Release 2021.03.15

3be098010f

.github/ISSUE_TEMPLATE/1_broken_site.md

@@ -21,7 +21,7 @@ assignees: ''
 
 <!--
 Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
-- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.07. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
 - Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
 - Make sure that all URLs and arguments with special characters are properly quoted or escaped as explained in https://github.com/yt-dlp/yt-dlp.
 - Search the bugtracker for similar issues: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
@@ -29,7 +29,7 @@ Carefully read and work through this check list in order to prevent the most com
 -->
 
 - [ ] I'm reporting a broken site support
-- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.07**
 - [ ] 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
@@ -44,7 +44,7 @@ Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your com
  [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] yt-dlp version 2021.03.01
+ [debug] yt-dlp version 2021.03.07
  [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: {}

.github/ISSUE_TEMPLATE/2_site_support_request.md

@@ -21,7 +21,7 @@ assignees: ''
 
 <!--
 Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
-- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.07. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
 - Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
 - Make sure that site you are requesting is not dedicated to copyright infringement, see https://github.com/yt-dlp/yt-dlp. yt-dlp does not support such sites. In order for site support request to be accepted all provided example URLs should not violate any copyrights.
 - Search the bugtracker for similar site support requests: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
@@ -29,7 +29,7 @@ Carefully read and work through this check list in order to prevent the most com
 -->
 
 - [ ] I'm reporting a new site support request
-- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.07**
 - [ ] 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

.github/ISSUE_TEMPLATE/3_site_feature_request.md

@@ -21,13 +21,13 @@ assignees: ''
 
 <!--
 Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
-- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.07. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
 - Search the bugtracker for similar site feature requests: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
 - Finally, put x into all relevant boxes like this [x] (Dont forget to delete the empty space)
 -->
 
 - [ ] I'm reporting a site feature request
-- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.07**
 - [ ] I've searched the bugtracker for similar site feature requests including closed ones
 
 

.github/ISSUE_TEMPLATE/4_bug_report.md

@@ -21,7 +21,7 @@ assignees: ''
 
 <!--
 Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
-- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.07. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
 - Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
 - Make sure that all URLs and arguments with special characters are properly quoted or escaped as explained in https://github.com/yt-dlp/yt-dlp.
 - Search the bugtracker for similar issues: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
@@ -30,7 +30,7 @@ Carefully read and work through this check list in order to prevent the most com
 -->
 
 - [ ] I'm reporting a broken site support issue
-- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.07**
 - [ ] 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
@@ -46,7 +46,7 @@ Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your com
  [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] yt-dlp version 2021.03.01
+ [debug] yt-dlp version 2021.03.07
  [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: {}

.github/ISSUE_TEMPLATE/5_feature_request.md

@@ -21,13 +21,13 @@ assignees: ''
 
 <!--
 Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
-- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.07. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
 - Search the bugtracker for similar feature requests: https://github.com/yt-dlp/yt-dlp. DO NOT post duplicates.
 - Finally, put x into all relevant boxes like this [x] (Dont forget to delete the empty space)
 -->
 
 - [ ] I'm reporting a feature request
-- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.07**
 - [ ] I've searched the bugtracker for similar feature requests including closed ones
 
 

.github/workflows/build.yml

@@ -29,7 +29,7 @@ jobs:
     - name: Print version
       run: echo "${{ steps.bump_version.outputs.ytdlp_version }}"
     - name: Run Make
-      run: make yt-dlp
+      run: make
     - name: Create Release
       id: create_release
       uses: actions/create-release@v1

.gitignore

@@ -60,6 +60,7 @@ yt-dlp.zip
 *.mkv
 *.swf
 *.part
+*.part-*
 *.ytdl
 *.dump
 *.frag

CONTRIBUTORS

@@ -21,6 +21,7 @@ nao20010128nao
 kurumigi
 tsukumi
 bbepis
+animelover1984
 Pccode66
 Ashish
 RobinD42
@@ -28,3 +29,4 @@ hseg
 colethedj
 DennyDai
 codeasashu
+teesid

Changelog.md

@@ -17,16 +17,58 @@
 -->
 
 
+### 2021.03.15
+* **Split video by chapters**: using option `--split-chapters`
+    * The output file of the split files can be set with `-o`/`-P` using the prefix `chapter:`
+    * Additional keys `section_title`, `section_number`, `section_start`, `section_end` are available in the output template
+* **Parallel fragment downloads** by [shirt](https://github.com/shirt-dev)
+    * Use option `--concurrent-fragments` (`-N`) to set the number of threads (default 1)
+* Merge youtube-dl: Upto [commit/3be0980](https://github.com/ytdl-org/youtube-dl/commit/3be098010f667b14075e3dfad1e74e5e2becc8ea)
+* [Zee5] Add Show Extractor by [Ashish](https://github.com/Ashish) and [pukkandan](https://github.com/pukkandan)
+* [rai] fix drm check [nixxo](https://github.com/nixxo)
+* [zee5] Support zee5originals
+* [wimtv] Add extractor by [nixxo](https://github.com/nixxo)
+* [mtv] Add mtv.it and extract series metadata by [nixxo](https://github.com/nixxo)
+* [pluto.tv] Add extractor by [kevinoconnor7](https://github.com/kevinoconnor7)
+* [Youtube] Rewrite comment extraction by [colethedj](https://github.com/colethedj)
+* [embedthumbnail] Set mtime correctly
+* Refactor some postprocessor/downloader code by [pukkandan](https://github.com/pukkandan) and [shirt](https://github.com/shirt-dev)
+
+
+### 2021.03.07
+* [youtube] Fix history, mixes, community pages and trending by [pukkandan](https://github.com/pukkandan) and [colethedj](https://github.com/colethedj)
+* [youtube] Fix private feeds/playlists on multi-channel accounts by [colethedj](https://github.com/colethedj)
+* [youtube] Extract alerts from continuation by [colethedj](https://github.com/colethedj)
+* [cbs] Add support for ParamountPlus by [shirt](https://github.com/shirt-dev)
+* [mxplayer] Rewrite extractor with show support by [pukkandan](https://github.com/pukkandan) and [Ashish](https://github.com/Ashish)
+* [gedi] Improvements from youtube-dl by [nixxo](https://github.com/nixxo)
+* [vimeo] Fix videos with password by [teesid](https://github.com/teesid)
+* [lbry] Support `lbry://` url by [nixxo](https://github.com/nixxo)
+* [bilibili] Change `Accept` header by [pukkandan](https://github.com/pukkandan) and [animelover1984](https://github.com/animelover1984)
+* [trovo] Pass origin header
+* [rai] Check for DRM by [nixxo](https://github.com/nixxo)
+* [downloader] Fix bug for `ffmpeg`/`httpie`
+* [update] Fix updater removing the executable bit on some UNIX distros
+* [update] Fix current build hash for UNIX
+* [documentation] Include wget/curl/aria2c install instructions for Unix by [Ashish](https://github.com/Ashish)
+* Fix some videos downloading with `m3u8` extension
+* Remove "fixup is ignored" warning when fixup wasn't passed by user
+
+
+### 2021.03.03.2
+* [build] Fix bug
+
+
 ### 2021.03.03
-* [youtube] Use new browse API for continuation page extraction by @colethedj and @pukkandan
+* [youtube] Use new browse API for continuation page extraction by [colethedj](https://github.com/colethedj) and [pukkandan](https://github.com/pukkandan)
-* Fix HLS playlist downloading by @shirt
+* Fix HLS playlist downloading by [shirt](https://github.com/shirt-dev)
-* **Merge youtube-dl:** Upto [2021.03.03](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.03.03)
+* Merge youtube-dl: Upto [2021.03.03](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.03.03)
 * [mtv] Fix extractor
-* [nick] Fix extractor by @DennyDai
+* [nick] Fix extractor by [DennyDai](https://github.com/DennyDai)
-* [mxplayer] Add new extractor by@codeasashu
+* [mxplayer] Add new extractor by [codeasashu](https://github.com/codeasashu)
 * [youtube] Throw error when `--extractor-retries` are exhausted
 * Reduce default of `--extractor-retries` to 3
-* Fix packaging bugs by @hseg
+* Fix packaging bugs by [hseg](https://github.com/hseg)
 
 
 ### 2021.03.01
@@ -55,7 +97,7 @@
 * Moved project to an organization [yt-dlp](https://github.com/yt-dlp)
 * **Completely changed project name to yt-dlp** by [Pccode66](https://github.com/Pccode66) and [pukkandan](https://github.com/pukkandan)
     * Also, `youtube-dlc` config files are no longer loaded
-* **Merge youtube-dl:** Upto [commit/4460329](https://github.com/ytdl-org/youtube-dl/commit/44603290e5002153f3ebad6230cc73aef42cc2cd) (except tmz, gedi)
+* Merge youtube-dl: Upto [commit/4460329](https://github.com/ytdl-org/youtube-dl/commit/44603290e5002153f3ebad6230cc73aef42cc2cd) (except tmz, gedi)
 * [Readthedocs](https://yt-dlp.readthedocs.io) support by [shirt](https://github.com/shirt-dev)
 * [youtube] Show if video was a live stream in info (`was_live`)
 * [Zee5] Add new extractor by [Ashish](https://github.com/Ashish) and [pukkandan](https://github.com/pukkandan)
@@ -73,7 +115,7 @@
 
 
 ### 2021.02.19
-* **Merge youtube-dl:** Upto [commit/cf2dbec](https://github.com/ytdl-org/youtube-dl/commit/cf2dbec6301177a1fddf72862de05fa912d9869d) (except kakao)
+* Merge youtube-dl: Upto [commit/cf2dbec](https://github.com/ytdl-org/youtube-dl/commit/cf2dbec6301177a1fddf72862de05fa912d9869d) (except kakao)
 * [viki] Fix extractor
 * [niconico] Extract `channel` and `channel_id` by [kurumigi](https://github.com/kurumigi)
 * [youtube] Multiple page support for hashtag URLs
@@ -98,7 +140,7 @@
 
 
 ### 2021.02.15
-* **Merge youtube-dl:** Upto [2021.02.10](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.02.10) (except archive.org)
+* Merge youtube-dl: Upto [2021.02.10](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.02.10) (except archive.org)
 * [niconico] Improved extraction and support encrypted/SMILE movies by [kurumigi](https://github.com/kurumigi), [tsukumi](https://github.com/tsukumi), [bbepis](https://github.com/bbepis), [pukkandan](https://github.com/pukkandan)
 * Fix HLS AES-128 with multiple keys in external downloaders by [shirt](https://github.com/shirt-dev)
 * [youtube_live_chat] Fix by using POST API by [siikamiika](https://github.com/siikamiika)
@@ -141,7 +183,7 @@
 
 
 ### 2021.02.04
-* **Merge youtube-dl:** Upto [2021.02.04.1](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.02.04.1)
+* Merge youtube-dl: Upto [2021.02.04.1](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.02.04.1)
 * **Date/time formatting in output template:**
     * You can use [`strftime`](https://docs.python.org/3/library/datetime.html#strftime-and-strptime-format-codes) to format date/time fields. Example: `%(upload_date>%Y-%m-%d)s`
 * **Multiple output templates:**
@@ -195,7 +237,7 @@
 
 
 ### 2021.01.24
-* **Merge youtube-dl:** Upto [2021.01.24](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.01.16)
+* Merge youtube-dl: Upto [2021.01.24](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.01.16)
 * Plugin support ([documentation](https://github.com/yt-dlp/yt-dlp#plugins))
 * **Multiple paths**: New option `-P`/`--paths` to give different paths for different types of files
     * The syntax is `-P "type:path" -P "type:path"` ([documentation](https://github.com/yt-dlp/yt-dlp#:~:text=-P,%20--paths%20TYPE:PATH))
@@ -224,7 +266,7 @@
 
 
 ### 2021.01.16
-* **Merge youtube-dl:** Upto [2021.01.16](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.01.16)
+* Merge youtube-dl: Upto [2021.01.16](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.01.16)
 * **Configuration files:**
     * Portable configuration file: `./yt-dlp.conf`
     * Allow the configuration files to be named `yt-dlp` instead of `youtube-dlc`. See [this](https://github.com/yt-dlp/yt-dlp#configuration) for details
@@ -270,8 +312,7 @@
 
 
 ### 2021.01.08
-* **Merge youtube-dl:** Upto [2021.01.08](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.01.08)
+* Merge youtube-dl: Upto [2021.01.08](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.01.08) except stitcher ([1](https://github.com/ytdl-org/youtube-dl/commit/bb38a1215718cdf36d73ff0a7830a64cd9fa37cc), [2](https://github.com/ytdl-org/youtube-dl/commit/a563c97c5cddf55f8989ed7ea8314ef78e30107f))
-    * Extractor stitcher ([1](https://github.com/ytdl-org/youtube-dl/commit/bb38a1215718cdf36d73ff0a7830a64cd9fa37cc), [2](https://github.com/ytdl-org/youtube-dl/commit/a563c97c5cddf55f8989ed7ea8314ef78e30107f)) have not been merged
 * Moved changelog to seperate file
 
 
@@ -310,7 +351,7 @@
     * Changed video format sorting to show video only files and video+audio files together.
     * Added `--video-multistreams`, `--no-video-multistreams`, `--audio-multistreams`, `--no-audio-multistreams`
     * Added `b`,`w`,`v`,`a` as alias for `best`, `worst`, `video` and `audio` respectively
-* **Shortcut Options:** Added `--write-link`, `--write-url-link`, `--write-webloc-link`, `--write-desktop-link` by [h-h-h-h](https://github.com/h-h-h-h) - See [Internet Shortcut Options](README.md#internet-shortcut-options) for details
+* Shortcut Options: Added `--write-link`, `--write-url-link`, `--write-webloc-link`, `--write-desktop-link` by [h-h-h-h](https://github.com/h-h-h-h) - See [Internet Shortcut Options](README.md#internet-shortcut-options) for details
 * **Sponskrub integration:** Added `--sponskrub`, `--sponskrub-cut`, `--sponskrub-force`, `--sponskrub-location`, `--sponskrub-args` - See [SponSkrub Options](README.md#sponskrub-sponsorblock-options) for details
 * Added `--force-download-archive` (`--force-write-archive`) by [h-h-h-h](https://github.com/h-h-h-h)
 * Added `--list-formats-as-table`,  `--list-formats-old`
@@ -320,36 +361,38 @@
 * Relaxed validation for format filters so that any arbitrary field can be used
 * Fix for embedding thumbnail in mp3 by [pauldubois98](https://github.com/pauldubois98) ([ytdl-org/youtube-dl#21569](https://github.com/ytdl-org/youtube-dl/pull/21569))
 * Make Twitch Video ID output from Playlist and VOD extractor same. This is only a temporary fix
-* **Merge youtube-dl:** Upto [2021.01.03](https://github.com/ytdl-org/youtube-dl/commit/8e953dcbb10a1a42f4e12e4e132657cb0100a1f8) - See [blackjack4494/yt-dlc#280](https://github.com/blackjack4494/yt-dlc/pull/280) for details
+* Merge youtube-dl: Upto [2021.01.03](https://github.com/ytdl-org/youtube-dl/commit/8e953dcbb10a1a42f4e12e4e132657cb0100a1f8) - See [blackjack4494/yt-dlc#280](https://github.com/blackjack4494/yt-dlc/pull/280) for details
     * Extractors [tiktok](https://github.com/ytdl-org/youtube-dl/commit/fb626c05867deab04425bad0c0b16b55473841a2) and [hotstar](https://github.com/ytdl-org/youtube-dl/commit/bb38a1215718cdf36d73ff0a7830a64cd9fa37cc) have not been merged
 * Cleaned up the fork for public use
 
 
+**PS**: All uncredited changes above this point are authored by [pukkandan](https://github.com/pukkandan)
+
 ### Unreleased changes in [blackjack4494/yt-dlc](https://github.com/blackjack4494/yt-dlc)
-* Updated to youtube-dl release 2020.11.26
+* Updated to youtube-dl release 2020.11.26 by [pukkandan](https://github.com/pukkandan)
-* [youtube]
+* Youtube improvements by [pukkandan](https://github.com/pukkandan)
     * Implemented all Youtube Feeds (ytfav, ytwatchlater, ytsubs, ythistory, ytrec) and SearchURL
-    * Fix ytsearch not returning results sometimes due to promoted content
-    * Temporary fix for automatic captions - disable json3
     * Fix some improper Youtube URLs
     * Redirect channel home to /video
     * Print youtube's warning message
-    * Multiple pages are handled better for feeds
+    * Handle Multiple pages for feeds better
+* [youtube] Fix ytsearch not returning results sometimes due to promoted content by [colethedj](https://github.com/colethedj)
+* [youtube] Temporary fix for automatic captions - disable json3 by [blackjack4494](https://github.com/blackjack4494)
 * Add --break-on-existing by [gergesh](https://github.com/gergesh)
-* Pre-check video IDs in the archive before downloading
+* Pre-check video IDs in the archive before downloading by [pukkandan](https://github.com/pukkandan)
-* [bitwave.tv] New extractor
+* [bitwave.tv] New extractor by [lorpus](https://github.com/lorpus)
-* [Gedi] Add extractor
+* [Gedi] Add extractor by [nixxo](https://github.com/nixxo)
-* [Rcs] Add new extractor
+* [Rcs] Add new extractor by [nixxo](https://github.com/nixxo)
-* [skyit] Add support for multiple Sky Italia website and removed old skyitalia extractor
+* [skyit] New skyitalia extractor by [nixxo](https://github.com/nixxo)
-* [france.tv] Fix thumbnail URL
+* [france.tv] Fix thumbnail URL by [renalid](https://github.com/renalid)
-* [ina] support mobile links
+* [ina] support mobile links by [B0pol](https://github.com/B0pol)
-* [instagram] Fix extractor
+* [instagram] Fix thumbnail extractor by [nao20010128nao](https://github.com/nao20010128nao)
-* [itv] BTCC new pages' URL update (articles instead of races)
+* [SouthparkDe] Support for English URLs by [xypwn](https://github.com/xypwn)
-* [SouthparkDe] Support for English URLs
+* [spreaker] fix SpreakerShowIE test URL by [pukkandan](https://github.com/pukkandan)
-* [spreaker] fix SpreakerShowIE test URL
+* [Vlive] Fix playlist handling when downloading a channel by [kyuyeunk](https://github.com/kyuyeunk)
-* [Vlive] Fix playlist handling when downloading a channel
+* [tmz] Fix extractor by [diegorodriguezv](https://github.com/diegorodriguezv)
-* [generic] Detect embedded bitchute videos
+* [generic] Detect embedded bitchute videos by [pukkandan](https://github.com/pukkandan)
-* [generic] Extract embedded youtube and twitter videos
+* [generic] Extract embedded youtube and twitter videos by [diegorodriguezv](https://github.com/diegorodriguezv)
-* [ffmpeg] Ensure all streams are copied
+* [ffmpeg] Ensure all streams are copied by [pukkandan](https://github.com/pukkandan)
-* Fix for os.rename error when embedding thumbnail to video in a different drive
+* [embedthumbnail] Fix for os.rename error by [pukkandan](https://github.com/pukkandan)
-* make_win.bat: don't use UPX to pack vcruntime140.dll
+* make_win.bat: don't use UPX to pack vcruntime140.dll by [jbruchon](https://github.com/jbruchon)

Makefile

@@ -1,4 +1,4 @@
-all: yt-dlp doc
+all: yt-dlp doc pypi-files
 clean: clean-test clean-dist clean-cache
 completions: completion-bash completion-fish completion-zsh
 doc: README.md CONTRIBUTING.md issuetemplates supportedsites

README.md

@@ -57,7 +57,7 @@ The major new features from the latest release of [blackjack4494/yt-dlc](https:/
 
 * **[Format Sorting](#sorting-formats)**: The default format sorting options have been changed so that higher resolution and better codecs will be now preferred instead of simply using larger bitrate. Furthermore, you can now specify the sort order using `-S`. This allows for much easier format selection that what is possible by simply using `--format` ([examples](#format-selection-examples))
 
-* **Merged with youtube-dl v2021.03.03**: You get all the latest features and patches of [youtube-dl](https://github.com/ytdl-org/youtube-dl) in addition to all the features of [youtube-dlc](https://github.com/blackjack4494/yt-dlc)
+* **Merged with youtube-dl v2021.03.14**: You get all the latest features and patches of [youtube-dl](https://github.com/ytdl-org/youtube-dl) in addition to all the features of [youtube-dlc](https://github.com/blackjack4494/yt-dlc)
 
 * **Merged with animelover1984/youtube-dl**: You get most of the features and improvements from [animelover1984/youtube-dl](https://github.com/animelover1984/youtube-dl) including `--get-comments`, `BiliBiliSearch`, `BilibiliChannel`, Embedding thumbnail in mp4/ogg/opus, Playlist infojson etc. Note that the NicoNico improvements are not available. See [#31](https://github.com/yt-dlp/yt-dlp/pull/31) for details.
 
