code', default=None) + if obf_code: + webpage = webpage.replace(obf_code, func(obf_code)) + + formats = [] + + jwplayer_data = self._search_regex( + [ + r'jwplayer\("[^"]+"\)\.load\(\[({.+?})\]\);', + r'jwplayer\("[^"]+"\)\.setup\(({.+?})\);', + ], webpage, + 'jwplayer data', default=None) + if jwplayer_data: + jwplayer_data = self._parse_json( + jwplayer_data.replace(r"\'", "'"), video_id, js_to_json) + if jwplayer_data: + formats = self._parse_jwplayer_data( + jwplayer_data, video_id, False, + m3u8_id='hls', mpd_id='dash')['formats'] + + if not formats: urls = [] for regex in ( r'(?:file|src)\s*:\s*(["\'])(?P<url>http(?:(?!\1).)+\.(?:m3u8|mp4|flv)(?:(?!\1).)*)\1', @@ -177,6 +159,12 @@ class XFileShareIE(InfoExtractor): video_url = mobj.group('url') if video_url not in urls: urls.append(video_url) + + sources = self._search_regex( + r'sources\s*:\s*(\[(?!{)[^\]]+\])', webpage, 'sources', default=None) + if sources: + urls.extend(self._parse_json(sources, video_id)) + formats = [] for video_url in urls: if determine_ext(video_url) == 'm3u8': @@ -189,21 +177,13 @@ class XFileShareIE(InfoExtractor): 'url': video_url, 'format_id': 'sd', }) - if not formats and default is not NO_DEFAULT: - return default - self._sort_formats(formats) - return formats - - formats = extract_formats(default=None) - - if not formats: - webpage = decode_packed_codes(self._search_regex( - r"(}\('(.+)',(\d+),(\d+),'[^']*\b(?:file|embed)\b[^']*'\.split\('\|'\))", - webpage, 'packed code')) - formats = extract_formats() + self._sort_formats(formats) thumbnail = self._search_regex( - r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', default=None) + [ + r'<video[^>]+poster="([^"]+)"', + r'(?:image|poster)\s*:\s*["\'](http[^"\']+)["\'],', + ], webpage, 'thumbnail', default=None) return { 'id': video_id, diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index d268372e6..0f7be6a7d 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -1,5 +1,6 @@ from __future__ import unicode_literals +import itertools import re from .common import InfoExtractor @@ -8,6 +9,7 @@ from ..utils import ( clean_html, determine_ext, dict_get, + extract_attributes, ExtractorError, int_or_none, parse_duration, @@ -18,21 +20,21 @@ from ..utils import ( class XHamsterIE(InfoExtractor): + _DOMAINS = r'(?:xhamster\.(?:com|one|desi)|xhms\.pro|xhamster[27]\.com)' _VALID_URL = r'''(?x) https?:// - (?:.+?\.)?xhamster\.(?:com|one)/ + (?:.+?\.)?%s/ (?: movies/(?P<id>\d+)/(?P<display_id>[^/]*)\.html| videos/(?P<display_id_2>[^/]*)-(?P<id_2>\d+) ) - ''' - + ''' % _DOMAINS _TESTS = [{ - 'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html', - 'md5': '8281348b8d3c53d39fffb377d24eac4e', + 'url': 'https://xhamster.com/videos/femaleagent-shy-beauty-takes-the-bait-1509445', + 'md5': '98b4687efb1ffd331c4197854dc09e8f', 'info_dict': { 'id': '1509445', - 'display_id': 'femaleagent_shy_beauty_takes_the_bait', + 'display_id': 'femaleagent-shy-beauty-takes-the-bait', 'ext': 'mp4', 'title': 'FemaleAgent Shy beauty takes the bait', 'timestamp': 1350194821, @@ -40,13 +42,12 @@ class XHamsterIE(InfoExtractor): 'uploader': 'Ruseful2011', 'duration': 893, 'age_limit': 18, - 'categories': ['Fake Hub', 'Amateur', 'MILFs', 'POV', 'Beauti', 'Beauties', 'Beautiful', 'Boss', 'Office', 'Oral', 'Reality', 'Sexy', 'Taking'], }, }, { - 'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd', + 'url': 'https://xhamster.com/videos/britney-spears-sexy-booty-2221348?hd=', 'info_dict': { 'id': '2221348', - 'display_id': 'britney_spears_sexy_booty', + 'display_id': 'britney-spears-sexy-booty', 'ext': 'mp4', 'title': 'Britney Spears Sexy Booty', 'timestamp': 1379123460, @@ -54,13 +55,12 @@ class XHamsterIE(InfoExtractor): 'uploader': 'jojo747400', 'duration': 200, 'age_limit': 18, - 'categories': ['Britney Spears', 'Celebrities', 'HD Videos', 'Sexy', 'Sexy Booty'], }, 'params': { 'skip_download': True, }, }, { - # empty seo + # empty seo, unavailable via new URL schema 'url': 'http://xhamster.com/movies/5667973/.html', 'info_dict': { 'id': '5667973', @@ -71,7 +71,6 @@ class XHamsterIE(InfoExtractor): 'uploader': 'parejafree', 'duration': 72, 'age_limit': 18, - 'categories': ['Amateur', 'Blowjobs'], }, 'params': { 'skip_download': True, @@ -94,6 +93,18 @@ class XHamsterIE(InfoExtractor): }, { 'url': 'https://xhamster.one/videos/femaleagent-shy-beauty-takes-the-bait-1509445', 'only_matching': True, + }, { + 'url': 'https://xhamster.desi/videos/femaleagent-shy-beauty-takes-the-bait-1509445', + 'only_matching': True, + }, { + 'url': 'https://xhamster2.com/videos/femaleagent-shy-beauty-takes-the-bait-1509445', + 'only_matching': True, + }, { + 'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html', + 'only_matching': True, + }, { + 'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd', + 'only_matching': True, }] def _real_extract(self, url): @@ -102,7 +113,7 @@ class XHamsterIE(InfoExtractor): display_id = mobj.group('display_id') or mobj.group('display_id_2') desktop_url = re.sub(r'^(https?://(?:.+?\.)