@@ -66,17 +66,19 @@ The major new features from the latest release of [blackjack4494/yt-dlc](https:/
     * Youtube search (`ytsearch:`, `ytsearchdate:`) along with Search URLs works correctly
     * Redirect channel's home URL automatically to `/video` to preserve the old behaviour
 
+* **Split video by chapters**: Videos can be split into multiple files based on chapters using `--split-chapters`
+
+* **Multithreaded fragment downloads**: Fragment downloads can be natively multi-threaded. Use `--concurrent-fragments` (`-N`) option to set the number of threads used
+
 * **Aria2c with HLS/DASH**: You can use aria2c as the external downloader for DASH(mpd) and HLS(m3u8) formats. No more slow ffmpeg/native downloads
 
-* **New extractors**: AnimeLab, Philo MSO, Rcs, Gedi, bitwave.tv, mildom, audius, zee5
+* **New extractors**: AnimeLab, Philo MSO, Rcs, Gedi, bitwave.tv, mildom, audius, zee5, mtv.it, wimtv, pluto.tv
 
 * **Fixed extractors**: archive.org, roosterteeth.com, skyit, instagram, itv, SouthparkDe, spreaker, Vlive, tiktok, akamai, ina, rumble, tennistv
 
 * **Plugin support**: Extractors can be loaded from an external file. See [plugins](#plugins) for details
 
-* **Multiple paths and output templates**: You can give different [output templates](#output-template) and download paths for different types of files. You can also set a temporary path where intermediary files are downloaded to. See [`--paths`](https://github.com/yt-dlp/yt-dlp/#:~:text=-P,%20--paths%20TYPE:PATH) for details
+* **Multiple paths and output templates**: You can give different [output templates](#output-template) and download paths for different types of files. You can also set a temporary path where intermediary files are downloaded to using `--paths` (`-P`)
-
-<!-- Relative link doesn't work for "#:~:text=" -->
 
 * **Portable Configuration**: Configuration files are automatically loaded from the home and root directories. See [configuration](#configuration) for details
 
@@ -103,6 +105,23 @@ You can install yt-dlp using one of the following methods:
 * Use pip+git: `python -m pip install --upgrade git+https://github.com/yt-dlp/yt-dlp.git@release`
 * Install master branch: `python -m pip install --upgrade git+https://github.com/yt-dlp/yt-dlp`
 
+UNIX users (Linux, macOS, BSD) can also install the [latest release](https://github.com/yt-dlp/yt-dlp/releases/latest) one of the following ways:
+
+```
+sudo curl -L https://github.com/yt-dlp/yt-dlp/releases/latest/download/yt-dlp -o /usr/local/bin/yt-dlp
+sudo chmod a+rx /usr/local/bin/yt-dlp
+```
+
+```
+sudo wget https://github.com/yt-dlp/yt-dlp/releases/latest/download/yt-dlp -O /usr/local/bin/yt-dlp
+sudo chmod a+rx /usr/local/bin/yt-dlp
+```
+
+```
+sudo aria2c https://github.com/yt-dlp/yt-dlp/releases/latest/download/yt-dlp -o /usr/local/bin/yt-dlp
+sudo chmod a+rx /usr/local/bin/yt-dlp
+```
+
 ### UPDATE
 Starting from version `2021.02.09`, you can use `yt-dlp -U` to update if you are using the provided release.
 If you are using `pip`, simply re-run the same command that was used to install the program.
@@ -177,7 +196,7 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
                                      only list them
     --no-flat-playlist               Extract the videos of a playlist
     --mark-watched                   Mark videos watched (YouTube only)
-    --no-mark-watched                Do not mark videos watched
+    --no-mark-watched                Do not mark videos watched (default)
     --no-colors                      Do not emit color codes in output
 
 ## Network Options:
@@ -280,6 +299,8 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
     --no-include-ads                 Do not download advertisements (default)
 
 ## Download Options:
+    -N, --concurrent-fragments N     Number of fragments to download
+                                     concurrently (default is 1)
     -r, --limit-rate RATE            Maximum download rate in bytes per second
                                      (e.g. 50K or 4.2M)
     -R, --retries RETRIES            Number of retries (default is 10), or
@@ -624,11 +645,12 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
                                      arguments to the specified executable only
                                      when being used by the specified
                                      postprocessor. Additionally, for
-                                     ffmpeg/ffprobe, a number can be appended to
+                                     ffmpeg/ffprobe, "_i"/"_o" can be appended
-                                     the exe name seperated by "_i" to pass the
+                                     to the prefix optionally followed by a
-                                     argument before the specified input file.
+                                     number to pass the argument before the
-                                     Eg: --ppa "Merger+ffmpeg_i1:-v quiet". You
+                                     specified input/output file. Eg: --ppa
-                                     can use this option multiple times to give
+                                     "Merger+ffmpeg_i1:-v quiet". You can use
+                                     this option multiple times to give
                                      different arguments to different
                                      postprocessors. (Alias: --ppa)
     -k, --keep-video                 Keep the intermediate video file on disk
@@ -674,6 +696,13 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
                                      push {} /sdcard/Music/ && rm {}'
     --convert-subs FORMAT            Convert the subtitles to other format
                                      (currently supported: srt|ass|vtt|lrc)
+    --split-chapters                 Split video into multiple files based on
+                                     internal chapters. The "chapter:" prefix
+                                     can be used with "--paths" and "--output"
+                                     to set the output filename for the split
+                                     files. See "OUTPUT TEMPLATE" for details
+    --no-split-chapters              Do not split video based on chapters
+                                     (default)
 
 ## SponSkrub (SponsorBlock) Options:
 [SponSkrub](https://github.com/yt-dlp/SponSkrub) is a utility to
@@ -789,9 +818,9 @@ The `-o` option is used to indicate a template for the output file names while `
 
 **tl;dr:** [navigate me to examples](#output-template-examples).
 
-The basic usage of `-o` is not to set any template arguments when downloading a single file, like in `yt-dlp -o funny_video.flv "https://some/video"`. However, it may contain special sequences that will be replaced when downloading each video. The special sequences may be formatted according to [python string formatting operations](https://docs.python.org/2/library/stdtypes.html#string-formatting). For example, `%(NAME)s` or `%(NAME)05d`. To clarify, that is a percent symbol followed by a name in parentheses, followed by formatting operations. Date/time fields can also be formatted according to [strftime formatting](https://docs.python.org/3/library/datetime.html#strftime-and-strptime-format-codes) by specifying it inside the parantheses seperated from the field name using a `>`. For example, `%(duration>%H-%M-%S)s`.
+The basic usage of `-o` is not to set any template arguments when downloading a single file, like in `yt-dlp -o funny_video.flv "https://some/video"` (hard-coding file extension like this is not recommended). However, it may contain special sequences that will be replaced when downloading each video. The special sequences may be formatted according to [python string formatting operations](https://docs.python.org/2/library/stdtypes.html#string-formatting). For example, `%(NAME)s` or `%(NAME)05d`. To clarify, that is a percent symbol followed by a name in parentheses, followed by formatting operations. Date/time fields can also be formatted according to [strftime formatting](https://docs.python.org/3/library/datetime.html#strftime-and-strptime-format-codes) by specifying it inside the parantheses seperated from the field name using a `>`. For example, `%(duration>%H-%M-%S)s`.
 
-Additionally, you can set different output templates for the various metadata files seperately from the general output template by specifying the type of file followed by the template seperated by a colon ":". The different filetypes supported are `subtitle|thumbnail|description|annotation|infojson|pl_description|pl_infojson`. For example, `-o '%(title)s.%(ext)s' -o 'thumbnail:%(title)s\%(title)s.%(ext)s'`  will put the thumbnails in a folder with the same name as the video.
+Additionally, you can set different output templates for the various metadata files seperately from the general output template by specifying the type of file followed by the template seperated by a colon ":". The different filetypes supported are `subtitle`, `thumbnail`, `description`, `annotation`, `infojson`, `pl_description`, `pl_infojson`, `chapter`. For example, `-o '%(title)s.%(ext)s' -o 'thumbnail:%(title)s\%(title)s.%(ext)s'`  will put the thumbnails in a folder with the same name as the video.
 
 The available fields are:
 
@@ -800,6 +829,7 @@ The available fields are:
  - `url` (string): Video URL
  - `ext` (string): Video filename extension
  - `alt_title` (string): A secondary title of the video
+ - `description` (string): The description of the video
  - `display_id` (string): An alternative identifier for the video
  - `uploader` (string): Full name of the video uploader
  - `license` (string): License name the video is licensed under
@@ -882,6 +912,13 @@ Available for the media that is a track or a part of a music album:
  - `disc_number` (numeric): Number of the disc or other physical medium the track belongs to
  - `release_year` (numeric): Year (YYYY) when the album was released
 
+Available when using `--split-chapters` for videos with internal chapters:
+
+ - `section_title` (string): Title of the chapter
+ - `section_number` (numeric): Number of the chapter within the file
+ - `section_start` (numeric): Start time of the chapter in seconds
+ - `section_end` (numeric): End time of the chapter in seconds
+
 Each aforementioned sequence when referenced in an output template will be replaced by the actual value corresponding to the sequence name. Note that some of the sequences are not guaranteed to be present since they depend on the metadata obtained by a particular extractor. Such sequences will be replaced with placeholder value provided with `--output-na-placeholder` (`NA` by default).
 
 For example for `-o %(title)s-%(id)s.%(ext)s` and an mp4 video with title `yt-dlp test video` and id `BaW_jenozKcj`, this will result in a `yt-dlp test video-BaW_jenozKcj.mp4` file created in the current directory.

supportedsites.md

@@ -347,8 +347,7 @@
  - **Gaskrank**
  - **Gazeta**
  - **GDCVault**
- - **Gedi**
+ - **GediDigital**
- - **GediEmbeds**
  - **generic**: Generic downloader that works on some sites
  - **Gfycat**
  - **GiantBomb**
@@ -562,6 +561,8 @@
  - **mtg**: MTG services
  - **mtv**
  - **mtv.de**
+ - **mtv.it**
+ - **mtv.it:programma**
  - **mtv:video**
  - **mtvjapan**
  - **mtvservices:embedded**
@@ -735,6 +736,7 @@
  - **Playwire**
  - **pluralsight**
  - **pluralsight:course**
+ - **PlutoTV**
  - **podomatic**
  - **Pokemon**
  - **PokemonWatch**
@@ -1172,6 +1174,7 @@
  - **Weibo**
  - **WeiboMobile**
  - **WeiqiTV**: WQTV
+ - **WimTV**
  - **Wistia**
  - **WistiaPlaylist**
  - **wnl**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
@@ -1242,6 +1245,7 @@
  - **ZDF**
  - **ZDFChannel**
  - **Zee5**
+ - **zee5:series**
  - **Zhihu**
  - **zingmp3**: mp3.zing.vn
  - **zoom**

test/test_all_urls.py

@@ -37,7 +37,6 @@ class TestAllURLsMatching(unittest.TestCase):
         assertPlaylist('PL63F0C78739B09958')
         assertTab('https://www.youtube.com/AsapSCIENCE')
         assertTab('https://www.youtube.com/embedded')
-        assertTab('https://www.youtube.com/feed')  # Own channel's home page
         assertTab('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
         assertTab('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
         assertTab('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')

yt_dlp/YoutubeDL.py

@@ -1171,6 +1171,9 @@ class YoutubeDL(object):
         else:
             raise Exception('Invalid result type: %s' % result_type)
 
+    def _ensure_dir_exists(self, path):
+        return make_dir(path, self.report_error)
+
     def __process_playlist(self, ie_result, download):
         # We process each entry in the playlist
         playlist = ie_result.get('title') or ie_result.get('id')
@@ -1187,12 +1190,9 @@ class YoutubeDL(object):
             }
             ie_copy.update(dict(ie_result))
 
-            def ensure_dir_exists(path):
-                return make_dir(path, self.report_error)
-
             if self.params.get('writeinfojson', False):
                 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
-                if not ensure_dir_exists(encodeFilename(infofn)):
+                if not self._ensure_dir_exists(encodeFilename(infofn)):
                     return
                 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
                     self.to_screen('[info] Playlist metadata is already present')
@@ -1208,7 +1208,7 @@ class YoutubeDL(object):
 
             if self.params.get('writedescription', False):
                 descfn = self.prepare_filename(ie_copy, 'pl_description')
-                if not ensure_dir_exists(encodeFilename(descfn)):
+                if not self._ensure_dir_exists(encodeFilename(descfn)):
                     return
                 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
                     self.to_screen('[info] Playlist description is already present')
@@ -1794,14 +1794,18 @@ class YoutubeDL(object):
         if 'display_id' not in info_dict and 'id' in info_dict:
             info_dict['display_id'] = info_dict['id']
 
-        if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
+        for ts_key, date_key in (
-            # Working around out-of-range timestamp values (e.g. negative ones on Windows,
+                ('timestamp', 'upload_date'),
-            # see http://bugs.python.org/issue1646728)
+                ('release_timestamp', 'release_date'),
-            try:
+        ):
-                upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
+            if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
-                info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
+                # Working around out-of-range timestamp values (e.g. negative ones on Windows,
-            except (ValueError, OverflowError, OSError):
+                # see http://bugs.python.org/issue1646728)
-                pass
+                try:
+                    upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
+                    info_dict[date_key] = upload_date.strftime('%Y%m%d')
+                except (ValueError, OverflowError, OSError):
+                    pass
 
         # Auto generate title fields corresponding to the *_number fields when missing
         # in order to always have clean titles. This is very common for TV series.
@@ -2089,17 +2093,14 @@ class YoutubeDL(object):
         if full_filename is None:
             return
 
-        def ensure_dir_exists(path):
+        if not self._ensure_dir_exists(encodeFilename(full_filename)):
-            return make_dir(path, self.report_error)
-
-        if not ensure_dir_exists(encodeFilename(full_filename)):
             return
-        if not ensure_dir_exists(encodeFilename(temp_filename)):
+        if not self._ensure_dir_exists(encodeFilename(temp_filename)):
             return
 
         if self.params.get('writedescription', False):
             descfn = self.prepare_filename(info_dict, 'description')
-            if not ensure_dir_exists(encodeFilename(descfn)):
+            if not self._ensure_dir_exists(encodeFilename(descfn)):
                 return
             if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
                 self.to_screen('[info] Video description is already present')
@@ -2116,7 +2117,7 @@ class YoutubeDL(object):
 
         if self.params.get('writeannotations', False):
             annofn = self.prepare_filename(info_dict, 'annotation')
-            if not ensure_dir_exists(encodeFilename(annofn)):
+            if not self._ensure_dir_exists(encodeFilename(annofn)):
                 return
             if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
                 self.to_screen('[info] Video annotations are already present')
@@ -2204,7 +2205,7 @@ class YoutubeDL(object):
 
         if self.params.get('writeinfojson', False):
             infofn = self.prepare_filename(info_dict, 'infojson')
-            if not ensure_dir_exists(encodeFilename(infofn)):
+            if not self._ensure_dir_exists(encodeFilename(infofn)):
                 return
             if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
                 self.to_screen('[info] Video metadata is already present')
@@ -2360,7 +2361,7 @@ class YoutubeDL(object):
                             fname = prepend_extension(
                                 self.prepare_filename(new_info, 'temp'),
                                 'f%s' % f['format_id'], new_info['ext'])
-                            if not ensure_dir_exists(fname):
+                            if not self._ensure_dir_exists(fname):
                                 return
                             downloaded.append(fname)
                             partial_success, real_download = dl(fname, new_info)
@@ -2437,9 +2438,8 @@ class YoutubeDL(object):
                     else:
                         assert fixup_policy in ('ignore', 'never')
 
-                if (info_dict.get('protocol') == 'm3u8_native'
+                if ('protocol' in info_dict
-                        or info_dict.get('protocol') == 'm3u8'
+                        and get_suitable_downloader(info_dict, self.params).__name__ == 'HlsFD'):
-                        and self.params.get('hls_prefer_native')):
                     if fixup_policy == 'warn':
                         self.report_warning('%s: malformed AAC bitstream detected.' % (
                             info_dict['id']))

yt_dlp/__init__.py

@@ -180,6 +180,8 @@ def _real_main(argv=None):
     if opts.overwrites:
         # --yes-overwrites implies --no-continue
         opts.continue_dl = False
+    if opts.concurrent_fragment_downloads <= 0:
+        raise ValueError('Concurrent fragments must be positive')
 
     def parse_retries(retries, name=''):
         if retries in ('inf', 'infinite'):
@@ -277,9 +279,14 @@ def _real_main(argv=None):
 
     def report_conflict(arg1, arg2):
         write_string('WARNING: %s is ignored since %s was given\n' % (arg2, arg1), out=sys.stderr)
+
     if opts.remuxvideo and opts.recodevideo:
         report_conflict('--recode-video', '--remux-video')
         opts.remuxvideo = False
+    if opts.sponskrub_cut and opts.split_chapters and opts.sponskrub is not False:
+        report_conflict('--split-chapter', '--sponskrub-cut')
+        opts.sponskrub_cut = False
+
     if opts.allow_unplayable_formats:
         if opts.extractaudio:
             report_conflict('--allow-unplayable-formats', '--extract-audio')
@@ -369,11 +376,7 @@ def _real_main(argv=None):
         })
         if not already_have_thumbnail:
             opts.writethumbnail = True
-    # XAttrMetadataPP should be run after post-processors that may change file
+    # This should be below most ffmpeg PP because it may cut parts out from the video
-    # contents
-    if opts.xattrs:
-        postprocessors.append({'key': 'XAttrMetadata'})
-    # This should be below all ffmpeg PP because it may cut parts out from the video
     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
     if opts.sponskrub is not False:
         postprocessors.append({
@@ -384,6 +387,11 @@ def _real_main(argv=None):
             'force': opts.sponskrub_force,
             'ignoreerror': opts.sponskrub is None,
         })
+    if opts.split_chapters:
+        postprocessors.append({'key': 'FFmpegSplitChapters'})
+    # XAttrMetadataPP should be run after post-processors that may change file contents
+    if opts.xattrs:
+        postprocessors.append({'key': 'XAttrMetadata'})
     # ExecAfterDownload must be the last PP
     if opts.exec_cmd:
         postprocessors.append({
@@ -463,6 +471,7 @@ def _real_main(argv=None):
         'extractor_retries': opts.extractor_retries,
         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
         'keep_fragments': opts.keep_fragments,
+        'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
         'buffersize': opts.buffersize,
         'noresizebuffer': opts.noresizebuffer,
         'http_chunk_size': opts.http_chunk_size,

yt_dlp/downloader/common.py

@@ -326,6 +326,12 @@ class FileDownloader(object):
         """Report it was impossible to resume download."""
         self.to_screen('[download] Unable to resume')
 
+    @staticmethod
+    def supports_manifest(manifest):
+        """ Whether the downloader can download the fragments from the manifest.
+        Redefine in subclasses if needed. """
+        pass
+
     def download(self, filename, info_dict, subtitle=False):
         """Download to a filename using the info from info_dict
         Return True on success and False otherwise

yt_dlp/downloader/dash.py

@@ -1,18 +1,26 @@
 from __future__ import unicode_literals
 
+try:
+    import concurrent.futures
+    can_threaded_download = True
+except ImportError:
+    can_threaded_download = False
+
 from ..downloader import _get_real_downloader
 from .fragment import FragmentFD
 
 from ..compat import compat_urllib_error
 from ..utils import (
     DownloadError,
+    sanitize_open,
     urljoin,
 )
 
 
 class DashSegmentsFD(FragmentFD):
     """
-    Download segments in a DASH manifest
+    Download segments in a DASH manifest. External downloaders can take over
+    the fragment downloads by supporting the 'frag_urls' protocol
     """
 
     FD_NAME = 'dashsegments'
@@ -37,7 +45,7 @@ class DashSegmentsFD(FragmentFD):
         fragment_retries = self.params.get('fragment_retries', 0)
         skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
 
-        fragment_urls = []
+        fragments_to_download = []
         frag_index = 0
         for i, fragment in enumerate(fragments):
             frag_index += 1
@@ -48,49 +56,15 @@ class DashSegmentsFD(FragmentFD):
                 assert fragment_base_url
                 fragment_url = urljoin(fragment_base_url, fragment['path'])
 