?)m\.', r'\1', url) - webpage = self._download_webpage(desktop_url, video_id) + webpage, urlh = self._download_webpage_handle(desktop_url, video_id) error = self._html_search_regex( r'<div[^>]+id=["\']videoClosed["\'][^>]*>(.+?)</div>', @@ -150,6 +161,9 @@ class XHamsterIE(InfoExtractor): 'ext': determine_ext(format_url, 'mp4'), 'height': get_height(quality), 'filesize': filesize, + 'http_headers': { + 'Referer': urlh.geturl(), + }, }) self._sort_formats(formats) @@ -285,7 +299,7 @@ class XHamsterIE(InfoExtractor): class XHamsterEmbedIE(InfoExtractor): - _VALID_URL = r'https?://(?:.+?\.)?xhamster\.com/xembed\.php\?video=(?P<id>\d+)' + _VALID_URL = r'https?://(?:.+?\.)?%s/xembed\.php\?video=(?P<id>\d+)' % XHamsterIE._DOMAINS _TEST = { 'url': 'http://xhamster.com/xembed.php?video=3328539', 'info_dict': { @@ -322,3 +336,49 @@ class XHamsterEmbedIE(InfoExtractor): video_url = dict_get(vars, ('downloadLink', 'homepageLink', 'commentsLink', 'shareUrl')) return self.url_result(video_url, 'XHamster') + + +class XHamsterUserIE(InfoExtractor): + _VALID_URL = r'https?://(?:.+?\.)?%s/users/(?P<id>[^/?#&]+)' % XHamsterIE._DOMAINS + _TESTS = [{ + # Paginated user profile + 'url': 'https://xhamster.com/users/netvideogirls/videos', + 'info_dict': { + 'id': 'netvideogirls', + }, + 'playlist_mincount': 267, + }, { + # Non-paginated user profile + 'url': 'https://xhamster.com/users/firatkaan/videos', + 'info_dict': { + 'id': 'firatkaan', + }, + 'playlist_mincount': 1, + }] + + def _entries(self, user_id): + next_page_url = 'https://xhamster.com/users/%s/videos/1' % user_id + for pagenum in itertools.count(1): + page = self._download_webpage( + next_page_url, user_id, 'Downloading page %s' % pagenum) + for video_tag in re.findall( + r'(<a[^>]+class=["\'].*?\bvideo-thumb__image-container[^>]+>)', + page): + video = extract_attributes(video_tag) + video_url = url_or_none(video.get('href')) + if not video_url or not XHamsterIE.suitable(video_url): + continue + video_id = XHamsterIE._match_id(video_url) + yield self.url_result( + video_url, ie=XHamsterIE.ie_key(), video_id=video_id) + mobj = re.search(r'<a[^>]+data-page=["\']next[^>]+>', page) + if not mobj: + break + next_page = extract_attributes(mobj.group(0)) + next_page_url = url_or_none(next_page.get('href')) + if not next_page_url: + break + + def _real_extract(self, url): + user_id = self._match_id(url) + return self.playlist_result(self._entries(user_id), user_id) diff --git a/youtube_dl/extractor/xiami.py b/youtube_dl/extractor/xiami.py index 8333fb534..618da8382 100644 --- a/youtube_dl/extractor/xiami.py +++ b/youtube_dl/extractor/xiami.py @@ -7,7 +7,7 @@ from ..utils import int_or_none class XiamiBaseIE(InfoExtractor): - _API_BASE_URL = 'http://www.xiami.com/song/playlist/cat/json/id' + _API_BASE_URL = 'https://emumo.xiami.com/song/playlist/cat/json/id' def _download_webpage_handle(self, *args, **kwargs): webpage = super(XiamiBaseIE, self)._download_webpage_handle(*args, **kwargs) diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py index c6c0b3291..47caec1de 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -47,7 +47,7 @@ class XTubeIE(InfoExtractor): 'display_id': 'A-Super-Run-Part-1-YT', 'ext': 'flv', 'title': 'A Super Run - Part 1 (YT)', - 'description': 'md5:ca0d47afff4a9b2942e4b41aa970fd93', + 'description': 'md5:4cc3af1aa1b0413289babc88f0d4f616', 'uploader': 'tshirtguy59', 'duration': 579, 'view_count': int, @@ -87,10 +87,24 @@ class XTubeIE(InfoExtractor): 'Cookie': 'age_verified=1; cookiesAccepted=1', }) - sources = self._parse_json(self._search_regex( - r'(["\'])?sources\1?\s*:\s*(?P<sources>{.+?}),', - webpage, 'sources', group='sources'), video_id, - transform_source=js_to_json) + title, thumbnail, duration = [None] * 3 + + config = self._parse_json(self._search_regex( + r'playerConf\s*=\s*({.+?})\s*,\s*\n', webpage, 'config', + default='{}'), video_id, transform_source=js_to_json, fatal=False) + if config: + config = config.get('mainRoll') + if isinstance(config, dict): + title = config.get('title') + thumbnail = config.get('poster') + duration = int_or_none(config.get('duration')) + sources = config.get('sources') + + if isinstance(sources, dict): + sources = self._parse_json(self._search_regex( + r'(["\'])?sources\1?\s*:\s*(?P<sources>{.+?}),', + webpage, 'sources', group='sources'), video_id, + transform_source=js_to_json) formats = [] for format_id, format_url in sources.items(): @@ -102,20 +116,25 @@ class XTubeIE(InfoExtractor): self._remove_duplicate_formats(formats) self._sort_formats(formats) - title = self._search_regex( - (r'<h1>\s*(?P<title>[^<]+?)\s*</h1>', r'videoTitle\s*:\s*(["\'])(?P<title>.+?)\1'), - webpage, 'title', group='title') - description = self._search_regex( + if not title: + title = self._search_regex( + (r'<h1>\s*(?P<title>[^<]+?)\s*</h1>', r'videoTitle\s*:\s*(["\'])(?P<title>.+?)\1'), + webpage, 'title', group='title') + description = self._og_search_description( + webpage, default=None) or self._html_search_meta( + 'twitter:description', webpage, default=None) or self._search_regex( r'</h1>\s*<p>([^<]+)', webpage, 'description', fatal=False) uploader = self._search_regex( (r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"', r'<span[^>]+class="nickname"[^>]*>([^<]+)'), webpage, 'uploader', fatal=False) - duration = parse_duration(self._search_regex( - r'<dt>Runtime:?</dt>\s*<dd>([^<]+)</dd>', - webpage, 'duration', fatal=False)) + if not duration: + duration = parse_duration(self._search_regex( + r'<dt>Runtime:?</dt>\s*<dd>([^<]+)</dd>', + webpage, 'duration', fatal=False)) view_count = str_to_int(self._search_regex( - r'<dt>Views:?</dt>\s*<dd>([\d,\.]+)</dd>', + (r'["\']viewsCount["\'][^>]*>(\d+)\s+views', + r'<dt>Views:?</dt>\s*<dd>([\d,\.]+)</dd>'), webpage, 'view count', fatal=False)) comment_count = str_to_int(self._html_search_regex( r'>Comments? \(([\d,\.]+)\)<', @@ -126,6 +145,7 @@ class XTubeIE(InfoExtractor): 'display_id': display_id, 'title': title, 'description': description, + 'thumbnail': thumbnail, 'uploader': uploader, 'duration': duration, 'view_count': view_count, @@ -144,7 +164,7 @@ class XTubeUserIE(InfoExtractor): 'id': 'greenshowers-4056496', 'age_limit': 18, }, - 'playlist_mincount': 155, + 'playlist_mincount': 154, } def _real_extract(self, url): diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py index 166bcf443..8fc64914c 100644 --- a/youtube_dl/extractor/xvideos.py +++ b/youtube_dl/extractor/xvideos.py @@ -17,7 +17,8 @@ class XVideosIE(InfoExtractor): _VALID_URL = r'''(?x) https?:// (?: - (?:www\.)?xvideos\.com/video| + (?:[^/]+\.)?xvideos2?\.com/video| + (?:www\.)?xvideos\.es/video| flashservice\.xvideos\.com/embedframe/| static-hw\.xvideos\.com/swf/xv-player\.swf\?.*?\bid_video= ) @@ -39,6 +40,42 @@ class XVideosIE(InfoExtractor): }, { 'url': 'http://static-hw.xvideos.com/swf/xv-player.swf?id_video=4588838', 'only_matching': True, + }, { + 'url': 'http://xvideos.com/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'https://xvideos.com/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'https://xvideos.es/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'https://www.xvideos.es/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'http://xvideos.es/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'http://www.xvideos.es/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'http://fr.xvideos.com/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'https://fr.xvideos.com/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'http://it.xvideos.com/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'https://it.xvideos.com/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'http://de.xvideos.com/video4588838/biker_takes_his_girl', + 'only_matching': True + }, { + 'url': 'https://de.xvideos.com/video4588838/biker_takes_his_girl', + 'only_matching': True }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index a3b5f00c8..238d9cea0 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -1,451 +1,315 @@ # coding: utf-8 from __future__ import unicode_literals +import hashlib import itertools -import json import re from .common import InfoExtractor, SearchInfoExtractor from ..compat import ( + compat_str, compat_urllib_parse, - compat_urlparse, ) from ..utils import ( clean_html, - determine_ext, - ExtractorError, - extract_attributes, int_or_none, mimetype2ext, + parse_iso8601, smuggle_url, - unescapeHTML, + try_get, + url_or_none, ) -from .brightcove import ( - BrightcoveLegacyIE, - BrightcoveNewIE, -) -from .nbc import NBCSportsVPlayerIE +from .brightcove import BrightcoveNewIE class YahooIE(InfoExtractor): IE_DESC = 'Yahoo screen and movies' - _VALID_URL = r'(?P<host>https?://(?:(?P<country>[a-zA-Z]{2})\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?:(?P<display_id>.+)?-)?(?P<id>[0-9]+)(?:-[a-z]+)?(?:\.html)?' - _TESTS = [ - { - 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', - 'info_dict': { - 'id': '2d25e626-2378-391f-ada0-ddaf1417e588', - 'ext': 'mp4', - 'title': 'Julian Smith & Travis Legg Watch Julian Smith', - 'description': 'Julian and Travis watch Julian Smith', - 'duration': 6863, - }, + _VALID_URL = r'(?P<url>https?://(?:(?P<country>[a-zA-Z]{2}(?:-[a-zA-Z]{2})?|malaysia)\.)?(?:[\da-zA-Z_-]+\.)?yahoo\.com/(?:[^/]+/)*(?P<id>[^?&#]*-[0-9]+(?:-[a-z]+)?)\.html)' + _TESTS = [{ + 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', + 'info_dict': { + 'id': '2d25e626-2378-391f-ada0-ddaf1417e588', + 'ext': 'mp4', + 'title': 'Julian Smith & Travis Legg Watch Julian Smith', + 'description': 'Julian and Travis watch Julian Smith', + 'duration': 6863, + 'timestamp': 1369812016, + 'upload_date': '20130529', }, - { - 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html', - 'md5': '251af144a19ebc4a033e8ba91ac726bb', - 'info_dict': { - 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9', - 'ext': 'mp4', - 'title': 'Codefellas - The Cougar Lies with Spanish Moss', - 'description': 'md5:66b627ab0a282b26352136ca96ce73c1', - 'duration': 151, - }, - 'skip': 'HTTP Error 404', + }, { + 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', + 'md5': '7993e572fac98e044588d0b5260f4352', + 'info_dict': { + 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', + 