-            if real_downloader:
+            fragments_to_download.append({
-                fragment_urls.append(fragment_url)
+                'frag_index': frag_index,
-                continue
+                'index': i,
-
+                'url': fragment_url,
-            # In DASH, the first segment contains necessary headers to
+            })
-            # generate a valid MP4 file, so always abort for the first segment
-            fatal = i == 0 or not skip_unavailable_fragments
-            count = 0
-            while count <= fragment_retries:
-                try:
-                    success, frag_content = self._download_fragment(ctx, fragment_url, info_dict)
-                    if not success:
-                        return False
-                    self._append_fragment(ctx, frag_content)
-                    break
-                except compat_urllib_error.HTTPError as err:
-                    # YouTube may often return 404 HTTP error for a fragment causing the
-                    # whole download to fail. However if the same fragment is immediately
-                    # 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.
-                    count += 1
-                    if count <= fragment_retries:
-                        self.report_retry_fragment(err, frag_index, count, fragment_retries)
-                except DownloadError:
-                    # Don't retry fragment if error occurred during HTTP downloading
-                    # itself since it has own retry settings
-                    if not fatal:
-                        self.report_skip_fragment(frag_index)
-                        break
-                    raise
-
-            if count > fragment_retries:
-                if not fatal:
-                    self.report_skip_fragment(frag_index)
-                    continue
-                self.report_error('giving up after %s fragment retries' % fragment_retries)
-                return False
 
         if real_downloader:
             info_copy = info_dict.copy()
-            info_copy['url_list'] = fragment_urls
+            info_copy['fragments'] = fragments_to_download
             fd = real_downloader(self.ydl, self.params)
             # TODO: Make progress updates work without hooking twice
             # for ph in self._progress_hooks:
@@ -99,5 +73,104 @@ class DashSegmentsFD(FragmentFD):
             if not success:
                 return False
         else:
+            def download_fragment(fragment):
+                i = fragment['index']
+                frag_index = fragment['frag_index']
+                fragment_url = fragment['url']
+
+                ctx['fragment_index'] = frag_index
+
+                # In DASH, the first segment contains necessary headers to
+                # generate a valid MP4 file, so always abort for the first segment
+                fatal = i == 0 or not skip_unavailable_fragments
+                count = 0
+                while count <= fragment_retries:
+                    try:
+                        success, frag_content = self._download_fragment(ctx, fragment_url, info_dict)
+                        if not success:
+                            return False, frag_index
+                        break
+                    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 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.
+                        count += 1
+                        if count <= fragment_retries:
+                            self.report_retry_fragment(err, frag_index, count, fragment_retries)
+                    except DownloadError:
+                        # Don't retry fragment if error occurred during HTTP downloading
+                        # itself since it has own retry settings
+                        if not fatal:
+                            break
+                        raise
+
+                if count > fragment_retries:
+                    if not fatal:
+                        return False, frag_index
+                    self.report_error('giving up after %s fragment retries' % fragment_retries)
+                    return False, frag_index
+
+                return frag_content, frag_index
+
+            def append_fragment(frag_content, frag_index):
+                if frag_content:
+                    fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], frag_index)
+                    try:
+                        file, frag_sanitized = sanitize_open(fragment_filename, 'rb')
+                        ctx['fragment_filename_sanitized'] = frag_sanitized
+                        file.close()
+                        self._append_fragment(ctx, frag_content)
+                        return True
+                    except FileNotFoundError:
+                        if skip_unavailable_fragments:
+                            self.report_skip_fragment(frag_index)
+                            return True
+                        else:
+                            self.report_error(
+                                'fragment %s not found, unable to continue' % frag_index)
+                            return False
+                else:
+                    if skip_unavailable_fragments:
+                        self.report_skip_fragment(frag_index)
+                        return True
+                    else:
+                        self.report_error(
+                            'fragment %s not found, unable to continue' % frag_index)
+                        return False
+
+            max_workers = self.params.get('concurrent_fragment_downloads', 1)
+            if can_threaded_download and max_workers > 1:
+                self.report_warning('The download speed shown is only of one thread. This is a known issue')
+                with concurrent.futures.ThreadPoolExecutor(max_workers) as pool:
+                    futures = [pool.submit(download_fragment, fragment) for fragment in fragments_to_download]
+                    # timeout must be 0 to return instantly
+                    done, not_done = concurrent.futures.wait(futures, timeout=0)
+                    try:
+                        while not_done:
+                            # Check every 1 second for KeyboardInterrupt
+                            freshly_done, not_done = concurrent.futures.wait(not_done, timeout=1)
+                            done |= freshly_done
+                    except KeyboardInterrupt:
+                        for future in not_done:
+                            future.cancel()
+                        # timeout must be none to cancel
+                        concurrent.futures.wait(not_done, timeout=None)
+                        raise KeyboardInterrupt
+                results = [future.result() for future in futures]
+
+                for frag_content, frag_index in results:
+                    result = append_fragment(frag_content, frag_index)
+                    if not result:
+                        return False
+            else:
+                for fragment in fragments_to_download:
+                    frag_content, frag_index = download_fragment(fragment)
+                    result = append_fragment(frag_content, frag_index)
+                    if not result:
+                        return False
+
             self._finish_frag_download(ctx)
         return True

yt_dlp/downloader/external.py

@@ -108,7 +108,8 @@ class ExternalFD(FileDownloader):
     def _configuration_args(self, *args, **kwargs):
         return cli_configuration_args(
             self.params.get('external_downloader_args'),
-            self.get_basename(), *args, **kwargs)
+            [self.get_basename(), 'default'],
+            *args, **kwargs)
 
     def _call_downloader(self, tmpfilename, info_dict):
         """ Either overwrite this or implement _make_cmd """
@@ -122,18 +123,14 @@ class ExternalFD(FileDownloader):
         if p.returncode != 0:
             self.to_stderr(stderr.decode('utf-8', 'replace'))
 
-        if 'url_list' in info_dict:
+        if 'fragments' in info_dict:
             file_list = []
-            for [i, url] in enumerate(info_dict['url_list']):
-                tmpsegmentname = '%s_%s.frag' % (tmpfilename, i)
-                file_list.append(tmpsegmentname)
-            key_list = info_dict.get('key_list')
-            decrypt_info = None
             dest, _ = sanitize_open(tmpfilename, 'wb')
-            for i, file in enumerate(file_list):
+            for i, fragment in enumerate(info_dict['fragments']):
+                file = '%s-Frag%d' % (tmpfilename, i)
+                decrypt_info = fragment.get('decrypt_info')
                 src, _ = sanitize_open(file, 'rb')
-                if key_list:
+                if decrypt_info:
-                    decrypt_info = next((x for x in key_list if x['INDEX'] == i), decrypt_info)
                     if decrypt_info['METHOD'] == 'AES-128':
                         iv = decrypt_info.get('IV')
                         decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen(
@@ -149,6 +146,7 @@ class ExternalFD(FileDownloader):
                     fragment_data = src.read()
                     dest.write(fragment_data)
                 src.close()
+                file_list.append(file)
             dest.close()
             if not self.params.get('keep_fragments', False):
                 for file_path in file_list:
@@ -245,10 +243,19 @@ class Aria2cFD(ExternalFD):
     AVAILABLE_OPT = '-v'
     SUPPORTED_PROTOCOLS = ('http', 'https', 'ftp', 'ftps', 'frag_urls')
 
+    @staticmethod
+    def supports_manifest(manifest):
+        UNSUPPORTED_FEATURES = [
+            r'#EXT-X-BYTERANGE',  # playlists composed of byte ranges of media files [1]
+            # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.2
+        ]
+        check_results = (not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES)
+        return all(check_results)
+
     def _make_cmd(self, tmpfilename, info_dict):
         cmd = [self.exe, '-c']
         dn = os.path.dirname(tmpfilename)
-        if 'url_list' not in info_dict:
+        if 'fragments' not in info_dict:
             cmd += ['--out', os.path.basename(tmpfilename)]
         verbose_level_args = ['--console-log-level=warn', '--summary-interval=0']
         cmd += self._configuration_args(['--file-allocation=none', '-x16', '-j16', '-s16'] + verbose_level_args)
@@ -262,14 +269,14 @@ class Aria2cFD(ExternalFD):
         cmd += self._bool_option('--check-certificate', 'nocheckcertificate', 'false', 'true', '=')
         cmd += self._bool_option('--remote-time', 'updatetime', 'true', 'false', '=')
         cmd += ['--auto-file-renaming=false']
-        if 'url_list' in info_dict:
+        if 'fragments' in info_dict:
             cmd += verbose_level_args
             cmd += ['--uri-selector', 'inorder', '--download-result=hide']
             url_list_file = '%s.frag.urls' % tmpfilename
             url_list = []
-            for [i, url] in enumerate(info_dict['url_list']):
+            for i, fragment in enumerate(info_dict['fragments']):
-                tmpsegmentname = '%s_%s.frag' % (os.path.basename(tmpfilename), i)
+                tmpsegmentname = '%s-Frag%d' % (os.path.basename(tmpfilename), i)
-                url_list.append('%s\n\tout=%s' % (url, tmpsegmentname))
+                url_list.append('%s\n\tout=%s' % (fragment['url'], tmpsegmentname))
             stream, _ = sanitize_open(url_list_file, 'wb')
             stream.write('\n'.join(url_list).encode('utf-8'))
             stream.close()
@@ -282,8 +289,8 @@ class Aria2cFD(ExternalFD):
 
 class HttpieFD(ExternalFD):
     @classmethod
-    def available(cls):
+    def available(cls, path=None):
-        return check_executable('http', ['--version'])
+        return check_executable(path or 'http', ['--version'])
 
     def _make_cmd(self, tmpfilename, info_dict):
         cmd = ['http', '--download', '--output', tmpfilename, info_dict['url']]
@@ -298,7 +305,7 @@ class FFmpegFD(ExternalFD):
     SUPPORTED_PROTOCOLS = ('http', 'https', 'ftp', 'ftps', 'm3u8', 'rtsp', 'rtmp', 'mms')
 
     @classmethod
-    def available(cls):
+    def available(cls, path=None):  # path is ignored for ffmpeg
         return FFmpegPostProcessor().available
 
     def _call_downloader(self, tmpfilename, info_dict):

yt_dlp/downloader/hls.py

@@ -7,6 +7,11 @@ try:
     can_decrypt_frag = True
 except ImportError:
     can_decrypt_frag = False
+try:
+    import concurrent.futures
+    can_threaded_download = True
+except ImportError:
+    can_threaded_download = False
 
 from ..downloader import _get_real_downloader
 from .fragment import FragmentFD
@@ -19,12 +24,17 @@ from ..compat import (
 )
 from ..utils import (
     parse_m3u8_attributes,
+    sanitize_open,
     update_url_query,
 )
 
 
 class HlsFD(FragmentFD):
-    """ A limited implementation that does not require ffmpeg """
+    """
+    Download segments in a m3u8 manifest. External downloaders can take over
+    the fragment downloads by supporting the 'frag_urls' protocol and
+    re-defining 'supports_manifest' function
+    """
 
     FD_NAME = 'hlsnative'
 
@@ -53,12 +63,15 @@ class HlsFD(FragmentFD):
             UNSUPPORTED_FEATURES += [
                 r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)',  # encrypted streams [1]
             ]
-        check_results = [not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES]
+
-        is_aes128_enc = '#EXT-X-KEY:METHOD=AES-128' in manifest
+        def check_results():
-        check_results.append(with_crypto or not is_aes128_enc)
+            yield not info_dict.get('is_live')
-        check_results.append(not (is_aes128_enc and r'#EXT-X-BYTERANGE' in manifest))
+            is_aes128_enc = '#EXT-X-KEY:METHOD=AES-128' in manifest
-        check_results.append(not info_dict.get('is_live'))
+            yield with_crypto or not is_aes128_enc
-        return all(check_results)
+            yield not (is_aes128_enc and r'#EXT-X-BYTERANGE' in manifest)
+            for feature in UNSUPPORTED_FEATURES:
+                yield not re.search(feature, manifest)
+        return all(check_results())
 
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
@@ -84,6 +97,8 @@ class HlsFD(FragmentFD):
             return fd.real_download(filename, info_dict)
 
         real_downloader = _get_real_downloader(info_dict, 'frag_urls', self.params, None)
+        if real_downloader and not real_downloader.supports_manifest(s):
+            real_downloader = None
 
         def is_ad_fragment_start(s):
             return (s.startswith('#ANVATO-SEGMENT-INFO') and 'type=ad' in s
@@ -93,7 +108,7 @@ class HlsFD(FragmentFD):
             return (s.startswith('#ANVATO-SEGMENT-INFO') and 'type=master' in s
                     or s.startswith('#UPLYNK-SEGMENT') and s.endswith(',segment'))
 
-        fragment_urls = []
+        fragments = []
 
         media_frags = 0
         ad_frags = 0
@@ -136,14 +151,12 @@ class HlsFD(FragmentFD):
         i = 0
         media_sequence = 0
         decrypt_info = {'METHOD': 'NONE'}
-        key_list = []
         byte_range = {}
         discontinuity_count = 0
         frag_index = 0
         ad_frag_next = False
         for line in s.splitlines():
             line = line.strip()
-            download_frag = False
             if line:
                 if not line.startswith('#'):
                     if format_index and discontinuity_count != format_index:
@@ -160,10 +173,13 @@ class HlsFD(FragmentFD):
                     if extra_query:
                         frag_url = update_url_query(frag_url, extra_query)
 
-                    if real_downloader:
+                    fragments.append({
-                        fragment_urls.append(frag_url)
+                        'frag_index': frag_index,
-                        continue
+                        'url': frag_url,
-                    download_frag = True
+                        'decrypt_info': decrypt_info,
+                        'byte_range': byte_range,
+                        'media_sequence': media_sequence,
+                    })
 
                 elif line.startswith('#EXT-X-MAP'):
                     if format_index and discontinuity_count != format_index:
@@ -180,9 +196,14 @@ class HlsFD(FragmentFD):
                         else compat_urlparse.urljoin(man_url, map_info.get('URI')))
                     if extra_query:
                         frag_url = update_url_query(frag_url, extra_query)
-                    if real_downloader:
+
-                        fragment_urls.append(frag_url)
+                    fragments.append({
-                        continue
+                        'frag_index': frag_index,
+                        'url': frag_url,
+                        'decrypt_info': decrypt_info,
+                        'byte_range': byte_range,
+                        'media_sequence': media_sequence
+                    })
 
                     if map_info.get('BYTERANGE'):
                         splitted_byte_range = map_info.get('BYTERANGE').split('@')
@@ -191,7 +212,6 @@ class HlsFD(FragmentFD):
                             'start': sub_range_start,
                             'end': sub_range_start + int(splitted_byte_range[0]),
                         }
-                    download_frag = True
 
                 elif line.startswith('#EXT-X-KEY'):
                     decrypt_url = decrypt_info.get('URI')
@@ -206,9 +226,6 @@ class HlsFD(FragmentFD):
                             decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_query)
                         if decrypt_url != decrypt_info['URI']:
                             decrypt_info['KEY'] = None
-                    key_data = decrypt_info.copy()
-                    key_data['INDEX'] = frag_index
-                    key_list.append(key_data)
 
                 elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):
                     media_sequence = int(line[22:])
@@ -225,58 +242,16 @@ class HlsFD(FragmentFD):
                     ad_frag_next = False
                 elif line.startswith('#EXT-X-DISCONTINUITY'):
                     discontinuity_count += 1
+                i += 1
+                media_sequence += 1
 
-                if download_frag:
+        # We only download the first fragment during the test
-                    count = 0
+        if test:
-                    headers = info_dict.get('http_headers', {})
+            fragments = [fragments[0] if fragments else None]
-                    if byte_range:
-                        headers['Range'] = 'bytes=%d-%d' % (byte_range['start'], byte_range['end'] - 1)
-                    while count <= fragment_retries:
-                        try:
-                            success, frag_content = self._download_fragment(
-                                ctx, frag_url, info_dict, headers)
-                            if not success:
-                                return False
-                            break
-                        except compat_urllib_error.HTTPError as err:
-                            # Unavailable (possibly temporary) fragments may be served.
-                            # First we try to retry then either skip or abort.
-                            # See https://github.com/ytdl-org/youtube-dl/issues/10165,
-                            # https://github.com/ytdl-org/youtube-dl/issues/10448).
-                            count += 1
-                            if count <= fragment_retries:
-                                self.report_retry_fragment(err, frag_index, count, fragment_retries)
-                    if count > fragment_retries:
-                        if skip_unavailable_fragments:
-                            i += 1
-                            media_sequence += 1
-                            self.report_skip_fragment(frag_index)
-                            continue
-                        self.report_error(
-                            'giving up after %s fragment retries' % fragment_retries)
-                        return False
-
-                    if decrypt_info['METHOD'] == 'AES-128':
-                        iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence)
-                        decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen(
-                            self._prepare_url(info_dict, info_dict.get('_decryption_key_url') or decrypt_info['URI'])).read()
-                        # Don't decrypt the content in tests since the data is explicitly truncated and it's not to a valid block
-                        # size (see https://github.com/ytdl-org/youtube-dl/pull/27660). Tests only care that the correct data downloaded,
-                        # not what it decrypts to.
-                        if not test:
-                            frag_content = AES.new(
-                                decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
-                    self._append_fragment(ctx, frag_content)
-                    # We only download the first fragment during the test
-                    if test:
-                        break
-                    i += 1
-                    media_sequence += 1
 
         if real_downloader:
             info_copy = info_dict.copy()
-            info_copy['url_list'] = fragment_urls
+            info_copy['fragments'] = fragments
-            info_copy['key_list'] = key_list
             fd = real_downloader(self.ydl, self.params)
             # TODO: Make progress updates work without hooking twice
             # for ph in self._progress_hooks:
@@ -285,5 +260,106 @@ class HlsFD(FragmentFD):
             if not success:
                 return False
         else:
+            def download_fragment(fragment):
+                frag_index = fragment['frag_index']
+                frag_url = fragment['url']
+                decrypt_info = fragment['decrypt_info']
+                byte_range = fragment['byte_range']
+                media_sequence = fragment['media_sequence']
+
+                ctx['fragment_index'] = frag_index
+
+                count = 0
+                headers = info_dict.get('http_headers', {})
+                if byte_range:
+                    headers['Range'] = 'bytes=%d-%d' % (byte_range['start'], byte_range['end'] - 1)
+                while count <= fragment_retries:
+                    try:
+                        success, frag_content = self._download_fragment(
+                            ctx, frag_url, info_dict, headers)
+                        if not success:
+                            return False, frag_index
+                        break
+                    except compat_urllib_error.HTTPError as err:
+                        # Unavailable (possibly temporary) fragments may be served.
+                        # First we try to retry then either skip or abort.
+                        # See https://github.com/ytdl-org/youtube-dl/issues/10165,
+                        # https://github.com/ytdl-org/youtube-dl/issues/10448).
+                        count += 1
+                        if count <= fragment_retries:
+                            self.report_retry_fragment(err, frag_index, count, fragment_retries)
+                if count > fragment_retries:
+                    return False, frag_index
+
+                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, info_dict.get('_decryption_key_url') or decrypt_info['URI'])).read()
+                    # Don't decrypt the content in tests since the data is explicitly truncated and it's not to a valid block
+                    # size (see https://github.com/ytdl-org/youtube-dl/pull/27660). Tests only care that the correct data downloaded,
+                    # not what it decrypts to.
+                    if not test:
+                        frag_content = AES.new(
+                            decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
+
+                return frag_content, frag_index
+
+            def append_fragment(frag_content, frag_index):
+                if frag_content:
+                    fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], frag_index)
+                    try:
+                        file, frag_sanitized = sanitize_open(fragment_filename, 'rb')
+                        ctx['fragment_filename_sanitized'] = frag_sanitized
+                        file.close()
+                        self._append_fragment(ctx, frag_content)
+                        return True
+                    except FileNotFoundError:
+                        if skip_unavailable_fragments:
+                            self.report_skip_fragment(frag_index)
+                            return True
+                        else:
+                            self.report_error(
+                                'fragment %s not found, unable to continue' % frag_index)
+                            return False
+                else:
+                    if skip_unavailable_fragments:
+                        self.report_skip_fragment(frag_index)
+                        return True
+                    else:
+                        self.report_error(
+                            'fragment %s not found, unable to continue' % frag_index)
+                        return False
+
+            max_workers = self.params.get('concurrent_fragment_downloads', 1)
+            if can_threaded_download and max_workers > 1:
+                self.report_warning('The download speed shown is only of one thread. This is a known issue')
+                with concurrent.futures.ThreadPoolExecutor(max_workers) as pool:
+                    futures = [pool.submit(download_fragment, fragment) for fragment in fragments]
+                    # timeout must be 0 to return instantly
+                    done, not_done = concurrent.futures.wait(futures, timeout=0)
+                    try:
+                        while not_done:
+                            # Check every 1 second for KeyboardInterrupt
+                            freshly_done, not_done = concurrent.futures.wait(not_done, timeout=1)
+                            done |= freshly_done
+                    except KeyboardInterrupt:
+                        for future in not_done:
+                            future.cancel()
+                        # timeout must be none to cancel
+                        concurrent.futures.wait(not_done, timeout=None)
+                        raise KeyboardInterrupt
+                results = [future.result() for future in futures]
+
+                for frag_content, frag_index in results:
+                    result = append_fragment(frag_content, frag_index)
+                    if not result:
+                        return False
+            else:
+                for fragment in fragments:
+                    frag_content, frag_index = download_fragment(fragment)
+                    result = append_fragment(frag_content, frag_index)
+                    if not result:
+                        return False
+
             self._finish_frag_download(ctx)
         return True

yt_dlp/extractor/applepodcasts.py

@@ -42,6 +42,7 @@ class ApplePodcastsIE(InfoExtractor):
         ember_data = self._parse_json(self._search_regex(
             r'id="shoebox-ember-data-store"[^>]*>\s*({.+?})\s*<',
             webpage, 'ember data'), episode_id)
+        ember_data = ember_data.get(episode_id) or ember_data
         episode = ember_data['data']['attributes']
         description = episode.get('description') or {}
 

yt_dlp/extractor/bandcamp.py

@@ -49,6 +49,7 @@ class BandcampIE(InfoExtractor):
             'uploader': 'Ben Prunty',
             'timestamp': 1396508491,
             'upload_date': '20140403',
+            'release_timestamp': 1396483200,
             'release_date': '20140403',
             'duration': 260.877,
             'track': 'Lanius (Battle)',
@@ -69,6 +70,7 @@ class BandcampIE(InfoExtractor):
             'uploader': 'Mastodon',
             'timestamp': 1322005399,
             'upload_date': '20111122',
+            'release_timestamp': 1076112000,
             'release_date': '20040207',
             'duration': 120.79,
             'track': 'Hail to Fire',
@@ -197,7 +199,7 @@ class BandcampIE(InfoExtractor):
             'thumbnail': thumbnail,
             'uploader': artist,
             'timestamp': timestamp,
-            'release_date': unified_strdate(tralbum.get('album_release_date')),
+            'release_timestamp': unified_timestamp(tralbum.get('album_release_date')),
             'duration': duration,
             'track': track,
             'track_number': track_number,

yt_dlp/extractor/bilibili.py

@@ -170,6 +170,7 @@ class BiliBiliIE(InfoExtractor):
             cid = js['result']['cid']
 
         headers = {
+            'Accept': 'application/json',
             'Referer': url
         }
         headers.update(self.geo_verification_headers())

yt_dlp/extractor/cbs.py

@@ -27,10 +27,10 @@ class CBSBaseIE(ThePlatformFeedIE):
 
 
 class CBSIE(CBSBaseIE):
-    _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
+    _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:(?:cbs|paramountplus)\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
 
     _TESTS = [{
-        'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
+        'url': 'https://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
         'info_dict': {
             'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
             'ext': 'mp4',
@@ -52,16 +52,19 @@ class CBSIE(CBSBaseIE):
     }, {
         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
         'only_matching': True,
+    }, {
+        'url': 'https://www.paramountplus.com/shows/all-rise/video/QmR1WhNkh1a_IrdHZrbcRklm176X_rVc/all-rise-space/',
+        'only_matching': True,
     }]
 
     def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
         items_data = self._download_xml(
-            'http://can.cbs.com/thunder/player/videoPlayerService.php',
+            'https://can.cbs.com/thunder/player/videoPlayerService.php',
             content_id, query={'partner': site, 'contentId': content_id})
         video_data = xpath_element(items_data, './/item')
         title = xpath_text(video_data, 'videoTitle', 'title') or xpath_text(video_data, 'videotitle', 'title')
         tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
-        tp_release_url = 'http://link.theplatform.com/s/' + tp_path
+        tp_release_url = 'https://link.theplatform.com/s/' + tp_path
 
         asset_types = []
         subtitles = {}

yt_dlp/extractor/common.py

@@ -231,8 +231,9 @@ class InfoExtractor(object):
     uploader:       Full name of the video uploader.
     license:        License name the video is licensed under.
     creator:        The creator of the video.
+    release_timestamp: UNIX timestamp of the moment the video was released.
     release_date:   The date (YYYYMMDD) when the video was released.
-    timestamp:      UNIX timestamp of the moment the video became available.
+    timestamp:      UNIX timestamp of the moment the video was uploaded
     upload_date:    Video upload date (YYYYMMDD).
                     If not explicitly set, calculated from timestamp.
     uploader_id:    Nickname or id of the video uploader.
@@ -264,6 +265,7 @@ class InfoExtractor(object):
                     properties (all but one of text or html optional):
                         * "author" - human-readable name of the comment author
                         * "author_id" - user ID of the comment author
+                        * "author_thumbnail" - The thumbnail of the comment author
                         * "id" - Comment ID
                         * "html" - Comment as HTML
                         * "text" - Plain text of the comment
@@ -271,6 +273,12 @@ class InfoExtractor(object):
                         * "parent" - ID of the comment this one is replying to.
                                      Set to "root" to indicate that this is a
                                      comment to the original video.
+                        * "like_count" - Number of positive ratings of the comment
+                        * "dislike_count" - Number of negative ratings of the comment
+                        * "is_favorited" - Whether the comment is marked as
+                                           favorite by the video uploader
+                        * "author_is_uploader" - Whether the comment is made by
+                                                 the video uploader
     age_limit:      Age restriction for the video, as an integer (years)
     webpage_url:    The URL to the video webpage, if given to yt-dlp it
                     should allow to get the same result again. (It will be set
@@ -1849,8 +1857,9 @@ class InfoExtractor(object):
 
     def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
                               entry_protocol='m3u8', preference=None, quality=None,
-                              m3u8_id=None, live=False, note=None, errnote=None,
+                              m3u8_id=None, note=None, errnote=None,
-                              fatal=True, data=None, headers={}, query={}):
+                              fatal=True, live=False, data=None, headers={},
+                              query={}):
         res = self._download_webpage_handle(
             m3u8_url, video_id,
             note=note or 'Downloading m3u8 information',
@@ -2050,11 +2059,11 @@ class InfoExtractor(object):
                 playlist_formats = _extract_m3u8_playlist_formats(manifest_url, video_id=video_id,
                                                                   fatal=fatal, data=data, headers=headers)
 
-                for format in playlist_formats:
+                for frmt in playlist_formats:
                     format_id = []
                     if m3u8_id:
                         format_id.append(m3u8_id)
-                    format_index = format.get('index')
+                    format_index = frmt.get('index')
                     stream_name = build_stream_name()
                     # Bandwidth of live streams may differ over time thus making
                     # format_id unpredictable. So it's better to keep provided
@@ -2109,6 +2118,8 @@ class InfoExtractor(object):
                             # TODO: update acodec for audio only formats with
                             # the same GROUP-ID
                             f['acodec'] = 'none'
+                    if not f.get('ext'):
+                        f['ext'] = 'm4a' if f.get('vcodec') == 'none' else 'mp4'
                     formats.append(f)
 
                     # for DailyMotion

yt_dlp/extractor/extractors.py

@@ -450,10 +450,7 @@ from .gamestar import GameStarIE
 from .gaskrank import GaskrankIE
 from .gazeta import GazetaIE
 from .gdcvault import GDCVaultIE
-from .gedi import (
+from .gedidigital import GediDigitalIE
-    GediIE,
-    GediEmbedsIE,
-)
 from .generic import GenericIE
 from .gfycat import GfycatIE
 from .giantbomb import GiantBombIE
@@ -735,6 +732,8 @@ from .mtv import (
     MTVServicesEmbeddedIE,
     MTVDEIE,
     MTVJapanIE,
+    MTVItaliaIE,
+    MTVItaliaProgrammaIE,
 )
 from .muenchentv import MuenchenTVIE
 from .mwave import MwaveIE, MwaveMeetGreetIE
@@ -954,6 +953,7 @@ from .plays import PlaysTVIE
 from .playtvak import PlaytvakIE
 from .playvid import PlayvidIE
 from .playwire import PlaywireIE
+from .plutotv import PlutoTVIE
 from .pluralsight import (
     PluralsightIE,
     PluralsightCourseIE,
@@ -1560,6 +1560,7 @@ from .weibo import (
     WeiboMobileIE
 )
 from .weiqitv import WeiqiTVIE
+from .wimtv import WimTVIE
 from .wistia import (
     WistiaIE,
     WistiaPlaylistIE,
@@ -1669,5 +1670,6 @@ from .zdf import ZDFIE, ZDFChannelIE
 from .zhihu import ZhihuIE
 from .zingmp3 import ZingMp3IE
 from .zee5 import Zee5IE
+from .zee5 import Zee5SeriesIE
 from .zoom import ZoomIE
 from .zype import ZypeIE

yt_dlp/extractor/fujitv.py

@@ -17,7 +17,7 @@ class FujiTVFODPlus7IE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
         formats = self._extract_m3u8_formats(
-            self._BASE_URL + 'abr/pc_html5/%s.m3u8' % video_id, video_id)
+            self._BASE_URL + 'abr/pc_html5/%s.m3u8' % video_id, video_id, 'mp4')
         for f in formats:
             wh = self._BITRATE_MAP.get(f.get('tbr'))
             if wh:

yt_dlp/extractor/gedi.py

@@ -1,266 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..compat import compat_str
-from ..utils import (
-    base_url,
-    url_basename,
-    urljoin,
-)
-
-
-class GediBaseIE(InfoExtractor):
-    @staticmethod
-    def _clean_audio_fmts(formats):
-        unique_formats = []
-        for f in formats:
-            if 'acodec' in f:
-                unique_formats.append(f)
-        formats[:] = unique_formats
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-        player_data = re.findall(
-            r'PlayerFactory\.setParam\(\'(?P<type>.+?)\',\s*\'(?P<name>.+?)\',\s*\'(?P<val>.+?)\'\);',
-            webpage)
-
-        formats = []
-        audio_fmts = []
-        hls_fmts = []
-        http_fmts = []
-        title = ''
-        thumb = ''
-
-        fmt_reg = r'(?P<t>video|audio)-(?P<p>rrtv|hls)-(?P<h>[\w\d]+)(?:-(?P<br>[\w\d]+))?$'
-        br_reg = r'video-rrtv-(?P<br>\d+)-'
-
-        for t, n, v in player_data:
-            if t == 'format':
-                m = re.match(fmt_reg, n)
-                if m:
-                    # audio formats
-                    if m.group('t') == 'audio':
-                        if m.group('p') == 'hls':
-                            audio_fmts.extend(self._extract_m3u8_formats(
-                                v, video_id, 'm4a', m3u8_id='hls', fatal=False))
-                        elif m.group('p') == 'rrtv':
-                            audio_fmts.append({
-                                'format_id': 'mp3',
-                                'url': v,
-                                'tbr': 128,
-                                'ext': 'mp3',
-                                'vcodec': 'none',
-                                'acodec': 'mp3',
-                            })
-
-                    # video formats
-                    elif m.group('t') == 'video':
-                        # hls manifest video
-                        if m.group('p') == 'hls':
-                            hls_fmts.extend(self._extract_m3u8_formats(
-                                v, video_id, 'mp4', m3u8_id='hls', fatal=False))
-                        # direct mp4 video
-                        elif m.group('p') == 'rrtv':
-                            if not m.group('br'):
-                                mm = re.search(br_reg, v)
-                            http_fmts.append({
-                                'format_id': 'https-' + m.group('h'),
-                                'protocol': 'https',
-                                'url': v,
-                                'tbr': int(m.group('br')) if m.group('br') else
-                                (int(mm.group('br')) if mm.group('br') else 0),
-                                'height': int(m.group('h'))
-                            })
-
-            elif t == 'param':
-                if n == 'videotitle':
-                    title = v
-                if n == 'image_full_play':
-                    thumb = v
-
-        title = self._og_search_title(webpage) if title == '' else title
-
-        # clean weird char
-        title = compat_str(title).encode('utf8', 'replace').replace(b'\xc3\x82', b'').decode('utf8', 'replace')
-
-        if audio_fmts:
-            self._clean_audio_fmts(audio_fmts)
-            self._sort_formats(audio_fmts)
-        if hls_fmts:
-            self._sort_formats(hls_fmts)
-        if http_fmts:
-            self._sort_formats(http_fmts)
-
-        formats.extend(audio_fmts)
-        formats.extend(hls_fmts)
-        formats.extend(http_fmts)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': self._html_search_meta('twitter:description', webpage),
-            'thumbnail': thumb,
-            'formats': formats,
-        }
-
-
-class GediIE(GediBaseIE):
-    _VALID_URL = r'''(?x)https?://video\.
-                    (?:
-                        (?:espresso\.)?repubblica
-                        |lastampa
-                        |huffingtonpost
-                        |ilsecoloxix
-                        |iltirreno
-                        |messaggeroveneto
-                        |ilpiccolo
-                        |gazzettadimantova
-                        |mattinopadova
-                        |laprovinciapavese
-                        |tribunatreviso
-                        |nuovavenezia
-                        |gazzettadimodena
-                        |lanuovaferrara
-                        |corrierealpi
-                        |lasentinella
-                    )
-                    (?:\.gelocal)?\.it/(?!embed/).+?/(?P<id>[\d/]+)(?:\?|\&|$)'''
-    _TESTS = [{
-        'url': 'https://video.lastampa.it/politica/il-paradosso-delle-regionali-la-lega-vince-ma-sembra-aver-perso/121559/121683',
-        'md5': '84658d7fb9e55a6e57ecc77b73137494',
-        'info_dict': {
-            'id': '121559/121683',
-            'ext': 'mp4',
-            'title': 'Il paradosso delle Regionali: ecco perché la Lega vince ma sembra aver perso',
-            'description': 'md5:de7f4d6eaaaf36c153b599b10f8ce7ca',
-            'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-social-play\.jpg$',
-        },
-    }, {
-        'url': 'https://video.repubblica.it/motori/record-della-pista-a-spa-francorchamps-la-pagani-huayra-roadster-bc-stupisce/367415/367963',
-        'md5': 'e763b94b7920799a0e0e23ffefa2d157',
-        'info_dict': {
-            'id': '367415/367963',
-            'ext': 'mp4',
-            'title': 'Record della pista a Spa Francorchamps, la Pagani Huayra Roadster BC stupisce',
-            'description': 'md5:5deb503cefe734a3eb3f07ed74303920',
-            'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-social-play\.jpg$',
-        },
-    }, {
-        'url': 'https://video.ilsecoloxix.it/sport/cassani-e-i-brividi-azzurri-ai-mondiali-di-imola-qui-mi-sono-innamorato-del-ciclismo-da-ragazzino-incredibile-tornarci-da-ct/66184/66267',
-        'md5': 'e48108e97b1af137d22a8469f2019057',
-        'info_dict': {
-            'id': '66184/66267',
-            'ext': 'mp4',
-            'title': 'Cassani e i brividi azzurri ai Mondiali di Imola: \\"Qui mi sono innamorato del ciclismo da ragazzino, incredibile tornarci da ct\\"',
-            'description': 'md5:fc9c50894f70a2469bb9b54d3d0a3d3b',
-            'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-social-play\.jpg$',
-        },
-    }, {
-        'url': 'https://video.iltirreno.gelocal.it/sport/dentro-la-notizia-ferrari-cosa-succede-a-maranello/141059/142723',
-        'md5': 'a6e39f3bdc1842bbd92abbbbef230817',
-        'info_dict': {
-            'id': '141059/142723',
-            'ext': 'mp4',
-            'title': 'Dentro la notizia - Ferrari, cosa succede a Maranello',
-            'description': 'md5:9907d65b53765681fa3a0b3122617c1f',
-            'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-social-play\.jpg$',
-        },
-    }]
-
-
-class GediEmbedsIE(GediBaseIE):
-    _VALID_URL = r'''(?x)https?://video\.
-                    (?:
-                        (?:espresso\.)?repubblica
-                        |lastampa
-                        |huffingtonpost
-                        |ilsecoloxix
-                        |iltirreno
-                        |messaggeroveneto
-                        |ilpiccolo
-                        |gazzettadimantova
-                        |mattinopadova
-                        |laprovinciapavese
-                        |tribunatreviso
-                        |nuovavenezia
-                        |gazzettadimodena
-                        |lanuovaferrara
-                        |corrierealpi
-                        |lasentinella
-                    )
-                    (?:\.gelocal)?\.it/embed/.+?/(?P<id>[\d/]+)(?:\?|\&|$)'''
-    _TESTS = [{
-        'url': 'https://video.huffingtonpost.it/embed/politica/cotticelli-non-so-cosa-mi-sia-successo-sto-cercando-di-capire-se-ho-avuto-un-malore/29312/29276?responsive=true&el=video971040871621586700',
-        'md5': 'f4ac23cadfea7fef89bea536583fa7ed',
-        'info_dict': {
-            'id': '29312/29276',
-            'ext': 'mp4',
-            'title': 'Cotticelli: \\"Non so cosa mi sia successo. Sto cercando di capire se ho avuto un malore\\"',
-            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
-            'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-social-play\.jpg$',
-        },
-    }, {
-        'url': 'https://video.espresso.repubblica.it/embed/tutti-i-video/01-ted-villa/14772/14870&width=640&height=360',
-        'md5': '0391c2c83c6506581003aaf0255889c0',
-        'info_dict': {
-            'id': '14772/14870',
-            'ext': 'mp4',
-            'title': 'Festival EMERGENCY, Villa: «La buona informazione aiuta la salute» (14772-14870)',
-            'description': 'md5:2bce954d278248f3c950be355b7c2226',
-            'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-social-play\.jpg$',
-        },
-    }]
-
-    @staticmethod
-    def _sanitize_urls(urls):
-        # add protocol if missing
-        for i, e in enumerate(urls):
-            if e.startswith('//'):
-                urls[i] = 'https:%s' % e
-        # clean iframes urls
-        for i, e in enumerate(urls):
-            urls[i] = urljoin(base_url(e), url_basename(e))
-        return urls
-
-    @staticmethod
-    def _extract_urls(webpage):
-        entries = [
-            mobj.group('url')
-            for mobj in re.finditer(r'''(?x)
-            (?:
-                data-frame-src=|
-                <iframe[^\n]+src=
-            )
-            (["'])
-                (?P<url>https?://video\.
-                    (?:
-                        (?:espresso\.)?repubblica
-                        |lastampa
-                        |huffingtonpost
-                        |ilsecoloxix
-                        |iltirreno
-                        |messaggeroveneto
-                        |ilpiccolo
-                        |gazzettadimantova
-                        |mattinopadova
-                        |laprovinciapavese
-                        |tribunatreviso
-                        |nuovavenezia
-                        |gazzettadimodena
-                        |lanuovaferrara
-                        |corrierealpi
-                        |lasentinella
-                    )
-                    (?:\.gelocal)?\.it/embed/.+?)
-            \1''', webpage)]
-        return GediEmbedsIE._sanitize_urls(entries)
-
-    @staticmethod
-    def _extract_url(webpage):
-        urls = GediEmbedsIE._extract_urls(webpage)
-        return urls[0] if urls else None