'ext': 'mp4', + 'title': "Yahoo Saves 'Community'", + 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053', + 'duration': 170, + 'timestamp': 1406838636, + 'upload_date': '20140731', }, - { - 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', - 'md5': '7993e572fac98e044588d0b5260f4352', - 'info_dict': { - 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', - 'ext': 'mp4', - 'title': "Yahoo Saves 'Community'", - 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053', - 'duration': 170, - } - }, - { - 'url': 'https://tw.news.yahoo.com/%E6%95%A2%E5%95%8F%E5%B8%82%E9%95%B7%20%E9%BB%83%E7%A7%80%E9%9C%9C%E6%89%B9%E8%B3%B4%E6%B8%85%E5%BE%B7%20%E9%9D%9E%E5%B8%B8%E9%AB%98%E5%82%B2-034024051.html', - 'md5': '45c024bad51e63e9b6f6fad7a43a8c23', - 'info_dict': { - 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f', - 'ext': 'mp4', - 'title': '敢問市長／黃秀霜批賴清德「非常高傲」', - 'description': '直言台南沒捷運 交通居五都之末', - 'duration': 396, - }, - }, - { - 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', - 'md5': '71298482f7c64cbb7fa064e4553ff1c1', - 'info_dict': { - 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58', - 'ext': 'webm', - 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder', - 'description': 'md5:f66c890e1490f4910a9953c941dee944', - 'duration': 97, - } - }, - { - 'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html', - 'md5': '57e06440778b1828a6079d2f744212c4', - 'info_dict': { - 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73', - 'ext': 'mp4', - 'title': 'Program that makes hockey more affordable not offered in Manitoba', - 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4', - 'duration': 121, - }, - 'skip': 'Video gone', - }, { - 'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html', - 'info_dict': { - 'id': '154609075', - }, - 'playlist': [{ - 'md5': '000887d0dc609bc3a47c974151a40fb8', - 'info_dict': { - 'id': 'e624c4bc-3389-34de-9dfc-025f74943409', - 'ext': 'mp4', - 'title': '\'The Interview\' TV Spot: War', - 'description': 'The Interview', - 'duration': 30, - }, - }, { - 'md5': '81bc74faf10750fe36e4542f9a184c66', - 'info_dict': { - 'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9', - 'ext': 'mp4', - 'title': '\'The Interview\' TV Spot: Guys', - 'description': 'The Interview', - 'duration': 30, - }, - }], - }, { - 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', - 'md5': '88e209b417f173d86186bef6e4d1f160', - 'info_dict': { - 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521', - 'ext': 'mp4', - 'title': 'China Moses Is Crazy About the Blues', - 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0', - 'duration': 128, - } - }, { - 'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html', - 'md5': 'd9a083ccf1379127bf25699d67e4791b', - 'info_dict': { - 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c', - 'ext': 'mp4', - 'title': 'Connect the Dots: Dark Side of Virgo', - 'description': 'md5:1428185051cfd1949807ad4ff6d3686a', - 'duration': 201, - }, - 'skip': 'Domain name in.lifestyle.yahoo.com gone', - }, { - 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', - 'md5': '989396ae73d20c6f057746fb226aa215', - 'info_dict': { - 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', - 'ext': 'mp4', - 'title': '\'True Story\' Trailer', - 'description': 'True Story', - 'duration': 150, - }, - }, { - 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html', - 'only_matching': True, - }, { - 'note': 'NBC Sports embeds', - 'url': 'http://sports.yahoo.com/blogs/ncaab-the-dagger/tyler-kalinoski-s-buzzer-beater-caps-davidson-s-comeback-win-185609842.html?guid=nbc_cbk_davidsonbuzzerbeater_150313', - 'info_dict': { - 'id': '9CsDKds0kvHI', - 'ext': 'flv', - 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d', - 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson', - 'upload_date': '20150313', - 'uploader': 'NBCU-SPORTS', - 'timestamp': 1426270238, - } - }, { - 'url': 'https://tw.news.yahoo.com/-100120367.html', - 'only_matching': True, - }, { - # Query result is embedded in webpage, but explicit request to video API fails with geo restriction - 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html', - 'md5': '4fbafb9c9b6f07aa8f870629f6671b35', - 'info_dict': { - 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', - 'ext': 'mp4', - 'title': 'Communitary - Community Episode 1: Ladders', - 'description': 'md5:8fc39608213295748e1e289807838c97', - 'duration': 1646, - }, - }, { - # it uses an alias to get the video_id - 'url': 'https://www.yahoo.com/movies/the-stars-of-daddys-home-have-very-different-212843197.html', - 'info_dict': { - 'id': '40eda9c8-8e5f-3552-8745-830f67d0c737', - 'ext': 'mp4', - 'title': 'Will Ferrell & Mark Wahlberg Are Pro-Spanking', - 'description': 'While they play feuding fathers in \'Daddy\'s Home,\' star Will Ferrell & Mark Wahlberg share their true feelings on parenthood.', - }, - }, - { - # config['models']['applet_model']['data']['sapi'] has no query - 'url': 'https://www.yahoo.com/music/livenation/event/galactic-2016', - 'md5': 