yt_dlp/extractor/gedidigital.py

@@ -0,0 +1,210 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    base_url,
+    determine_ext,
+    int_or_none,
+    url_basename,
+    urljoin,
+)
+
+
+class GediDigitalIE(InfoExtractor):
+    _VALID_URL = r'''(?x)(?P<url>(?:https?:)//video\.
+        (?:
+            (?:
+                (?:espresso\.)?repubblica
+                |lastampa
+                |ilsecoloxix
+                |huffingtonpost
+            )|
+            (?:
+                iltirreno
+                |messaggeroveneto
+                |ilpiccolo
+                |gazzettadimantova
+                |mattinopadova
+                |laprovinciapavese
+                |tribunatreviso
+                |nuovavenezia
+                |gazzettadimodena
+                |lanuovaferrara
+                |corrierealpi
+                |lasentinella
+            )\.gelocal
+        )\.it(?:/[^/]+){2,4}/(?P<id>\d+))(?:$|[?&].*)'''
+    _TESTS = [{
+        'url': 'https://video.lastampa.it/politica/il-paradosso-delle-regionali-la-lega-vince-ma-sembra-aver-perso/121559/121683',
+        'md5': '84658d7fb9e55a6e57ecc77b73137494',
+        'info_dict': {
+            'id': '121683',
+            'ext': 'mp4',
+            'title': 'Il paradosso delle Regionali: ecco perché la Lega vince ma sembra aver perso',
+            'description': 'md5:de7f4d6eaaaf36c153b599b10f8ce7ca',
+            'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-full-.+?\.jpg$',
+            'duration': 125,
+        },
+    }, {
+        'url': 'https://video.huffingtonpost.it/embed/politica/cotticelli-non-so-cosa-mi-sia-successo-sto-cercando-di-capire-se-ho-avuto-un-malore/29312/29276?responsive=true&el=video971040871621586700',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.espresso.repubblica.it/embed/tutti-i-video/01-ted-villa/14772/14870&width=640&height=360',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.repubblica.it/motori/record-della-pista-a-spa-francorchamps-la-pagani-huayra-roadster-bc-stupisce/367415/367963',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.ilsecoloxix.it/sport/cassani-e-i-brividi-azzurri-ai-mondiali-di-imola-qui-mi-sono-innamorato-del-ciclismo-da-ragazzino-incredibile-tornarci-da-ct/66184/66267',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.iltirreno.gelocal.it/sport/dentro-la-notizia-ferrari-cosa-succede-a-maranello/141059/142723',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.messaggeroveneto.gelocal.it/locale/maria-giovanna-elmi-covid-vaccino/138155/139268',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.ilpiccolo.gelocal.it/dossier/big-john/dinosauro-big-john-al-via-le-visite-guidate-a-trieste/135226/135751',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.gazzettadimantova.gelocal.it/locale/dal-ponte-visconteo-di-valeggio-l-and-8217sos-dei-ristoratori-aprire-anche-a-cena/137310/137818',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.mattinopadova.gelocal.it/dossier/coronavirus-in-veneto/covid-a-vo-un-anno-dopo-un-cuore-tricolore-per-non-dimenticare/138402/138964',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.laprovinciapavese.gelocal.it/locale/mede-zona-rossa-via-alle-vaccinazioni-per-gli-over-80/137545/138120',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.tribunatreviso.gelocal.it/dossier/coronavirus-in-veneto/ecco-le-prima-vaccinazioni-di-massa-nella-marca/134485/135024',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.nuovavenezia.gelocal.it/locale/camion-troppo-alto-per-il-ponte-ferroviario-perde-il-carico/135734/136266',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.gazzettadimodena.gelocal.it/locale/modena-scoperta-la-proteina-che-predice-il-livello-di-gravita-del-covid/139109/139796',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.lanuovaferrara.gelocal.it/locale/due-bombole-di-gpl-aperte-e-abbandonate-i-vigili-bruciano-il-gas/134391/134957',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.corrierealpi.gelocal.it/dossier/cortina-2021-i-mondiali-di-sci-alpino/mondiali-di-sci-il-timelapse-sulla-splendida-olympia/133760/134331',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.lasentinella.gelocal.it/locale/vestigne-centra-un-auto-e-si-ribalta/138931/139466',
+        'only_matching': True,
+    }, {
+        'url': 'https://video.espresso.repubblica.it/tutti-i-video/01-ted-villa/14772',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _sanitize_urls(urls):
+        # add protocol if missing
+        for i, e in enumerate(urls):
+            if e.startswith('//'):
+                urls[i] = 'https:%s' % e
+        # clean iframes urls
+        for i, e in enumerate(urls):
+            urls[i] = urljoin(base_url(e), url_basename(e))
+        return urls
+
+    @staticmethod
+    def _extract_urls(webpage):
+        entries = [
+            mobj.group('eurl')
+            for mobj in re.finditer(r'''(?x)
+            (?:
+                data-frame-src=|
+                <iframe[^\n]+src=
+            )
+            (["'])(?P<eurl>%s)\1''' % GediDigitalIE._VALID_URL, webpage)]
+        return GediDigitalIE._sanitize_urls(entries)
+
+    @staticmethod
+    def _extract_url(webpage):
+        urls = GediDigitalIE._extract_urls(webpage)
+        return urls[0] if urls else None
+
+    @staticmethod
+    def _clean_formats(formats):
+        format_urls = set()
+        clean_formats = []
+        for f in formats:
+            if f['url'] not in format_urls:
+                if f.get('audio_ext') != 'none' and not f.get('acodec'):
+                    continue
+                format_urls.add(f['url'])
+                clean_formats.append(f)
+        formats[:] = clean_formats
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        url = re.match(self._VALID_URL, url).group('url')
+        webpage = self._download_webpage(url, video_id)
+        title = self._html_search_meta(
+            ['twitter:title', 'og:title'], webpage, fatal=True)
+        player_data = re.findall(
+            r"PlayerFactory\.setParam\('(?P<type>format|param)',\s*'(?P<name>[^']+)',\s*'(?P<val>[^']+)'\);",
+            webpage)
+
+        formats = []
+        duration = thumb = None
+        for t, n, v in player_data:
+            if t == 'format':
+                if n in ('video-hds-vod-ec', 'video-hls-vod-ec', 'video-viralize', 'video-youtube-pfp'):
+                    continue
+                elif n.endswith('-vod-ak'):
+                    formats.extend(self._extract_akamai_formats(
+                        v, video_id, {'http': 'media.gedidigital.it'}))
+                else:
+                    ext = determine_ext(v)
+                    if ext == 'm3u8':
+                        formats.extend(self._extract_m3u8_formats(
+                            v, video_id, 'mp4', 'm3u8_native', m3u8_id=n, fatal=False))
+                        continue
+                    f = {
+                        'format_id': n,
+                        'url': v,
+                    }
+                    if ext == 'mp3':
+                        abr = int_or_none(self._search_regex(
+                            r'-mp3-audio-(\d+)', v, 'abr', default=None))
+                        f.update({
+                            'abr': abr,
+                            'tbr': abr,
+                            'acodec': ext,
+                            'vcodec': 'none'
+                        })
+                    else:
+                        mobj = re.match(r'^video-rrtv-(\d+)(?:-(\d+))?$', n)
+                        if mobj:
+                            f.update({
+                                'height': int(mobj.group(1)),
+                                'vbr': int_or_none(mobj.group(2)),
+                            })
+                        if not f.get('vbr'):
+                            f['vbr'] = int_or_none(self._search_regex(
+                                r'-video-rrtv-(\d+)', v, 'abr', default=None))
+                    formats.append(f)
+            elif t == 'param':
+                if n in ['image_full', 'image']:
+                    thumb = v
+                elif n == 'videoDuration':
+                    duration = int_or_none(v)
+
+        self._clean_formats(formats)
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': self._html_search_meta(
+                ['twitter:description', 'og:description', 'description'], webpage),
+            'thumbnail': thumb or self._og_search_thumbnail(webpage),
+            'formats': formats,
+            'duration': duration,
+        }

yt_dlp/extractor/generic.py

@@ -127,13 +127,14 @@ from .expressen import ExpressenIE
 from .zype import ZypeIE
 from .odnoklassniki import OdnoklassnikiIE
 from .kinja import KinjaEmbedIE
-from .gedi import GediEmbedsIE
+from .gedidigital import GediDigitalIE
 from .rcs import RCSEmbedsIE
 from .bitchute import BitChuteIE
 from .rumble import RumbleEmbedIE
 from .arcpublishing import ArcPublishingIE
 from .medialaan import MedialaanIE
 from .simplecast import SimplecastIE
+from .wimtv import WimTVIE
 
 
 class GenericIE(InfoExtractor):
@@ -2250,6 +2251,15 @@ class GenericIE(InfoExtractor):
             },
             'playlist_mincount': 52,
         },
+        {
+            # WimTv embed player
+            'url': 'http://www.msmotor.tv/wearefmi-pt-2-2021/',
+            'info_dict': {
+                'id': 'wearefmi-pt-2-2021',
+                'title': '#WEAREFMI – PT.2 – 2021 – MsMotorTV',
+            },
+            'playlist_count': 1,
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -3339,17 +3349,22 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 zype_urls, video_id, video_title, ie=ZypeIE.ie_key())
 
-        # Look for RCS media group embeds
+        gedi_urls = GediDigitalIE._extract_urls(webpage)
-        gedi_urls = GediEmbedsIE._extract_urls(webpage)
         if gedi_urls:
             return self.playlist_from_matches(
-                gedi_urls, video_id, video_title, ie=GediEmbedsIE.ie_key())
+                gedi_urls, video_id, video_title, ie=GediDigitalIE.ie_key())
 
+        # Look for RCS media group embeds
         rcs_urls = RCSEmbedsIE._extract_urls(webpage)
         if rcs_urls:
             return self.playlist_from_matches(
                 rcs_urls, video_id, video_title, ie=RCSEmbedsIE.ie_key())
 
+        wimtv_urls = WimTVIE._extract_urls(webpage)
+        if wimtv_urls:
+            return self.playlist_from_matches(
+                wimtv_urls, video_id, video_title, ie=WimTVIE.ie_key())
+
         bitchute_urls = BitChuteIE._extract_urls(webpage)
         if bitchute_urls:
             return self.playlist_from_matches(

yt_dlp/extractor/lbry.py

@@ -6,8 +6,10 @@ import json
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_parse_qs,
     compat_str,
     compat_urllib_parse_unquote,
+    compat_urllib_parse_urlparse,
 )
 from ..utils import (
     determine_ext,
@@ -21,9 +23,9 @@ from ..utils import (
 
 
 class LBRYBaseIE(InfoExtractor):
-    _BASE_URL_REGEX = r'https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/'
+    _BASE_URL_REGEX = r'(?:https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/|lbry://)'
     _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
-    _OPT_CLAIM_ID = '[^:/?#&]+(?::%s)?' % _CLAIM_ID_REGEX
+    _OPT_CLAIM_ID = '[^:/?#&]+(?:[:#]%s)?' % _CLAIM_ID_REGEX
     _SUPPORTED_STREAM_TYPES = ['video', 'audio']
 
     def _call_api_proxy(self, method, display_id, params, resource):
@@ -41,7 +43,9 @@ class LBRYBaseIE(InfoExtractor):
             'resolve', display_id, {'urls': url}, resource)[url]
 
     def _permanent_url(self, url, claim_name, claim_id):
-        return urljoin(url, '/%s:%s' % (claim_name, claim_id))
+        return urljoin(
+            url.replace('lbry://', 'https://lbry.tv/'),
+            '/%s:%s' % (claim_name, claim_id))
 
     def _parse_stream(self, stream, url):
         stream_value = stream.get('value') or {}
@@ -60,6 +64,7 @@ class LBRYBaseIE(InfoExtractor):
             'description': stream_value.get('description'),
             'license': stream_value.get('license'),
             'timestamp': int_or_none(stream.get('timestamp')),
+            'release_timestamp': int_or_none(stream_value.get('release_time')),
             'tags': stream_value.get('tags'),
             'duration': int_or_none(media.get('duration')),
             'channel': try_get(signing_channel, lambda x: x['value']['title']),
@@ -92,6 +97,8 @@ class LBRYIE(LBRYBaseIE):
             'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
             'timestamp': 1595694354,
             'upload_date': '20200725',
+            'release_timestamp': 1595340697,
+            'release_date': '20200721',
             'width': 1280,
             'height': 720,
         }
@@ -106,6 +113,8 @@ class LBRYIE(LBRYBaseIE):
             'description': 'md5:661ac4f1db09f31728931d7b88807a61',
             'timestamp': 1591312601,
             'upload_date': '20200604',
+            'release_timestamp': 1591312421,
+            'release_date': '20200604',
             'tags': list,
             'duration': 2570,
             'channel': 'The LBRY Foundation',
@@ -137,6 +146,9 @@ class LBRYIE(LBRYBaseIE):
     }, {
         'url': 'https://lbry.tv/@lacajadepandora:a/TRUMP-EST%C3%81-BIEN-PUESTO-con-Pilar-Baselga,-Carlos-Senra,-Luis-Palacios-(720p_30fps_H264-192kbit_AAC):1',
         'only_matching': True,
+    }, {
+        'url': 'lbry://@lbry#3f/odysee#7',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -166,7 +178,7 @@ class LBRYIE(LBRYBaseIE):
 
 class LBRYChannelIE(LBRYBaseIE):
     IE_NAME = 'lbry:channel'
-    _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?#&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
+    _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
     _TESTS = [{
         'url': 'https://lbry.tv/@LBRYFoundation:0',
         'info_dict': {
@@ -178,20 +190,24 @@ class LBRYChannelIE(LBRYBaseIE):
     }, {
         'url': 'https://lbry.tv/@LBRYFoundation',
         'only_matching': True,
+    }, {
+        'url': 'lbry://@lbry#3f',
+        'only_matching': True,
     }]
     _PAGE_SIZE = 50
 
-    def _fetch_page(self, claim_id, url, page):
+    def _fetch_page(self, claim_id, url, params, page):
         page += 1
+        page_params = {
+            'channel_ids': [claim_id],
+            'claim_type': 'stream',
+            'no_totals': True,
+            'page': page,
+            'page_size': self._PAGE_SIZE,
+        }
+        page_params.update(params)
         result = self._call_api_proxy(
-            'claim_search', claim_id, {
+            'claim_search', claim_id, page_params, 'page %d' % page)
-                'channel_ids': [claim_id],
-                'claim_type': 'stream',
-                'no_totals': True,
-                'page': page,
-                'page_size': self._PAGE_SIZE,
-                'stream_types': self._SUPPORTED_STREAM_TYPES,
-            }, 'page %d' % page)
         for item in (result.get('items') or []):
             stream_claim_name = item.get('name')
             stream_claim_id = item.get('claim_id')
@@ -212,8 +228,31 @@ class LBRYChannelIE(LBRYBaseIE):
         result = self._resolve_url(
             'lbry://' + display_id, display_id, 'channel')
         claim_id = result['claim_id']
+        qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
+        content = qs.get('content', [None])[0]
+        params = {
+            'fee_amount': qs.get('fee_amount', ['>=0'])[0],
+            'order_by': {
+                'new': ['release_time'],
+                'top': ['effective_amount'],
+                'trending': ['trending_group', 'trending_mixed'],
+            }[qs.get('order', ['new'])[0]],
+            'stream_types': [content] if content in ['audio', 'video'] else self._SUPPORTED_STREAM_TYPES,
+        }
+        duration = qs.get('duration', [None])[0]
+        if duration:
+            params['duration'] = {
+                'long': '>=1200',
+                'short': '<=240',
+            }[duration]
+        language = qs.get('language', ['all'])[0]
+        if language != 'all':
+            languages = [language]
+            if language == 'en':
+                languages.append('none')
+            params['any_languages'] = languages
         entries = OnDemandPagedList(
-            functools.partial(self._fetch_page, claim_id, url),
+            functools.partial(self._fetch_page, claim_id, url, params),
             self._PAGE_SIZE)
         result_value = result.get('value') or {}
         return self.playlist_result(

yt_dlp/extractor/mtv.py

@@ -14,6 +14,7 @@ from ..utils import (
     fix_xml_ampersands,
     float_or_none,
     HEADRequest,
+    int_or_none,
     RegexNotFoundError,
     sanitized_Request,
     strip_or_none,
@@ -176,6 +177,22 @@ class MTVServicesInfoExtractor(InfoExtractor):
             raise ExtractorError('Could not find video title')
         title = title.strip()
 
+        series = find_xpath_attr(
+            itemdoc, './/{http://search.yahoo.com/mrss/}category',
+            'scheme', 'urn:mtvn:franchise')
+        season = find_xpath_attr(
+            itemdoc, './/{http://search.yahoo.com/mrss/}category',
+            'scheme', 'urn:mtvn:seasonN')
+        episode = find_xpath_attr(
+            itemdoc, './/{http://search.yahoo.com/mrss/}category',
+            'scheme', 'urn:mtvn:episodeN')
+        series = series.text if series is not None else None
+        season = season.text if season is not None else None
+        episode = episode.text if episode is not None else None
+        if season and episode:
+            # episode number includes season, so remove it
+            episode = re.sub(r'^%s' % season, '', episode)
+
         # This a short id that's used in the webpage urls
         mtvn_id = None
         mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
@@ -201,6 +218,9 @@ class MTVServicesInfoExtractor(InfoExtractor):
             'description': description,
             'duration': float_or_none(content_el.attrib.get('duration')),
             'timestamp': timestamp,
+            'series': series,
+            'season_number': int_or_none(season),
+            'episode_number': int_or_none(episode),
         }
 
     def _get_feed_query(self, uri):
@@ -483,3 +503,152 @@ class MTVDEIE(MTVServicesInfoExtractor):
             'arcEp': 'mtv.de',
             'mgid': uri,
         }
+
+
+class MTVItaliaIE(MTVServicesInfoExtractor):
+    IE_NAME = 'mtv.it'
+    _VALID_URL = r'https?://(?:www\.)?mtv\.it/(?:episodi|video|musica)/(?P<id>[0-9a-z]+)'
+    _TESTS = [{
+        'url': 'http://www.mtv.it/episodi/24bqab/mario-una-serie-di-maccio-capatonda-cavoli-amario-episodio-completo-S1-E1',
+        'info_dict': {
+            'id': '0f0fc78e-45fc-4cce-8f24-971c25477530',
+            'ext': 'mp4',
+            'title': 'Cavoli amario (episodio completo)',
+            'description': 'md5:4962bccea8fed5b7c03b295ae1340660',
+            'series': 'Mario - Una Serie Di Maccio Capatonda',
+            'season_number': 1,
+            'episode_number': 1,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
+    _GEO_COUNTRIES = ['IT']
+    _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
+
+    def _get_feed_query(self, uri):
+        return {
+            'arcEp': 'mtv.it',
+            'mgid': uri,
+        }
+
+
+class MTVItaliaProgrammaIE(MTVItaliaIE):
+    IE_NAME = 'mtv.it:programma'
+    _VALID_URL = r'https?://(?:www\.)?mtv\.it/(?:programmi|playlist)/(?P<id>[0-9a-z]+)'
+    _TESTS = [{
+        # program page: general
+        'url': 'http://www.mtv.it/programmi/s2rppv/mario-una-serie-di-maccio-capatonda',
+        'info_dict': {
+            'id': 'a6f155bc-8220-4640-aa43-9b95f64ffa3d',
+            'title': 'Mario - Una Serie Di Maccio Capatonda',
+            'description': 'md5:72fbffe1f77ccf4e90757dd4e3216153',
+        },
+        'playlist_count': 2,
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # program page: specific season
+        'url': 'http://www.mtv.it/programmi/d9ncjf/mario-una-serie-di-maccio-capatonda-S2',
+        'info_dict': {
+            'id': '4deeb5d8-f272-490c-bde2-ff8d261c6dd1',
+            'title': 'Mario - Una Serie Di Maccio Capatonda - Stagione 2',
+        },
+        'playlist_count': 34,
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # playlist page + redirect
+        'url': 'http://www.mtv.it/playlist/sexy-videos/ilctal',
+        'info_dict': {
+            'id': 'dee8f9ee-756d-493b-bf37-16d1d2783359',
+            'title': 'Sexy Videos',
+        },
+        'playlist_mincount': 145,
+        'params': {
+            'skip_download': True,
+        },
+    }]
+    _GEO_COUNTRIES = ['IT']
+    _FEED_URL = 'http://www.mtv.it/feeds/triforce/manifest/v8'
+
+    def _get_entries(self, title, url):
+        while True:
+            pg = self._search_regex(r'/(\d+)$', url, 'entries', '1')
+            entries = self._download_json(url, title, 'page %s' % pg)
+            url = try_get(
+                entries, lambda x: x['result']['nextPageURL'], compat_str)
+            entries = try_get(
+                entries, (
+                    lambda x: x['result']['data']['items'],
+                    lambda x: x['result']['data']['seasons']),
+                list)
+            for entry in entries or []:
+                if entry.get('canonicalURL'):
+                    yield self.url_result(entry['canonicalURL'])
+            if not url:
+                break
+
+    def _real_extract(self, url):
+        query = {'url': url}
+        info_url = update_url_query(self._FEED_URL, query)
+        video_id = self._match_id(url)
+        info = self._download_json(info_url, video_id).get('manifest')
+
+        redirect = try_get(
+            info, lambda x: x['newLocation']['url'], compat_str)
+        if redirect:
+            return self.url_result(redirect)
+
+        title = info.get('title')
+        video_id = try_get(
+            info, lambda x: x['reporting']['itemId'], compat_str)
+        parent_id = try_get(
+            info, lambda x: x['reporting']['parentId'], compat_str)
+
+        playlist_url = current_url = None
+        for z in (info.get('zones') or {}).values():
+            if z.get('moduleName') in ('INTL_M304', 'INTL_M209'):
+                info_url = z.get('feed')
+            if z.get('moduleName') in ('INTL_M308', 'INTL_M317'):
+                playlist_url = playlist_url or z.get('feed')
+            if z.get('moduleName') in ('INTL_M300',):
+                current_url = current_url or z.get('feed')
+
+        if not info_url:
+            raise ExtractorError('No info found')
+
+        if video_id == parent_id:
+            video_id = self._search_regex(
+                r'([^\/]+)/[^\/]+$', info_url, 'video_id')
+
+        info = self._download_json(info_url, video_id, 'Show infos')
+        info = try_get(info, lambda x: x['result']['data'], dict)
+        title = title or try_get(
+            info, (
+                lambda x: x['title'],
+                lambda x: x['headline']),
+            compat_str)
+        description = try_get(info, lambda x: x['content'], compat_str)
+
+        if current_url:
+            season = try_get(
+                self._download_json(playlist_url, video_id, 'Seasons info'),
+                lambda x: x['result']['data'], dict)
+            current = try_get(
+                season, lambda x: x['currentSeason'], compat_str)
+            seasons = try_get(
+                season, lambda x: x['seasons'], list) or []
+
+            if current in [s.get('eTitle') for s in seasons]:
+                playlist_url = current_url
+
+        title = re.sub(
+            r'[-|]\s*(?:mtv\s*italia|programma|playlist)',
+            '', title, flags=re.IGNORECASE).strip()
+
+        return self.playlist_result(
+            self._get_entries(title, playlist_url),
+            video_id, title, description)

yt_dlp/extractor/mxplayer.py

@@ -6,98 +6,122 @@ from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     js_to_json,
+    qualities,
+    try_get,
     url_or_none,
     urljoin,
 )
 