'dac0c72d502bc5facda80c9e6d5c98db', - 'info_dict': { - 'id': 'a6015640-e9e5-3efb-bb60-05589a183919', - 'ext': 'mp4', - 'description': 'Galactic', - 'title': 'Dolla Diva (feat. Maggie Koerner)', - }, - 'skip': 'redirect to https://www.yahoo.com/music', - }, - { - # yahoo://article/ - 'url': 'https://www.yahoo.com/movies/video/true-story-trailer-173000497.html', - 'info_dict': { - 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', - 'ext': 'mp4', - 'title': "'True Story' Trailer", - 'description': 'True Story', - }, - 'params': { - 'skip_download': True, - }, - }, - { - # ytwnews://cavideo/ - 'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html', - 'info_dict': { - 'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff', - 'ext': 'mp4', - 'title': '單車天使 - 中文版預', - 'description': '中文版預', - }, - 'params': { - 'skip_download': True, - }, - }, - { - # custom brightcove - 'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37083565/clown-entertainers-say-it-is-hurting-their-business/', - 'info_dict': { - 'id': '5575377707001', - 'ext': 'mp4', - 'title': "Clown entertainers say 'It' is hurting their business", - 'description': 'Stephen King s horror film has much to answer for. Jelby and Mr Loopy the Clowns join us.', - 'timestamp': 1505341164, - 'upload_date': '20170913', - 'uploader_id': '2376984109001', - }, - 'params': { - 'skip_download': True, - }, - }, - { - # custom brightcove, geo-restricted to Australia, bypassable - 'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37263964/sunrise-episode-wed-27-sep/', - 'only_matching': True, + }, { + 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', + 'md5': '71298482f7c64cbb7fa064e4553ff1c1', + 'info_dict': { + 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58', + 'ext': 'webm', + 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder', + 'description': 'md5:f66c890e1490f4910a9953c941dee944', + 'duration': 97, + 'timestamp': 1414489862, + 'upload_date': '20141028', } - ] + }, { + 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', + 'md5': '88e209b417f173d86186bef6e4d1f160', + 'info_dict': { + 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521', + 'ext': 'mp4', + 'title': 'China Moses Is Crazy About the Blues', + 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0', + 'duration': 128, + 'timestamp': 1385722202, + 'upload_date': '20131129', + } + }, { + 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', + 'md5': '2a9752f74cb898af5d1083ea9f661b58', + 'info_dict': { + 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', + 'ext': 'mp4', + 'title': '\'True Story\' Trailer', + 'description': 'True Story', + 'duration': 150, + 'timestamp': 1418919206, + 'upload_date': '20141218', + }, + }, { + 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html', + 'only_matching': True, + }, { + 'note': 'NBC Sports embeds', + 'url': 'http://sports.yahoo.com/blogs/ncaab-the-dagger/tyler-kalinoski-s-buzzer-beater-caps-davidson-s-comeback-win-185609842.html?guid=nbc_cbk_davidsonbuzzerbeater_150313', + 'info_dict': { + 'id': '9CsDKds0kvHI', + 'ext': 'flv', + 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d', + 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson', + 'upload_date': '20150313', + 'uploader': 'NBCU-SPORTS', + 'timestamp': 1426270238, + }, + }, { + 'url': 'https://tw.news.yahoo.com/-100120367.html', + 'only_matching': True, + }, { + # Query result is embedded in webpage, but explicit request to video API fails with geo restriction + 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html', + 'md5': '4fbafb9c9b6f07aa8f870629f6671b35', + 'info_dict': { + 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', + 'ext': 'mp4', + 'title': 'Communitary - Community Episode 1: Ladders', + 'description': 'md5:8fc39608213295748e1e289807838c97', + 'duration': 1646, + 'timestamp': 1440436550, + 'upload_date': '20150824', + 'series': 'Communitary', + 'season_number': 6, + 'episode_number': 1, + }, + }, { + # ytwnews://cavideo/ + 'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html', + 'info_dict': { + 'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff', + 'ext': 'mp4', + 'title': '單車天使 - 中文版預', + 'description': '中文版預', + 'timestamp': 1476696196, + 'upload_date': '20161017', + }, + 'params': { + 'skip_download': True, + }, + }, { + # Contains both a Yahoo hosted video and multiple Youtube embeds + 'url': 'https://www.yahoo.com/entertainment/gwen-stefani-reveals-the-pop-hit-she-passed-on-assigns-it-to-her-voice-contestant-instead-033045672.html', + 'info_dict': { + 'id': '46c5d95a-528f-3d03-b732-732fcadd51de', + 'title': 'Gwen Stefani reveals the pop hit she passed on, assigns it to her \'Voice\' contestant instead', + 'description': 'Gwen decided not to record this hit herself, but she decided it was the perfect fit for Kyndall Inskeep.', + }, + 'playlist': [{ + 'info_dict': { + 'id': '966d4262-4fd1-3aaa-b45b-049ca6e38ba6', + 'ext': 'mp4', + 'title': 'Gwen Stefani reveals she turned down one of Sia\'s best songs', + 'description': 'On "The Voice" Tuesday, Gwen Stefani told Taylor Swift which Sia hit was almost hers.', + 'timestamp': 1572406500, + 'upload_date': '20191030', + }, + }, { + 'info_dict': { + 'id': '352CFDOQrKg', + 'ext': 'mp4', + 'title': 'Kyndal Inskeep "Performs the Hell Out of" Sia\'s "Elastic Heart" - The Voice Knockouts 2019', + 'description': 'md5:35b61e94c2ae214bc965ff4245f80d11', + 'uploader': 'The Voice', + 'uploader_id': 'NBCTheVoice', + 'upload_date': '20191029', + }, + }], + 'params': { + 'playlistend': 2, + }, + 'expected_warnings': ['HTTP Error 404'], + }, { + 'url': 'https://malaysia.news.yahoo.com/video/bystanders-help-ontario-policeman-bust-190932818.html', + 'only_matching': True, + }, { + 'url': 'https://es-us.noticias.yahoo.com/es-la-puerta-irrompible-que-110539379.html', + 'only_matching': True, + }, { + 'url': 'https://www.yahoo.com/entertainment/v/longtime-cbs-news-60-minutes-032036500-cbs.html', + 'only_matching': True, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - page_id = mobj.group('id') - display_id = mobj.group('display_id') or page_id - host = mobj.group('host') - webpage, urlh = self._download_webpage_handle(url, display_id) - if 'err=404' in urlh.geturl(): - raise ExtractorError('Video gone', expected=True) - - # Look for iframed media first - entries = [] - iframe_urls = re.findall(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage) - for idx, iframe_url in enumerate(iframe_urls): - entries.append(self.url_result(host + iframe_url, 'Yahoo')) - if entries: - return self.playlist_result(entries, page_id) - - # Look for NBCSports iframes - nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage) - if nbc_sports_url: - return self.url_result(nbc_sports_url, NBCSportsVPlayerIE.ie_key()) - - # Look for Brightcove Legacy Studio embeds - bc_url = BrightcoveLegacyIE._extract_brightcove_url(webpage) - if bc_url: - return self.url_result(bc_url, BrightcoveLegacyIE.ie_key()) - - def brightcove_url_result(bc_url): - return self.url_result( - smuggle_url(bc_url, {'geo_countries': [mobj.group('country')]}), - BrightcoveNewIE.ie_key()) - - # Look for Brightcove New Studio embeds - bc_url = BrightcoveNewIE._extract_url(self, webpage) - if bc_url: - return brightcove_url_result(bc_url) - - brightcove_iframe = self._search_regex( - r'(<iframe[^>]+data-video-id=["\']\d+[^>]+>)', webpage, - 'brightcove iframe', default=None) - if brightcove_iframe: - attr = extract_attributes(brightcove_iframe) - src = attr.get('src') - if src: - parsed_src = compat_urlparse.urlparse(src) - qs = compat_urlparse.parse_qs(parsed_src.query) - account_id = qs.get('accountId', ['2376984109001'])[0] - brightcove_id = attr.get('data-video-id') or qs.get('videoId', [None])[0] - if account_id and brightcove_id: - return brightcove_url_result( - 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s' - % (account_id, brightcove_id)) - - # Query result is often embedded in webpage as JSON. Sometimes explicit requests - # to video API results in a failure with geo restriction reason therefore using - # embedded query result when present sounds reasonable. - config_json = self._search_regex( - r'window\.Af\.bootstrap\[[^\]]+\]\s*=\s*({.*?"applet_type"\s*:\s*"td-applet-videoplayer".*?});(?:</script>|$)', - webpage, 'videoplayer applet', default=None) - if config_json: - config = self._parse_json(config_json, display_id, fatal=False) - if config: - sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi') - if sapi and 'query' in sapi: - info = self._extract_info(display_id, sapi, webpage) - self._sort_formats(info['formats']) - return info - - items_json = self._search_regex( - r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE, - default=None) - if items_json is None: - alias = self._search_regex( - r'"aliases":{"video":"(.*?)"', webpage, 'alias', default=None) - if alias is not None: - alias_info = self._download_json( - 'https://www.yahoo.com/_td/api/resource/VideoService.videos;video_aliases=["%s"]' % alias, - display_id, 'Downloading alias info') - video_id = alias_info[0]['id'] - else: - CONTENT_ID_REGEXES = [ - r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"', - r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"', - r'"first_videoid"\s*:\s*"([^"]+)"', - r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id), - r'<article[^>]data-uuid=["\']([^"\']+)', - r'<meta[^<>]+yahoo://article/view\?.