 
-VALID_STREAMS = ('dash', )
-
-
 class MxplayerIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?mxplayer\.in/movie/(?P<slug>[a-z0-9]+(?:-[a-z0-9]+)*)'
+    _VALID_URL = r'https?://(?:www\.)?mxplayer\.in/(?:show|movie)/(?:(?P<display_id>[-/a-z0-9]+)-)?(?P<id>[a-z0-9]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'https://www.mxplayer.in/movie/watch-knock-knock-hindi-dubbed-movie-online-b9fa28df3bfb8758874735bbd7d2655a?watch=true',
         'info_dict': {
             'id': 'b9fa28df3bfb8758874735bbd7d2655a',
             'ext': 'mp4',
-            'title': 'Knock Knock Movie | Watch 2015 Knock Knock Full Movie Online- MX Player',
+            'title': 'Knock Knock (Hindi Dubbed)',
             'description': 'md5:b195ba93ff1987309cfa58e2839d2a5b'
         },
         'params': {
             'skip_download': True,
             'format': 'bestvideo'
         }
-    }
+    }, {
-
+        'url': 'https://www.mxplayer.in/show/watch-shaitaan/season-1/the-infamous-taxi-gang-of-meerut-online-45055d5bcff169ad48f2ad7552a83d6c',
-    def _get_best_stream_url(self, stream):
+        'info_dict': {
-        best_stream = list(filter(None, [v for k, v in stream.items()]))
+            'id': '45055d5bcff169ad48f2ad7552a83d6c',
-        return best_stream.pop(0) if len(best_stream) else None
+            'ext': 'm3u8',
+            'title': 'The infamous taxi gang of Meerut',
+            'description': 'md5:033a0a7e3fd147be4fb7e07a01a3dc28',
+            'season': 'Season 1',
+            'series': 'Shaitaan'
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        'url': 'https://www.mxplayer.in/show/watch-aashram/chapter-1/duh-swapna-online-d445579792b0135598ba1bc9088a84cb',
+        'info_dict': {
+            'id': 'd445579792b0135598ba1bc9088a84cb',
+            'ext': 'mp4',
+            'title': 'Duh Swapna',
+            'description': 'md5:35ff39c4bdac403c53be1e16a04192d8',
+            'season': 'Chapter 1',
+            'series': 'Aashram'
+        },
+        'expected_warnings': ['Unknown MIME type application/mp4 in DASH manifest'],
+        'params': {
+            'skip_download': True,
+            'format': 'bestvideo'
+        }
+    }]
 
     def _get_stream_urls(self, video_dict):
-        stream_dict = video_dict.get('stream', {'provider': {}})
+        stream_provider_dict = try_get(
-        stream_provider = stream_dict.get('provider')
+            video_dict,
+            lambda x: x['stream'][x['stream']['provider']])
+        if not stream_provider_dict:
+            raise ExtractorError('No stream provider found', expected=True)
 
-        if not stream_dict[stream_provider]:
+        for stream_name, stream in stream_provider_dict.items():
-            message = 'No stream provider found'
+            if stream_name in ('hls', 'dash', 'hlsUrl', 'dashUrl'):
-            raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+                stream_type = stream_name.replace('Url', '')
-
+                if isinstance(stream, dict):
-        streams = []
+                    for quality, stream_url in stream.items():
-        for stream_name, v in stream_dict[stream_provider].items():
+                        if stream_url:
-            if stream_name in VALID_STREAMS:
+                            yield stream_type, quality, stream_url
-                stream_url = self._get_best_stream_url(v)
+                else:
-                if stream_url is None:
+                    yield stream_type, 'base', stream
-                    continue
-                streams.append((stream_name, stream_url))
-        return streams
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        display_id, video_id = re.match(self._VALID_URL, url).groups()
-        video_slug = mobj.group('slug')
-
-        video_id = video_slug.split('-')[-1]
-
         webpage = self._download_webpage(url, video_id)
 
-        window_state_json = self._html_search_regex(
+        source = self._parse_json(
-            r'(?s)<script>window\.state\s*[:=]\s(\{.+\})\n(\w+).*(</script>).*',
+            js_to_json(self._html_search_regex(
-            webpage, 'WindowState')
+                r'(?s)<script>window\.state\s*[:=]\s(\{.+\})\n(\w+).*(</script>).*',
-
+                webpage, 'WindowState')),
-        source = self._parse_json(js_to_json(window_state_json), video_id)
+            video_id)
         if not source:
             raise ExtractorError('Cannot find source', expected=True)
 
         config_dict = source['config']
         video_dict = source['entities'][video_id]
-        stream_urls = self._get_stream_urls(video_dict)
 
-        title = self._og_search_title(webpage, fatal=True, default=video_dict['title'])
+        thumbnails = []
+        for i in video_dict.get('imageInfo') or []:
+            thumbnails.append({
+                'url': urljoin(config_dict['imageBaseUrl'], i['url']),
+                'width': i['width'],
+                'height': i['height'],
+            })
 
         formats = []
-        headers = {'Referer': url}
+        get_quality = qualities(['main', 'base', 'high'])
-        for stream_name, stream_url in stream_urls:
+        for stream_type, quality, stream_url in self._get_stream_urls(video_dict):
-            if stream_name == 'dash':
+            format_url = url_or_none(urljoin(config_dict['videoCdnBaseUrl'], stream_url))
-                format_url = url_or_none(urljoin(config_dict['videoCdnBaseUrl'], stream_url))
+            if not format_url:
-                if not format_url:
+                continue
-                    continue
+            if stream_type == 'dash':
-                formats.extend(self._extract_mpd_formats(
+                dash_formats = self._extract_mpd_formats(
-                    format_url, video_id, mpd_id='dash', headers=headers))
+                    format_url, video_id, mpd_id='dash-%s' % quality, headers={'Referer': url})
+                for frmt in dash_formats:
+                    frmt['quality'] = get_quality(quality)
+                formats.extend(dash_formats)
+            elif stream_type == 'hls':
+                formats.extend(self._extract_m3u8_formats(
+                    format_url, video_id, fatal=False,
+                    m3u8_id='hls-%s' % quality, quality=get_quality(quality)))
 
         self._sort_formats(formats)
-        info = {
+        return {
             'id': video_id,
-            'title': title,
+            'display_id': display_id.replace('/', '-'),
+            'title': video_dict['title'] or self._og_search_title(webpage),
+            'formats': formats,
             'description': video_dict.get('description'),
-            'formats': formats
+            'season': try_get(video_dict, lambda x: x['container']['title']),
+            'series': try_get(video_dict, lambda x: x['container']['container']['title']),
+            'thumbnails': thumbnails,
         }
-
-        if video_dict.get('imageInfo'):
-            info['thumbnails'] = list(map(lambda i: dict(i, **{
-                'url': urljoin(config_dict['imageBaseUrl'], i['url'])
-            }), video_dict['imageInfo']))
-
-        if video_dict.get('webUrl'):
-            last_part = video_dict['webUrl'].split("/")[-1]
-            info['display_id'] = last_part.replace(video_id, "").rstrip("-")
-
-        return info

yt_dlp/extractor/peertube.py

@@ -599,11 +599,13 @@ class PeerTubeIE(InfoExtractor):
         else:
             age_limit = None
 
+        webpage_url = 'https://%s/videos/watch/%s' % (host, video_id)
+
         return {
             'id': video_id,
             'title': title,
             'description': description,
-            'thumbnail': urljoin(url, video.get('thumbnailPath')),
+            'thumbnail': urljoin(webpage_url, video.get('thumbnailPath')),
             'timestamp': unified_timestamp(video.get('publishedAt')),
             'uploader': account_data('displayName', compat_str),
             'uploader_id': str_or_none(account_data('id', int)),
@@ -621,5 +623,6 @@ class PeerTubeIE(InfoExtractor):
             'tags': try_get(video, lambda x: x['tags'], list),
             'categories': categories,
             'formats': formats,
-            'subtitles': subtitles
+            'subtitles': subtitles,
+            'webpage_url': webpage_url,
         }

yt_dlp/extractor/pinterest.py

@@ -31,6 +31,7 @@ class PinterestBaseIE(InfoExtractor):
 
         title = (data.get('title') or data.get('grid_title') or video_id).strip()
 
+        urls = []
         formats = []
         duration = None
         if extract_formats:
@@ -38,8 +39,9 @@ class PinterestBaseIE(InfoExtractor):
                 if not isinstance(format_dict, dict):
                     continue
                 format_url = url_or_none(format_dict.get('url'))
-                if not format_url:
+                if not format_url or format_url in urls:
                     continue
+                urls.append(format_url)
                 duration = float_or_none(format_dict.get('duration'), scale=1000)
                 ext = determine_ext(format_url)
                 if 'hls' in format_id.lower() or ext == 'm3u8':

yt_dlp/extractor/plutotv.py

@@ -0,0 +1,164 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+import uuid
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_str,
+    compat_urlparse,
+)
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    int_or_none,
+    try_get,
+    url_or_none,
+)
+
+
+class PlutoTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?pluto\.tv/on-demand/(?P<video_type>movies|series)/(?P<slug>.*)/?$'
+    _INFO_URL = 'https://service-vod.clusters.pluto.tv/v3/vod/slugs/'
+    _INFO_QUERY_PARAMS = {
+        'appName': 'web',
+        'appVersion': 'na',
+        'clientID': compat_str(uuid.uuid1()),
+        'clientModelNumber': 'na',
+        'serverSideAds': 'false',
+        'deviceMake': 'unknown',
+        'deviceModel': 'web',
+        'deviceType': 'web',
+        'deviceVersion': 'unknown',
+        'sid': compat_str(uuid.uuid1()),
+    }
+    _TESTS = [
+        {
+            'url': 'https://pluto.tv/on-demand/series/i-love-money/season/2/episode/its-in-the-cards-2009-2-3',
+            'md5': 'ebcdd8ed89aaace9df37924f722fd9bd',
+            'info_dict': {
+                'id': '5de6c598e9379ae4912df0a8',
+                'ext': 'mp4',
+                'title': 'It\'s In The Cards',
+                'episode': 'It\'s In The Cards',
+                'description': 'The teams face off against each other in a 3-on-2 soccer showdown.  Strategy comes into play, though, as each team gets to select their opposing teams’ two defenders.',
+                'series': 'I Love Money',
+                'season_number': 2,
+                'episode_number': 3,
+                'duration': 3600,
+            }
+        },
+        {
+            'url': 'https://pluto.tv/on-demand/series/i-love-money/season/1/',
+            'playlist_count': 11,
+            'info_dict': {
+                'id': '5de6c582e9379ae4912dedbd',
+                'title': 'I Love Money - Season 1',
+            }
+        },
+        {
+            'url': 'https://pluto.tv/on-demand/series/i-love-money/',
+            'playlist_count': 26,
+            'info_dict': {
+                'id': '5de6c582e9379ae4912dedbd',
+                'title': 'I Love Money',
+            }
+        },
+        {
+            'url': 'https://pluto.tv/on-demand/movies/arrival-2015-1-1',
+            'md5': '3cead001d317a018bf856a896dee1762',
+            'info_dict': {
+                'id': '5e83ac701fa6a9001bb9df24',
+                'ext': 'mp4',
+                'title': 'Arrival',
+                'description': 'When mysterious spacecraft touch down across the globe, an elite team - led by expert translator Louise Banks (Academy Award® nominee Amy Adams) – races against time to decipher their intent.',
+                'duration': 9000,
+            }
+        },
+    ]
+
+    def _to_ad_free_formats(self, video_id, formats):
+        ad_free_formats = []
+        m3u8_urls = set()
+        for format in formats:
+            res = self._download_webpage(
+                format.get('url'), video_id, note='Downloading m3u8 playlist',
+                fatal=False)
+            if not res:
+                continue
+            first_segment_url = re.search(
+                r'^(https?://.*/)0\-(end|[0-9]+)/[^/]+\.ts$', res,
+                re.MULTILINE)
+            if not first_segment_url:
+                continue
+            m3u8_urls.add(
+                compat_urlparse.urljoin(first_segment_url.group(1), '0-end/master.m3u8'))
+
+        for m3u8_url in m3u8_urls:
+            ad_free_formats.extend(
+                self._extract_m3u8_formats(
+                    m3u8_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
+        self._sort_formats(ad_free_formats)
+        return ad_free_formats
+
+    def _get_video_info(self, video_json, slug, series_name=None):
+        video_id = video_json.get('_id', slug)
+        formats = []
+        for video_url in try_get(video_json, lambda x: x['stitched']['urls'], list) or []:
+            if video_url.get('type') != 'hls':
+                continue
+            url = url_or_none(video_url.get('url'))
+            formats.extend(
+                self._extract_m3u8_formats(
+                    url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
+        info = {
+            'id': video_id,
+            'formats': self._to_ad_free_formats(video_id, formats),
+            'title': video_json.get('name'),
+            'description': video_json.get('description'),
+            'duration': float_or_none(video_json.get('duration'), scale=1000),
+        }
+        if series_name:
+            info.update({
+                'series': series_name,
+                'episode': video_json.get('name'),
+                'season_number': int_or_none(video_json.get('season')),
+                'episode_number': int_or_none(video_json.get('number')),
+            })
+        return info
+
+    def _real_extract(self, url):
+        path = compat_urlparse.urlparse(url).path
+        path_components = path.split('/')
+        video_type = path_components[2]
+        info_slug = path_components[3]
+        video_json = self._download_json(self._INFO_URL + info_slug, info_slug,
+                                         query=self._INFO_QUERY_PARAMS)
+
+        if video_type == 'series':
+            series_name = video_json.get('name', info_slug)
+            season_number = int_or_none(try_get(path_components, lambda x: x[5]))
+            episode_slug = try_get(path_components, lambda x: x[7])
+
+            videos = []
+            for season in video_json['seasons']:
+                if season_number is not None and season_number != int_or_none(season.get('number')):
+                    continue
+                for episode in season['episodes']:
+                    if episode_slug is not None and episode_slug != episode.get('slug'):
+                        continue
+                    videos.append(self._get_video_info(episode, episode_slug, series_name))
+            if not videos:
+                raise ExtractorError('Failed to find any videos to extract')
+            if episode_slug is not None and len(videos) == 1:
+                return videos[0]
+            playlist_title = series_name
+            if season_number is not None:
+                playlist_title += ' - Season %d' % season_number
+            return self.playlist_result(videos,
+                                        playlist_id=video_json.get('_id', info_slug),
+                                        playlist_title=playlist_title)
+        return self._get_video_info(video_json, info_slug)

yt_dlp/extractor/pornhub.py

@@ -167,6 +167,7 @@ class PornHubIE(PornHubBaseIE):
         'params': {
             'skip_download': True,
         },
+        'skip': 'Video has been flagged for verification in accordance with our trust and safety policy',
     }, {
         # subtitles
         'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5af5fef7c2aa7',
@@ -265,7 +266,8 @@ class PornHubIE(PornHubBaseIE):
         webpage = dl_webpage('pc')
 
         error_msg = self._html_search_regex(
-            r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
+            (r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>',
+             r'(?s)<section[^>]+class=["\']noVideo["\'][^>]*>(?P<error>.+?)</section>'),
             webpage, 'error message', default=None, group='error')
         if error_msg:
             error_msg = re.sub(r'\s+', ' ', error_msg)
@@ -394,6 +396,21 @@ class PornHubIE(PornHubBaseIE):
 
         upload_date = None
         formats = []
+
+        def add_format(format_url, height=None):
+            tbr = None
+            mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', format_url)
+            if mobj:
+                if not height:
+                    height = int(mobj.group('height'))
+                tbr = int(mobj.group('tbr'))
+            formats.append({
+                'url': format_url,
+                'format_id': '%dp' % height if height else None,
+                'height': height,
+                'tbr': tbr,
+            })
+
         for video_url, height in video_urls:
             if not upload_date:
                 upload_date = self._search_regex(
@@ -410,18 +427,19 @@ class PornHubIE(PornHubBaseIE):
                     video_url, video_id, 'mp4', entry_protocol='m3u8_native',
                     m3u8_id='hls', fatal=False))
                 continue
-            tbr = None
+            if '/video/get_media' in video_url:
-            mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url)
+                medias = self._download_json(video_url, video_id, fatal=False)
-            if mobj:
+                if isinstance(medias, list):
-                if not height:
+                    for media in medias:
-                    height = int(mobj.group('height'))
+                        if not isinstance(media, dict):
-                tbr = int(mobj.group('tbr'))
+                            continue
-            formats.append({
+                        video_url = url_or_none(media.get('videoUrl'))
-                'url': video_url,
+                        if not video_url:
-                'format_id': '%dp' % height if height else None,
+                            continue
-                'height': height,
+                        height = int_or_none(media.get('quality'))
-                'tbr': tbr,
+                        add_format(video_url, height)
-            })
+                continue
+            add_format(video_url)
         self._sort_formats(formats)
 
         video_uploader = self._html_search_regex(

yt_dlp/extractor/rai.py

@@ -158,6 +158,10 @@ class RaiPlayIE(RaiBaseIE):
         # subtitles at 'subtitlesArray' key (see #27698)
         'url': 'https://www.raiplay.it/video/2020/12/Report---04-01-2021-2e90f1de-8eee-4de4-ac0e-78d21db5b600.html',
         'only_matching': True,
+    }, {
+        # DRM protected
+        'url': 'https://www.raiplay.it/video/2020/09/Lo-straordinario-mondo-di-Zoey-S1E1-Lo-straordinario-potere-di-Zoey-ed493918-1d32-44b7-8454-862e473d00ff.html',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -166,6 +170,14 @@ class RaiPlayIE(RaiBaseIE):
         media = self._download_json(
             base + '.json', video_id, 'Downloading video JSON')
 
+        if not self._downloader.params.get('allow_unplayable_formats'):
+            if try_get(
+                    media,
+                    (lambda x: x['rights_management']['rights']['drm'],
+                     lambda x: x['program_info']['rights_management']['rights']['drm']),
+                    dict):
+                raise ExtractorError('This video is DRM protected.', expected=True)
+
         title = media['name']
         video = media['video']
 

yt_dlp/extractor/rtve.py

@@ -2,8 +2,9 @@
 from __future__ import unicode_literals
 
 import base64
+import io
 import re
-import time
+import sys
 
 from .common import InfoExtractor
 from ..compat import (
@@ -14,56 +15,13 @@ from ..utils import (
     determine_ext,
     ExtractorError,
     float_or_none,
+    qualities,
     remove_end,
     remove_start,
-    sanitized_Request,
     std_headers,
 )
 