*\buuid=([^&"\']+)', - r'<meta[^<>]+["\']ytwnews://cavideo/(?:[^/]+/)+([\da-fA-F-]+)[&"\']', - ] - video_id = self._search_regex( - CONTENT_ID_REGEXES, webpage, 'content ID') + url, country, display_id = re.match(self._VALID_URL, url).groups() + if not country: + country = 'us' else: - items = json.loads(items_json) - info = items['mediaItems']['query']['results']['mediaObj'][0] - # The 'meta' field is not always in the video webpage, we request it - # from another page - video_id = info['id'] - return self._get_info(video_id, display_id, webpage) + country = country.split('-')[0] + api_base = 'https://%s.yahoo.com/_td/api/resource/' % country - def _extract_info(self, display_id, query, webpage): - info = query['query']['results']['mediaObj'][0] - meta = info.get('meta') - video_id = info.get('id') + for i, uuid in enumerate(['url=' + url, 'ymedia-alias=' + display_id]): + content = self._download_json( + api_base + 'content;getDetailView=true;uuids=["%s"]' % uuid, + display_id, 'Downloading content JSON metadata', fatal=i == 1) + if content: + item = content['items'][0] + break - if not meta: - msg = info['status'].get('msg') - if msg: - raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, msg), expected=True) - raise ExtractorError('Unable to extract media object meta') + if item.get('type') != 'video': + entries = [] + cover = item.get('cover') or {} + if cover.get('type') == 'yvideo': + cover_url = cover.get('url') + if cover_url: + entries.append(self.url_result( + cover_url, 'Yahoo', cover.get('uuid'))) + + for e in item.get('body', []): + if e.get('type') == 'videoIframe': + iframe_url = e.get('url') + if not iframe_url: + continue + entries.append(self.url_result(iframe_url)) + + return self.playlist_result( + entries, item.get('uuid'), + item.get('title'), item.get('summary')) + + video_id = item['uuid'] + video = self._download_json( + api_base + 'VideoService.videos;view=full;video_ids=["%s"]' % video_id, + video_id, 'Downloading video JSON metadata')[0] + title = video['title'] + + if country == 'malaysia': + country = 'my' + + is_live = video.get('live_state') == 'live' + fmts = ('m3u8',) if is_live else ('webm', 'mp4') + + urls = [] formats = [] - for s in info['streams']: - tbr = int_or_none(s.get('bitrate')) - format_info = { - 'width': int_or_none(s.get('width')), - 'height': int_or_none(s.get('height')), - 'tbr': tbr, - } - - host = s['host'] - path = s['path'] - if host.startswith('rtmp'): - fmt = 'rtmp' - format_info.update({ - 'url': host, - 'play_path': path, - 'ext': 'flv', - }) - else: - if s.get('format') == 'm3u8_playlist': - fmt = 'hls' - format_info.update({ - 'protocol': 'm3u8_native', - 'ext': 'mp4', - }) - else: - fmt = format_info['ext'] = determine_ext(path) - format_url = compat_urlparse.urljoin(host, path) - format_info['url'] = format_url - format_info['format_id'] = fmt + ('-%d' % tbr if tbr else '') - formats.append(format_info) - - closed_captions = self._html_search_regex( - r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions', - default='[]') - - cc_json = self._parse_json(closed_captions, video_id, fatal=False) subtitles = {} - if cc_json: - for closed_caption in cc_json: - lang = closed_caption['lang'] - if lang not in subtitles: - subtitles[lang] = [] - subtitles[lang].append({ - 'url': closed_caption['url'], - 'ext': mimetype2ext(closed_caption['content_type']), + for fmt in fmts: + media_obj = self._download_json( + 'https://video-api.yql.yahoo.com/v1/video/sapi/streams/' + video_id, + video_id, 'Downloading %s JSON metadata' % fmt, + headers=self.geo_verification_headers(), query={ + 'format': fmt, + 'region': country.upper(), + })['query']['results']['mediaObj'][0] + msg = media_obj.get('status', {}).get('msg') + + for s in media_obj.get('streams', []): + host = s.get('host') + path = s.get('path') + if not host or not path: + continue + s_url = host + path + if s.get('format') == 'm3u8': + formats.extend(self._extract_m3u8_formats( + s_url, video_id, 'mp4', m3u8_id='hls', fatal=False)) + continue + tbr = int_or_none(s.get('bitrate')) + formats.append({ + 'url': s_url, + 'format_id': fmt + ('-%d' % tbr if tbr else ''), + 'width': int_or_none(s.get('width')), + 'height': int_or_none(s.get('height')), + 'tbr': tbr, + 'fps': int_or_none(s.get('framerate')), }) + for cc in media_obj.get('closedcaptions', []): + cc_url = cc.get('url') + if not cc_url or cc_url in urls: + continue + urls.append(cc_url) + subtitles.setdefault(cc.get('lang') or 'en-US', []).append({ + 'url': cc_url, + 'ext': mimetype2ext(cc.get('content_type')), + }) + + streaming_url = video.get('streaming_url') + if streaming_url and not is_live: + formats.extend(self._extract_m3u8_formats( + streaming_url, video_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) + + if not formats and msg == 'geo restricted': + self.raise_geo_restricted() + + self._sort_formats(formats) + + thumbnails = [] + for thumb in video.get('thumbnails', []): + thumb_url = thumb.get('url') + if not thumb_url: + continue + thumbnails.append({ + 'id': thumb.get('tag'), + 'url': thumb.get('url'), + 'width': int_or_none(thumb.get('width')), + 'height': int_or_none(thumb.get('height')), + }) + + series_info = video.get('series_info') or {} + return { 'id': video_id, - 'display_id': display_id, - 'title': unescapeHTML(meta['title']), + 'title': self._live_title(title) if is_live else title, 'formats': formats, - 'description': clean_html(meta['description']), - 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage), - 'duration': int_or_none(meta.get('duration')), + 'display_id': display_id, + 'thumbnails': thumbnails, + 'description': clean_html(video.get('description')), + 'timestamp': parse_iso8601(video.get('publish_time')), 'subtitles': subtitles, + 'duration': int_or_none(video.get('duration')), + 