-
+_bytes_to_chr = (lambda x: x) if sys.version_info[0] == 2 else (lambda x: map(chr, x))
-def _decrypt_url(png):
-    encrypted_data = compat_b64decode(png)
-    text_index = encrypted_data.find(b'tEXt')
-    text_chunk = encrypted_data[text_index - 4:]
-    length = compat_struct_unpack('!I', text_chunk[:4])[0]
-    # Use bytearray to get integers when iterating in both python 2.x and 3.x
-    data = bytearray(text_chunk[8:8 + length])
-    data = [chr(b) for b in data if b != 0]
-    hash_index = data.index('#')
-    alphabet_data = data[:hash_index]
-    url_data = data[hash_index + 1:]
-    if url_data[0] == 'H' and url_data[3] == '%':
-        # remove useless HQ%% at the start
-        url_data = url_data[4:]
-
-    alphabet = []
-    e = 0
-    d = 0
-    for l in alphabet_data:
-        if d == 0:
-            alphabet.append(l)
-            d = e = (e + 1) % 4
-        else:
-            d -= 1
-    url = ''
-    f = 0
-    e = 3
-    b = 1
-    for letter in url_data:
-        if f == 0:
-            l = int(letter) * 10
-            f = 1
-        else:
-            if e == 0:
-                l += int(letter)
-                url += alphabet[l]
-                e = (b + 3) % 4
-                f = 0
-                b += 1
-            else:
-                e -= 1
-
-    return url
 
 
 class RTVEALaCartaIE(InfoExtractor):
@@ -79,28 +37,31 @@ class RTVEALaCartaIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Balonmano - Swiss Cup masculina. Final: España-Suecia',
             'duration': 5024.566,
+            'series': 'Balonmano',
         },
+        'expected_warnings': ['Failed to download MPD manifest', 'Failed to download m3u8 information'],
     }, {
         'note': 'Live stream',
         'url': 'http://www.rtve.es/alacarta/videos/television/24h-live/1694255/',
         'info_dict': {
             'id': '1694255',
-            'ext': 'flv',
+            'ext': 'mp4',
-            'title': 'TODO',
+            'title': 're:^24H LIVE [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': 'live stream',
         },
-        'skip': 'The f4m manifest can\'t be used yet',
     }, {
         'url': 'http://www.rtve.es/alacarta/videos/servir-y-proteger/servir-proteger-capitulo-104/4236788/',
-        'md5': 'e55e162379ad587e9640eda4f7353c0f',
+        'md5': 'd850f3c8731ea53952ebab489cf81cbf',
         'info_dict': {
             'id': '4236788',
             'ext': 'mp4',
-            'title': 'Servir y proteger - Capítulo 104 ',
+            'title': 'Servir y proteger - Capítulo 104',
             'duration': 3222.0,
         },
-        'params': {
+        'expected_warnings': ['Failed to download MPD manifest', 'Failed to download m3u8 information'],
-            'skip_download': True,  # requires ffmpeg
-        },
     }, {
         'url': 'http://www.rtve.es/m/alacarta/videos/cuentame-como-paso/cuentame-como-paso-t16-ultimo-minuto-nuestra-vida-capitulo-276/2969138/?media=tve',
         'only_matching': True,
@@ -111,58 +72,102 @@ class RTVEALaCartaIE(InfoExtractor):
 
     def _real_initialize(self):
         user_agent_b64 = base64.b64encode(std_headers['User-Agent'].encode('utf-8')).decode('utf-8')
-        manager_info = self._download_json(
+        self._manager = self._download_json(
             'http://www.rtve.es/odin/loki/' + user_agent_b64,
-            None, 'Fetching manager info')
+            None, 'Fetching manager info')['manager']
-        self._manager = manager_info['manager']
+
+    @staticmethod
+    def _decrypt_url(png):
+        encrypted_data = io.BytesIO(compat_b64decode(png)[8:])
+        while True:
+            length = compat_struct_unpack('!I', encrypted_data.read(4))[0]
+            chunk_type = encrypted_data.read(4)
+            if chunk_type == b'IEND':
+                break
+            data = encrypted_data.read(length)
+            if chunk_type == b'tEXt':
+                alphabet_data, text = data.split(b'\0')
+                quality, url_data = text.split(b'%%')
+                alphabet = []
+                e = 0
+                d = 0
+                for l in _bytes_to_chr(alphabet_data):
+                    if d == 0:
+                        alphabet.append(l)
+                        d = e = (e + 1) % 4
+                    else:
+                        d -= 1
+                url = ''
+                f = 0
+                e = 3
+                b = 1
+                for letter in _bytes_to_chr(url_data):
+                    if f == 0:
+                        l = int(letter) * 10
+                        f = 1
+                    else:
+                        if e == 0:
+                            l += int(letter)
+                            url += alphabet[l]
+                            e = (b + 3) % 4
+                            f = 0
+                            b += 1
+                        else:
+                            e -= 1
+
+                yield quality.decode(), url
+            encrypted_data.read(4)  # CRC
+
+    def _extract_png_formats(self, video_id):
+        png = self._download_webpage(
+            'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id),
+            video_id, 'Downloading url information', query={'q': 'v2'})
+        q = qualities(['Media', 'Alta', 'HQ', 'HD_READY', 'HD_FULL'])
+        formats = []
+        for quality, video_url in self._decrypt_url(png):
+            ext = determine_ext(video_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif ext == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    video_url, video_id, 'dash', fatal=False))
+            else:
+                formats.append({
+                    'format_id': quality,
+                    'quality': q(quality),
+                    'url': video_url,
+                })
+        self._sort_formats(formats)
+        return formats
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
         info = self._download_json(
             'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
             video_id)['page']['items'][0]
         if info['state'] == 'DESPU':
             raise ExtractorError('The video is no longer available', expected=True)
-        title = info['title']
+        title = info['title'].strip()
-        png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id)
+        formats = self._extract_png_formats(video_id)
-        png_request = sanitized_Request(png_url)
-        png_request.add_header('Referer', url)
-        png = self._download_webpage(png_request, video_id, 'Downloading url information')
-        video_url = _decrypt_url(png)
-        ext = determine_ext(video_url)
-
-        formats = []
-        if not video_url.endswith('.f4m') and ext != 'm3u8':
-            if '?' not in video_url:
-                video_url = video_url.replace('resources/', 'auth/resources/')
-            video_url = video_url.replace('.net.rtve', '.multimedia.cdn.rtve')
-
-        if ext == 'm3u8':
-            formats.extend(self._extract_m3u8_formats(
-                video_url, video_id, ext='mp4', entry_protocol='m3u8_native',
-                m3u8_id='hls', fatal=False))
-        elif ext == 'f4m':
-            formats.extend(self._extract_f4m_formats(
-                video_url, video_id, f4m_id='hds', fatal=False))
-        else:
-            formats.append({
-                'url': video_url,
-            })
-        self._sort_formats(formats)
 
         subtitles = None
-        if info.get('sbtFile') is not None:
+        sbt_file = info.get('sbtFile')
-            subtitles = self.extract_subtitles(video_id, info['sbtFile'])
+        if sbt_file:
+            subtitles = self.extract_subtitles(video_id, sbt_file)
+
+        is_live = info.get('live') is True
 
         return {
             'id': video_id,
-            'title': title,
+            'title': self._live_title(title) if is_live else title,
             'formats': formats,
             'thumbnail': info.get('image'),
-            'page_url': url,
             'subtitles': subtitles,
-            'duration': float_or_none(info.get('duration'), scale=1000),
+            'duration': float_or_none(info.get('duration'), 1000),
+            'is_live': is_live,
+            'series': info.get('programTitle'),
         }
 
     def _get_subtitles(self, video_id, sub_file):
@@ -174,48 +179,26 @@ class RTVEALaCartaIE(InfoExtractor):
             for s in subs)
 
 
-class RTVEInfantilIE(InfoExtractor):
+class RTVEInfantilIE(RTVEALaCartaIE):
     IE_NAME = 'rtve.es:infantil'
     IE_DESC = 'RTVE infantil'
-    _VALID_URL = r'https?://(?:www\.)?rtve\.es/infantil/serie/(?P<show>[^/]*)/video/(?P<short_title>[^/]*)/(?P<id>[0-9]+)/'
+    _VALID_URL = r'https?://(?:www\.)?rtve\.es/infantil/serie/[^/]+/video/[^/]+/(?P<id>[0-9]+)/'
 
     _TESTS = [{
         'url': 'http://www.rtve.es/infantil/serie/cleo/video/maneras-vivir/3040283/',
-        'md5': '915319587b33720b8e0357caaa6617e6',
+        'md5': '5747454717aedf9f9fdf212d1bcfc48d',
         'info_dict': {
             'id': '3040283',
             'ext': 'mp4',
             'title': 'Maneras de vivir',
-            'thumbnail': 'http://www.rtve.es/resources/jpg/6/5/1426182947956.JPG',
+            'thumbnail': r're:https?://.+/1426182947956\.JPG',
             'duration': 357.958,
         },
+        'expected_warnings': ['Failed to download MPD manifest', 'Failed to download m3u8 information'],
     }]
 
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        info = self._download_json(
-            'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
-            video_id)['page']['items'][0]
 
-        webpage = self._download_webpage(url, video_id)
+class RTVELiveIE(RTVEALaCartaIE):
-        vidplayer_id = self._search_regex(
-            r' id="vidplayer([0-9]+)"', webpage, 'internal video ID')
-
-        png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % vidplayer_id
-        png = self._download_webpage(png_url, video_id, 'Downloading url information')
-        video_url = _decrypt_url(png)
-
-        return {
-            'id': video_id,
-            'ext': 'mp4',
-            'title': info['title'],
-            'url': video_url,
-            'thumbnail': info.get('image'),
-            'duration': float_or_none(info.get('duration'), scale=1000),
-        }
-
-
-class RTVELiveIE(InfoExtractor):
     IE_NAME = 'rtve.es:live'
     IE_DESC = 'RTVE.es live streams'
     _VALID_URL = r'https?://(?:www\.)?rtve\.es/directo/(?P<id>[a-zA-Z0-9-]+)'
@@ -225,7 +208,7 @@ class RTVELiveIE(InfoExtractor):
         'info_dict': {
             'id': 'la-1',
             'ext': 'mp4',
-            'title': 're:^La 1 [0-9]{4}-[0-9]{2}-[0-9]{2}Z[0-9]{6}$',
+            'title': 're:^La 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
         },
         'params': {
             'skip_download': 'live stream',
@@ -234,29 +217,22 @@ class RTVELiveIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        start_time = time.gmtime()
         video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id)
         title = remove_end(self._og_search_title(webpage), ' en directo en RTVE.es')
         title = remove_start(title, 'Estoy viendo ')
-        title += ' ' + time.strftime('%Y-%m-%dZ%H%M%S', start_time)
 
         vidplayer_id = self._search_regex(
             (r'playerId=player([0-9]+)',
              r'class=["\'].*?\blive_mod\b.*?["\'][^>]+data-assetid=["\'](\d+)',
              r'data-id=["\'](\d+)'),
             webpage, 'internal video ID')
-        png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/amonet/videos/%s.png' % vidplayer_id
-        png = self._download_webpage(png_url, video_id, 'Downloading url information')
-        m3u8_url = _decrypt_url(png)
-        formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
-        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': title,
+            'title': self._live_title(title),
-            'formats': formats,
+            'formats': self._extract_png_formats(vidplayer_id),
             'is_live': True,
         }
 

yt_dlp/extractor/shahid.py

@@ -51,13 +51,16 @@ class ShahidIE(ShahidBaseIE):
     _NETRC_MACHINE = 'shahid'
     _VALID_URL = r'https?://shahid\.mbc\.net/ar/(?:serie|show|movie)s/[^/]+/(?P<type>episode|clip|movie)-(?P<id>\d+)'
     _TESTS = [{
-        'url': 'https://shahid.mbc.net/ar/shows/%D9%85%D8%AC%D9%84%D8%B3-%D8%A7%D9%84%D8%B4%D8%A8%D8%A7%D8%A8-%D8%A7%D9%84%D9%85%D9%88%D8%B3%D9%85-1-%D9%83%D9%84%D9%8A%D8%A8-1/clip-275286',
+        'url': 'https://shahid.mbc.net/ar/shows/%D9%85%D8%AA%D8%AD%D9%81-%D8%A7%D9%84%D8%AF%D8%AD%D9%8A%D8%AD-%D8%A7%D9%84%D9%85%D9%88%D8%B3%D9%85-1-%D9%83%D9%84%D9%8A%D8%A8-1/clip-816924',
         'info_dict': {
-            'id': '275286',
+            'id': '816924',
             'ext': 'mp4',
-            'title': 'مجلس الشباب الموسم 1 كليب 1',
+            'title': 'متحف الدحيح الموسم 1 كليب 1',
-            'timestamp': 1506988800,
+            'timestamp': 1602806400,
-            'upload_date': '20171003',
+            'upload_date': '20201016',
+            'description': 'برومو',
+            'duration': 22,
+            'categories': ['كوميديا'],
         },
         'params': {
             # m3u8 download
@@ -109,12 +112,15 @@ class ShahidIE(ShahidBaseIE):
             page_type = 'episode'
 
         playout = self._call_api(
-            'playout/url/' + video_id, video_id)['playout']
+            'playout/new/url/' + video_id, video_id)['playout']
 
         if not self._downloader.params.get('allow_unplayable_formats') and playout.get('drm'):
             raise ExtractorError('This video is DRM protected.', expected=True)
 
-        formats = self._extract_m3u8_formats(playout['url'], video_id, 'mp4')
+        formats = self._extract_m3u8_formats(re.sub(
+            # https://docs.aws.amazon.com/mediapackage/latest/ug/manifest-filtering.html
+            r'aws\.manifestfilter=[\w:;,-]+&?',
+            '', playout['url']), video_id, 'mp4')
         self._sort_formats(formats)
 
         # video = self._call_api(

yt_dlp/extractor/southpark.py

@@ -6,9 +6,9 @@ from .mtv import MTVServicesInfoExtractor
 
 class SouthParkIE(MTVServicesInfoExtractor):
     IE_NAME = 'southpark.cc.com'
-    _VALID_URL = r'https?://(?:www\.)?(?P<url>southpark\.cc\.com/(?:clips|(?:full-)?episodes|collections)/(?P<id>.+?)(\?|#|$))'
+    _VALID_URL = r'https?://(?:www\.)?(?P<url>southpark(?:\.cc|studios)\.com/(?:clips|(?:full-)?episodes|collections)/(?P<id>.+?)(\?|#|$))'
 
-    _FEED_URL = 'http://www.southparkstudios.com/feeds/video-player/mrss'
+    _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
 
     _TESTS = [{
         'url': 'http://southpark.cc.com/clips/104437/bat-daded#tab=featured',
@@ -23,8 +23,20 @@ class SouthParkIE(MTVServicesInfoExtractor):
     }, {
         'url': 'http://southpark.cc.com/collections/7758/fan-favorites/1',
         'only_matching': True,
+    }, {
+        'url': 'https://www.southparkstudios.com/episodes/h4o269/south-park-stunning-and-brave-season-19-ep-1',
+        'only_matching': True,
     }]
 
+    def _get_feed_query(self, uri):
+        return {
+            'accountOverride': 'intl.mtvi.com',
+            'arcEp': 'shared.southpark.global',
+            'ep': '90877963',
+            'imageEp': 'shared.southpark.global',
+            'mgid': uri,
+        }
+
 
 class SouthParkEsIE(SouthParkIE):
     IE_NAME = 'southpark.cc.com:español'

yt_dlp/extractor/sportdeutschland.py

@@ -1,82 +1,105 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..compat import (
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+)
 from ..utils import (
+    clean_html,
+    float_or_none,
+    int_or_none,
     parse_iso8601,
-    sanitized_Request,
+    strip_or_none,
+    try_get,
 )
 
 
 class SportDeutschlandIE(InfoExtractor):
-    _VALID_URL = r'https?://sportdeutschland\.tv/(?P<sport>[^/?#]+)/(?P<id>[^?#/]+)(?:$|[?#])'
+    _VALID_URL = r'https?://sportdeutschland\.tv/(?P<id>(?:[^/]+/)?[^?#/&]+)'
     _TESTS = [{
         'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0',
         'info_dict': {
-            'id': 're-live-deutsche-meisterschaften-2020-halbfinals',
+            'id': '5318cac0275701382770543d7edaf0a0',
             'ext': 'mp4',
-            'title': 're:Re-live: Deutsche Meisterschaften 2020.*Halbfinals',
+            'title': 'Re-live: Deutsche Meisterschaften 2020 - Halbfinals - Teil 1',
-            'categories': ['Badminton-Deutschland'],
+            'duration': 16106.36,
-            'view_count': int,
-            'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
-            'timestamp': int,
-            'upload_date': '20200201',
-            'description': 're:.*',  # meaningless description for THIS video
         },
+        'params': {
+            'noplaylist': True,
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0',
+        'info_dict': {
+            'id': 'c6e2fdd01f63013854c47054d2ab776f',
+            'title': 'Re-live: Deutsche Meisterschaften 2020 - Halbfinals',
+            'description': 'md5:5263ff4c31c04bb780c9f91130b48530',
+            'duration': 31397,
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'https://sportdeutschland.tv/freeride-world-tour-2021-fieberbrunn-oesterreich',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        display_id = self._match_id(url)
-        video_id = mobj.group('id')
+        data = self._download_json(
-        sport_id = mobj.group('sport')
+            'https://backend.sportdeutschland.tv/api/permalinks/' + display_id,
-
+            display_id, query={'access_token': 'true'})
-        api_url = 'https://proxy.vidibusdynamic.net/ssl/backend.sportdeutschland.tv/api/permalinks/%s/%s?access_token=true' % (
-            sport_id, video_id)
-        req = sanitized_Request(api_url, headers={
-            'Accept': 'application/vnd.vidibus.v2.html+json',
-            'Referer': url,
-        })
-        data = self._download_json(req, video_id)
-
         asset = data['asset']
-        categories = [data['section']['title']]
+        title = (asset.get('title') or asset['label']).strip()
-
+        asset_id = asset.get('id') or asset.get('uuid')
-        formats = []
+        info = {
-        smil_url = asset['video']
+            'id': asset_id,
-        if '.smil' in smil_url:
+            'title': title,
-            m3u8_url = smil_url.replace('.smil', '.m3u8')
+            'description': clean_html(asset.get('body') or asset.get('description')) or asset.get('teaser'),
-            formats.extend(
+            'duration': int_or_none(asset.get('seconds')),
-                self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4'))
-
-            smil_doc = self._download_xml(
-                smil_url, video_id, note='Downloading SMIL metadata')
-            base_url_el = smil_doc.find('./head/meta')
-            if base_url_el:
-                base_url = base_url_el.attrib['base']
-            formats.extend([{
-                'format_id': 'rmtp',
-                'url': base_url if base_url_el else n.attrib['src'],
-                'play_path': n.attrib['src'],
-                'ext': 'flv',
-                'preference': -100,
-                'format_note': 'Seems to fail at example stream',
-            } for n in smil_doc.findall('./body/video')])
-        else:
-            formats.append({'url': smil_url})
-
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'formats': formats,
-            'title': asset['title'],
-            'thumbnail': asset.get('image'),
-            'description': asset.get('teaser'),
-            'duration': asset.get('duration'),
-            'categories': categories,
-            'view_count': asset.get('views'),
-            'rtmp_live': asset.get('live'),
-            'timestamp': parse_iso8601(asset.get('date')),
         }
+        videos = asset.get('videos') or []
+        if len(videos) > 1:
+            playlist_id = compat_parse_qs(compat_urllib_parse_urlparse(url).query).get('playlistId', [None])[0]
+            if playlist_id:
+                if self._downloader.params.get('noplaylist'):
+                    videos = [videos[int(playlist_id)]]
+                    self.to_screen('Downloading just a single video because of --no-playlist')
+                else:
+                    self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % asset_id)
+
+            def entries():
+                for i, video in enumerate(videos, 1):
+                    video_id = video.get('uuid')
+                    video_url = video.get('url')
+                    if not (video_id and video_url):
+                        continue
+                    formats = self._extract_m3u8_formats(
+                        video_url.replace('.smil', '.m3u8'), video_id, 'mp4', fatal=False)
+                    if not formats:
+                        continue
+                    yield {
+                        'id': video_id,
+                        'formats': formats,
+                        'title': title + ' - ' + (video.get('label') or 'Teil %d' % i),
+                        'duration': float_or_none(video.get('duration')),
+                    }
+            info.update({
+                '_type': 'multi_video',
+                'entries': entries(),
+            })
+        else:
+            formats = self._extract_m3u8_formats(
+                videos[0]['url'].replace('.smil', '.m3u8'), asset_id, 'mp4')
+            section_title = strip_or_none(try_get(data, lambda x: x['section']['title']))
+            info.update({
+                'formats': formats,
+                'display_id': asset.get('permalink'),
+                'thumbnail': try_get(asset, lambda x: x['images'][0]),
+                'categories': [section_title] if section_title else None,
+                'view_count': int_or_none(asset.get('views')),
+                'is_live': asset.get('is_live') is True,
+                'timestamp': parse_iso8601(asset.get('date') or asset.get('published_at')),
+            })
+        return info

yt_dlp/extractor/trovo.py

@@ -14,6 +14,7 @@ from ..utils import (
 
 class TrovoBaseIE(InfoExtractor):
     _VALID_URL_BASE = r'https?://(?:www\.)?trovo\.live/'
+    _HEADERS = {'Origin': 'https://trovo.live'}
 
     def _extract_streamer_info(self, data):
         streamer_info = data.get('streamerInfo') or {}
@@ -68,6 +69,7 @@ class TrovoIE(TrovoBaseIE):
                 'format_id': format_id,
                 'height': int_or_none(format_id[:-1]) if format_id else None,
                 'url': play_url,
+                'http_headers': self._HEADERS,
             })
         self._sort_formats(formats)
 
@@ -153,6 +155,7 @@ class TrovoVodIE(TrovoBaseIE):
                 'protocol': 'm3u8_native',
                 'tbr': int_or_none(play_info.get('bitrate')),
                 'url': play_url,
+                'http_headers': self._HEADERS,
             })
         self._sort_formats(formats)
 

yt_dlp/extractor/tver.py

@@ -9,6 +9,7 @@ from ..utils import (
     int_or_none,
     remove_start,
     smuggle_url,
+    strip_or_none,
     try_get,
 )
 
@@ -25,6 +26,10 @@ class TVerIE(InfoExtractor):
     }, {
         'url': 'https://tver.jp/episode/79622438',
         'only_matching': True,
+    }, {
+        # subtitle = ' '
+        'url': 'https://tver.jp/corner/f0068870',
+        'only_matching': True,
     }]
     _TOKEN = None
     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s'
@@ -47,8 +52,12 @@ class TVerIE(InfoExtractor):
         }
 
         if service == 'cx':
+            title = main['title']
+            subtitle = strip_or_none(main.get('subtitle'))
+            if subtitle:
+                title += ' - ' + subtitle
             info.update({
-                'title': main.get('subtitle') or main['title'],
+                'title': title,
                 'url': 'https://i.fod.fujitv.co.jp/plus7/web/%s/%s.html' % (p_id[:4], p_id),
                 'ie_key': 'FujiTVFODPlus7',
             })

yt_dlp/extractor/vimeo.py

@@ -498,6 +498,24 @@ class VimeoIE(VimeoBaseInfoExtractor):
             'url': 'https://vimeo.com/album/2632481/video/79010983',
             'only_matching': True,
         },
+        {
+            'url': 'https://vimeo.com/showcase/3253534/video/119195465',
+            'note': 'A video in a password protected album (showcase)',
+            'info_dict': {
+                'id': '119195465',
+                'ext': 'mp4',
+                'title': 'youtube-dl test video \'ä"BaW_jenozKc',
+                'uploader': 'Philipp Hagemeister',
+                'uploader_id': 'user20132939',
+                'description': 'md5:fa7b6c6d8db0bdc353893df2f111855b',
+                'upload_date': '20150209',
+                'timestamp': 1423518307,
+            },
+            'params': {
+                'format': 'best[protocol=https]',
+                'videopassword': 'youtube-dl',
+            },
+        },
         {
             # source file returns 403: Forbidden
             'url': 'https://vimeo.com/7809605',
@@ -564,6 +582,44 @@ class VimeoIE(VimeoBaseInfoExtractor):
     def _real_initialize(self):
         self._login()
 