'view_count': int_or_none(video.get('view_count')), + 'is_live': is_live, + 'series': video.get('show_name'), + 'season_number': int_or_none(series_info.get('season_number')), + 'episode_number': int_or_none(series_info.get('episode_number')), } - def _get_info(self, video_id, display_id, webpage): - region = self._search_regex( - r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"', - webpage, 'region', fatal=False, default='US').upper() - formats = [] - info = {} - for fmt in ('webm', 'mp4'): - query_result = self._download_json( - 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' + video_id, - display_id, 'Downloading %s video info' % fmt, query={ - 'protocol': 'http', - 'region': region, - 'format': fmt, - }) - info = self._extract_info(display_id, query_result, webpage) - formats.extend(info['formats']) - formats.extend(self._extract_m3u8_formats( - 'http://video.media.yql.yahoo.com/v1/hls/%s?region=%s' % (video_id, region), - video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) - self._sort_formats(formats) - info['formats'] = formats - return info - class YahooSearchIE(SearchInfoExtractor): IE_DESC = 'Yahoo screen search' @@ -519,7 +383,7 @@ class YahooGyaOPlayerIE(InfoExtractor): 'id': video_id, 'title': video['title'], 'url': smuggle_url( - 'http://players.brightcove.net/4235717419001/default_default/index.html?videoId=' + video['videoId'], + 'http://players.brightcove.net/4235717419001/SyG5P0gjb_default/index.html?videoId=' + video['videoId'], {'geo_countries': ['JP']}), 'description': video.get('longDescription'), 'ie_key': BrightcoveNewIE.ie_key(), @@ -556,3 +420,130 @@ class YahooGyaOIE(InfoExtractor): 'https://gyao.yahoo.co.jp/player/%s/' % video_id.replace(':', '/'), YahooGyaOPlayerIE.ie_key(), video_id)) return self.playlist_result(entries, program_id) + + +class YahooJapanNewsIE(InfoExtractor): + IE_NAME = 'yahoo:japannews' + IE_DESC = 'Yahoo! Japan News' + _VALID_URL = r'https?://(?P<host>(?:news|headlines)\.yahoo\.co\.jp)[^\d]*(?P<id>\d[\d-]*\d)?' + _GEO_COUNTRIES = ['JP'] + _TESTS = [{ + 'url': 'https://headlines.yahoo.co.jp/videonews/ann?a=20190716-00000071-ann-int', + 'info_dict': { + 'id': '1736242', + 'ext': 'mp4', + 'title': 'ムン大統領が対日批判を強化“現金化”効果は？（テレビ朝日系（ANN）） - Yahoo!ニュース', + 'description': '韓国の元徴用工らを巡る裁判の原告が弁護士が差し押さえた三菱重工業の資産を売却して - Yahoo!ニュース(テレビ朝日系（ANN）)', + 'thumbnail': r're:^https?://.*\.[a-zA-Z\d]{3,4}$', + }, + 'params': { + 'skip_download': True, + }, + }, { + # geo restricted + 'url': 'https://headlines.yahoo.co.jp/hl?a=20190721-00000001-oxv-l04', + 'only_matching': True, + }, { + 'url': 'https://headlines.yahoo.co.jp/videonews/', + 'only_matching': True, + }, { + 'url': 'https://news.yahoo.co.jp', + 'only_matching': True, + }, { + 'url': 'https://news.yahoo.co.jp/byline/hashimotojunji/20190628-00131977/', + 'only_matching': True, + }, { + 'url': 'https://news.yahoo.co.jp/feature/1356', + 'only_matching': True + }] + + def _extract_formats(self, json_data, content_id): + formats = [] + + video_data = try_get( + json_data, + lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'], + list) + for vid in video_data or []: + delivery = vid.get('delivery') + url = url_or_none(vid.get('Url')) + if not delivery or not url: + continue + elif delivery == 'hls': + formats.extend( + self._extract_m3u8_formats( + url, content_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + else: + formats.append({ + 'url': url, + 'format_id': 'http-%s' % compat_str(vid.get('bitrate', '')), + 'height': int_or_none(vid.get('height')), + 'width': int_or_none(vid.get('width')), + 'tbr': int_or_none(vid.get('bitrate')), + }) + self._remove_duplicate_formats(formats) + self._sort_formats(formats) + + return formats + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') + display_id = mobj.group('id') or host + + webpage = self._download_webpage(url, display_id) + + title = self._html_search_meta( + ['og:title', 'twitter:title'], webpage, 'title', default=None + ) or self._html_search_regex('<title>([^<]+)', webpage, 'title') + + if display_id == host: + # Headline page (w/ multiple BC playlists) ('news.yahoo.co.jp', 'headlines.yahoo.co.jp/videonews/', ...) + stream_plists = re.findall(r'plist=(\d+)', webpage) or re.findall(r'plist["\']:\s*["\']([^"\']+)', webpage) + entries = [ + self.url_result( + smuggle_url( + 'http://players.brightcove.net/5690807595001/HyZNerRl7_default/index.html?playlistId=%s' % plist_id, + {'geo_countries': ['JP']}), + ie='BrightcoveNew', video_id=plist_id) + for plist_id in stream_plists] + return self.playlist_result(entries, playlist_title=title) + + # Article page + description = self._html_search_meta( + ['og:description', 'description', 'twitter:description'], + webpage, 'description', default=None) + thumbnail = self._og_search_thumbnail( + webpage, default=None) or self._html_search_meta( + 'twitter:image', webpage, 'thumbnail', default=None) + space_id = self._search_regex([ + r']+class=["\']yvpub-player["\'][^>]+spaceid=([^&"\']+)', + r'YAHOO\.JP\.srch\.\w+link\.onLoad[^;]+spaceID["\' ]*:["\' ]+([^"\']+)', + r'