+    def _try_album_password(self, url):
+        album_id = self._search_regex(
+            r'vimeo\.com/(?:album|showcase)/([^/]+)', url, 'album id', default=None)
+        if not album_id:
+            return
+        viewer = self._download_json(
+            'https://vimeo.com/_rv/viewer', album_id, fatal=False)
+        if not viewer:
+            webpage = self._download_webpage(url, album_id)
+            viewer = self._parse_json(self._search_regex(
+                r'bootstrap_data\s*=\s*({.+?})</script>',
+                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'})
+        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:
+                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',
+                    })
+            except ExtractorError as e:
+                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                    raise ExtractorError('Wrong password', expected=True)
+                raise
+
     def _real_extract(self, url):
         url, data = unsmuggle_url(url, {})
         headers = std_headers.copy()
@@ -591,6 +647,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
             url = 'https://vimeo.com/' + video_id
 
+        self._try_album_password(url)
         try:
             # Retrieve video webpage to extract further information
             webpage, urlh = self._download_webpage_handle(

yt_dlp/extractor/voxmedia.py

@@ -7,6 +7,8 @@ from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     ExtractorError,
     int_or_none,
+    try_get,
+    unified_timestamp,
 )
 
 
@@ -19,14 +21,17 @@ class VoxMediaVolumeIE(OnceIE):
 
         setup = self._parse_json(self._search_regex(
             r'setup\s*=\s*({.+});', webpage, 'setup'), video_id)
-        video_data = setup.get('video') or {}
+        player_setup = setup.get('player_setup') or setup
+        video_data = player_setup.get('video') or {}
+        formatted_metadata = video_data.get('formatted_metadata') or {}
         info = {
             'id': video_id,
-            'title': video_data.get('title_short'),
+            'title': player_setup.get('title') or video_data.get('title_short'),
             'description': video_data.get('description_long') or video_data.get('description_short'),
-            'thumbnail': video_data.get('brightcove_thumbnail')
+            'thumbnail': formatted_metadata.get('thumbnail') or video_data.get('brightcove_thumbnail'),
+            'timestamp': unified_timestamp(formatted_metadata.get('video_publish_date')),
         }
-        asset = setup.get('asset') or setup.get('params') or {}
+        asset = try_get(setup, lambda x: x['embed_assets']['chorus'], dict) or {}
 
         formats = []
         hls_url = asset.get('hls_url')
@@ -47,6 +52,7 @@ class VoxMediaVolumeIE(OnceIE):
         if formats:
             self._sort_formats(formats)
             info['formats'] = formats
+            info['duration'] = int_or_none(asset.get('duration'))
             return info
 
         for provider_video_type in ('ooyala', 'youtube', 'brightcove'):
@@ -84,7 +90,7 @@ class VoxMediaIE(InfoExtractor):
     }, {
         # Volume embed, Youtube
         'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',
-        'md5': '4c8f4a0937752b437c3ebc0ed24802b5',
+        'md5': 'fd19aa0cf3a0eea515d4fd5c8c0e9d68',
         'info_dict': {
             'id': 'Gy8Md3Eky38',
             'ext': 'mp4',
@@ -93,6 +99,7 @@ class VoxMediaIE(InfoExtractor):
             'uploader_id': 'TheVerge',
             'upload_date': '20141021',
             'uploader': 'The Verge',
+            'timestamp': 1413907200,
         },
         'add_ie': ['Youtube'],
         'skip': 'similar to the previous test',
@@ -100,13 +107,13 @@ class VoxMediaIE(InfoExtractor):
         # Volume embed, Youtube
         'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',
         'info_dict': {
-            'id': 'YCjDnX-Xzhg',
+            'id': '22986359b',
             'ext': 'mp4',
             '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',
+            'timestamp': 1442332800,
+            'duration': 285,
         },
         'add_ie': ['Youtube'],
         'skip': 'similar to the previous test',
@@ -160,6 +167,9 @@ class VoxMediaIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Post-Post-PC CEO: The Full Code Conference Video of Microsoft\'s Satya Nadella',
             'description': 'The longtime veteran was chosen earlier this year as the software giant\'s third leader in its history.',
+            'timestamp': 1402938000,
+            'upload_date': '20140616',
+            'duration': 4114,
         },
         'add_ie': ['VoxMediaVolume'],
     }]

yt_dlp/extractor/wimtv.py

@@ -0,0 +1,163 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    parse_duration,
+    urlencode_postdata,
+    ExtractorError,
+)
+
+
+class WimTVIE(InfoExtractor):
+    _player = None
+    _UUID_RE = r'[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
+    _VALID_URL = r'''(?x)
+        https?://platform.wim.tv/
+        (?:
+            (?:embed/)?\?
+            |\#/webtv/.+?/
+        )
+        (?P<type>vod|live|cast)[=/]
+        (?P<id>%s).*?''' % _UUID_RE
+    _TESTS = [{
+        # vod stream
+        'url': 'https://platform.wim.tv/embed/?vod=db29fb32-bade-47b6-a3a6-cb69fe80267a',
+        'md5': 'db29fb32-bade-47b6-a3a6-cb69fe80267a',
+        'info_dict': {
+            'id': 'db29fb32-bade-47b6-a3a6-cb69fe80267a',
+            'ext': 'mp4',
+            'title': 'AMA SUPERCROSS 2020 - R2 ST. LOUIS',
+            'duration': 6481,
+            'thumbnail': r're:https?://.+?/thumbnail/.+?/720$'
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # live stream
+        'url': 'https://platform.wim.tv/embed/?live=28e22c22-49db-40f3-8c37-8cbb0ff44556&autostart=true',
+        'info_dict': {
+            'id': '28e22c22-49db-40f3-8c37-8cbb0ff44556',
+            'ext': 'mp4',
+            'title': 'Streaming MSmotorTV',
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://platform.wim.tv/#/webtv/automotornews/vod/422492b6-539e-474d-9c6b-68c9d5893365',
+        'only_matching': True,
+    }, {
+        'url': 'https://platform.wim.tv/#/webtv/renzoarborechannel/cast/f47e0d15-5b45-455e-bf0d-dba8ffa96365',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return [
+            mobj.group('url')
+            for mobj in re.finditer(
+                r'<iframe[^>]+src=["\'](?P<url>%s)' % WimTVIE._VALID_URL,
+                webpage)]
+
+    def _real_initialize(self):
+        if not self._player:
+            self._get_player_data()
+
+    def _get_player_data(self):
+        msg_id = 'Player data'
+        self._player = {}
+
+        datas = [{
+            'url': 'https://platform.wim.tv/common/libs/player/wimtv/wim-rest.js',
+            'vars': [{
+                'regex': r'appAuth = "(.+?)"',
+                'variable': 'app_auth',
+            }]
+        }, {
+            'url': 'https://platform.wim.tv/common/config/endpointconfig.js',
+            'vars': [{
+                'regex': r'PRODUCTION_HOSTNAME_THUMB = "(.+?)"',
+                'variable': 'thumb_server',
+            }, {
+                'regex': r'PRODUCTION_HOSTNAME_THUMB\s*\+\s*"(.+?)"',
+                'variable': 'thumb_server_path',
+            }]
+        }]
+
+        for data in datas:
+            temp = self._download_webpage(data['url'], msg_id)
+            for var in data['vars']:
+                val = self._search_regex(var['regex'], temp, msg_id)
+                if not val:
+                    raise ExtractorError('%s not found' % var['variable'])
+                self._player[var['variable']] = val
+
+    def _generate_token(self):
+        json = self._download_json(
+            'https://platform.wim.tv/wimtv-server/oauth/token', 'Token generation',
+            headers={'Authorization': 'Basic %s' % self._player['app_auth']},
+            data=urlencode_postdata({'grant_type': 'client_credentials'}))
+        token = json.get('access_token')
+        if not token:
+            raise ExtractorError('access token not generated')
+        return token
+
+    def _generate_thumbnail(self, thumb_id, width='720'):
+        if not thumb_id or not self._player.get('thumb_server'):
+            return None
+        if not self._player.get('thumb_server_path'):
+            self._player['thumb_server_path'] = ''
+        return '%s%s/asset/thumbnail/%s/%s' % (
+            self._player['thumb_server'],
+            self._player['thumb_server_path'],
+            thumb_id, width)
+
+    def _real_extract(self, url):
+        urlc = re.match(self._VALID_URL, url).groupdict()
+        video_id = urlc['id']
+        stream_type = is_live = None
+        if urlc['type'] in {'live', 'cast'}:
+            stream_type = urlc['type'] + '/channel'
+            is_live = True
+        else:
+            stream_type = 'vod'
+            is_live = False
+        token = self._generate_token()
+        json = self._download_json(
+            'https://platform.wim.tv/wimtv-server/api/public/%s/%s/play' % (
+                stream_type, video_id), video_id,
+            headers={'Authorization': 'Bearer %s' % token,
+                     'Content-Type': 'application/json'},
+            data=bytes('{}', 'utf-8'))
+
+        formats = []
+        for src in json.get('srcs') or []:
+            if src.get('mimeType') == 'application/x-mpegurl':
+                formats.extend(
+                    self._extract_m3u8_formats(
+                        src.get('uniqueStreamer'), video_id, 'mp4'))
+            if src.get('mimeType') == 'video/flash':
+                formats.append({
+                    'format_id': 'rtmp',
+                    'url': src.get('uniqueStreamer'),
+                    'ext': determine_ext(src.get('uniqueStreamer'), 'flv'),
+                    'rtmp_live': is_live,
+                })
+        json = json.get('resource')
+        thumb = self._generate_thumbnail(json.get('thumbnailId'))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': json.get('title') or json.get('name'),
+            'duration': parse_duration(json.get('duration')),
+            'formats': formats,
+            'thumbnail': thumb,
+            'is_live': is_live,
+        }

yt_dlp/extractor/youtube.py

@@ -26,6 +26,7 @@ from ..compat import (
 from ..jsinterp import JSInterpreter
 from ..utils import (
     clean_html,
+    dict_get,
     ExtractorError,
     format_field,
     float_or_none,
@@ -59,9 +60,9 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
     _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
 
     _RESERVED_NAMES = (
-        r'embed|e|watch_popup|channel|c|user|playlist|watch|w|v|movies|results|shared|hashtag|'
+        r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|'
-        r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout|'
+        r'movies|results|shared|hashtag|trending|feed|feeds|'
-        r'feed/(?:watch_later|history|subscriptions|library|trending|recommended)')
+        r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
 
     _NETRC_MACHINE = 'youtube'
     # If True it will raise an error if no login info is provided
@@ -271,15 +272,21 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         if not self._login():
             return
 
+    _YT_WEB_CLIENT_VERSION = '2.20210301.08.00'
     _DEFAULT_API_DATA = {
         'context': {
             'client': {
                 'clientName': 'WEB',
-                'clientVersion': '2.20210301.08.00',
+                'clientVersion': _YT_WEB_CLIENT_VERSION,
             }
         },
     }
 
+    _DEFAULT_BASIC_API_HEADERS = {
+        'X-YouTube-Client-Name': '1',
+        'X-YouTube-Client-Version': _YT_WEB_CLIENT_VERSION
+    }
+
     _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
     _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
     _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
@@ -301,7 +308,6 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         auth = self._generate_sapisidhash_header()
         if auth is not None:
             headers.update({'Authorization': auth, 'X-Origin': 'https://www.youtube.com'})
-
         return self._download_json(
             'https://www.youtube.com/youtubei/v1/%s' % ep,
             video_id=video_id, fatal=fatal, note=note, errnote=errnote,
@@ -315,6 +321,27 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                  self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'),
             video_id)
 
+    def _extract_identity_token(self, webpage, item_id):
+        ytcfg = self._extract_ytcfg(item_id, webpage)
+        if ytcfg:
+            token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
+            if token:
+                return token
+        return self._search_regex(
+            r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
+            'identity token', default=None)
+
+    @staticmethod
+    def _extract_account_syncid(data):
+        """Extract syncId required to download private playlists of secondary channels"""
+        sync_ids = (
+            try_get(data, lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'], compat_str)
+            or '').split("||")
+        if len(sync_ids) >= 2 and sync_ids[1]:
+            # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
+            # and just "user_syncid||" for primary channel. We only want the channel_syncid
+            return sync_ids[0]
+
     def _extract_ytcfg(self, video_id, webpage):
         return self._parse_json(
             self._search_regex(
@@ -1462,6 +1489,270 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
              regex), webpage, name, default='{}'), video_id, fatal=False)
 
+    @staticmethod
+    def _join_text_entries(runs):
+        text = None
+        for run in runs:
+            if not isinstance(run, dict):
+                continue
+            sub_text = try_get(run, lambda x: x['text'], compat_str)
+            if sub_text:
+                if not text:
+                    text = sub_text
+                    continue
+                text += sub_text
+        return text
+
+    def _extract_comment(self, comment_renderer, parent=None):
+        comment_id = comment_renderer.get('commentId')
+        if not comment_id:
+            return
+        comment_text_runs = try_get(comment_renderer, lambda x: x['contentText']['runs']) or []
+        text = self._join_text_entries(comment_text_runs) or ''
+        comment_time_text = try_get(comment_renderer, lambda x: x['publishedTimeText']['runs']) or []
+        time_text = self._join_text_entries(comment_time_text)
+
+        author = try_get(comment_renderer, lambda x: x['authorText']['simpleText'], compat_str)
+        author_id = try_get(comment_renderer,
+                            lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
+        votes = str_to_int(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
+                                                      lambda x: x['likeCount']), compat_str)) or 0
+        author_thumbnail = try_get(comment_renderer,
+                                   lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
+
+        author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
+        is_liked = try_get(comment_renderer, lambda x: x['isLiked'], bool)
+
+        return {
+            'id': comment_id,
+            'text': text,
+            # TODO: This should be parsed to timestamp
+            'time_text': time_text,
+            'like_count': votes,
+            'is_favorited': is_liked,
+            'author': author,
+            'author_id': author_id,
+            'author_thumbnail': author_thumbnail,
+            'author_is_uploader': author_is_uploader,
+            'parent': parent or 'root'
+        }
+
+    def _comment_entries(self, root_continuation_data, identity_token, account_syncid,
+                         session_token_list, parent=None, comment_counts=None):
+
+        def extract_thread(parent_renderer):
+            contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
+            if not parent:
+                comment_counts[2] = 0
+            for content in contents:
+                comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
+                comment_renderer = try_get(
+                    comment_thread_renderer, (lambda x: x['comment']['commentRenderer'], dict)) or try_get(
+                    content, (lambda x: x['commentRenderer'], dict))
+
+                if not comment_renderer:
+                    continue
+                comment = self._extract_comment(comment_renderer, parent)
+                if not comment:
+                    continue
+                comment_counts[0] += 1
+                yield comment
+                # Attempt to get the replies
+                comment_replies_renderer = try_get(
+                    comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
+
+                if comment_replies_renderer:
+                    comment_counts[2] += 1
+                    comment_entries_iter = self._comment_entries(
+                        comment_replies_renderer, identity_token, account_syncid,
+                        parent=comment.get('id'), session_token_list=session_token_list,
+                        comment_counts=comment_counts)
+
+                    for reply_comment in comment_entries_iter:
+                        yield reply_comment
+
+        if not comment_counts:
+            # comment so far, est. total comments, current comment thread #
+            comment_counts = [0, 0, 0]
+        headers = self._DEFAULT_BASIC_API_HEADERS.copy()
+
+        # TODO: Generalize the download code with TabIE
+        if identity_token:
+            headers['x-youtube-identity-token'] = identity_token
+
+        if account_syncid:
+            headers['X-Goog-PageId'] = account_syncid
+            headers['X-Goog-AuthUser'] = 0
+
+        continuation = YoutubeTabIE._extract_continuation(root_continuation_data)  # TODO
+        first_continuation = False
+        if parent is None:
+            first_continuation = True
+
+        for page_num in itertools.count(0):
+            if not continuation:
+                break
+            retries = self._downloader.params.get('extractor_retries', 3)
+            count = -1
+            last_error = None
+
+            while count < retries:
+                count += 1
+                if last_error:
+                    self.report_warning('%s. Retrying ...' % last_error)
+                try:
+                    query = {
+                        'ctoken': continuation['ctoken'],
+                        'pbj': 1,
+                        'type': 'next',
+                    }
+                    if parent:
+                        query['action_get_comment_replies'] = 1
+                    else:
+                        query['action_get_comments'] = 1
+
+                    comment_prog_str = '(%d/%d)' % (comment_counts[0], comment_counts[1])
+                    if page_num == 0:
+                        if first_continuation:
+                            note_prefix = "Downloading initial comment continuation page"
+                        else:
+                            note_prefix = "    Downloading comment reply thread %d %s" % (comment_counts[2], comment_prog_str)
+                    else:
+                        note_prefix = "%sDownloading comment%s page %d %s" % (
+                            "       " if parent else "",
+                            ' replies' if parent else '',
+                            page_num,
+                            comment_prog_str)
+
+                    browse = self._download_json(
+                        'https://www.youtube.com/comment_service_ajax', None,
+                        '%s %s' % (note_prefix, '(retry #%d)' % count if count else ''),
+                        headers=headers, query=query,
+                        data=urlencode_postdata({
+                            'session_token': session_token_list[0]
+                        }))
+                except ExtractorError as e:
+                    if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404, 413):
+                        if e.cause.code == 413:
+                            self.report_warning("Assumed end of comments (received HTTP Error 413)")
+                            return
+                        # Downloading page may result in intermittent 5xx HTTP error
+                        # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
+                        last_error = 'HTTP Error %s' % e.cause.code
+                        if e.cause.code == 404:
+                            last_error = last_error + " (this API is probably deprecated)"
+                        if count < retries:
+                            continue
+                    raise
+                else:
+                    session_token = try_get(browse, lambda x: x['xsrf_token'], compat_str)
+                    if session_token:
+                        session_token_list[0] = session_token
+
+                    response = try_get(browse,
+                                       (lambda x: x['response'],
+                                        lambda x: x[1]['response'])) or {}
+
+                    if response.get('continuationContents'):
+                        break
+
+                    # YouTube sometimes gives reload: now json if something went wrong (e.g. bad auth)
+                    if browse.get('reload'):
+                        raise ExtractorError("Invalid or missing params in continuation request", expected=False)
+
+                    # TODO: not tested, merged from old extractor
+                    err_msg = browse.get('externalErrorMessage')
+                    if err_msg:
+                        raise ExtractorError('YouTube said: %s' % err_msg, expected=False)
+
+                    # Youtube sometimes sends incomplete data
+                    # See: https://github.com/ytdl-org/youtube-dl/issues/28194
+                    last_error = 'Incomplete data received'
+                    if count >= retries:
+                        self._downloader.report_error(last_error)
+
+            if not response:
+                break
+
+            known_continuation_renderers = {
+                'itemSectionContinuation': extract_thread,
+                'commentRepliesContinuation': extract_thread
+            }
+
+            # extract next root continuation from the results
+            continuation_contents = try_get(
+                response, lambda x: x['continuationContents'], dict) or {}
+
+            for key, value in continuation_contents.items():
+                if key not in known_continuation_renderers:
+                    continue
+                continuation_renderer = value
+
+                if first_continuation:
+                    first_continuation = False
+                    expected_comment_count = try_get(
+                        continuation_renderer,
+                        (lambda x: x['header']['commentsHeaderRenderer']['countText']['runs'][0]['text'],
+                         lambda x: x['header']['commentsHeaderRenderer']['commentsCount']['runs'][0]['text']),
+                        compat_str)
+
+                    if expected_comment_count:
+                        comment_counts[1] = str_to_int(expected_comment_count)
+                        self.to_screen("Downloading ~%d comments" % str_to_int(expected_comment_count))
+                        yield comment_counts[1]
+
+                    # TODO: cli arg.
+                    # 1/True for newest, 0/False for popular (default)
+                    comment_sort_index = int(True)
+                    sort_continuation_renderer = try_get(
+                        continuation_renderer,
+                        lambda x: x['header']['commentsHeaderRenderer']['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems']
+                        [comment_sort_index]['continuation']['reloadContinuationData'], dict)
+                    # If this fails, the initial continuation page
+                    # starts off with popular anyways.
+                    if sort_continuation_renderer:
+                        continuation = YoutubeTabIE._build_continuation_query(
+                            continuation=sort_continuation_renderer.get('continuation'),
+                            ctp=sort_continuation_renderer.get('clickTrackingParams'))
+                        self.to_screen("Sorting comments by %s" % ('popular' if comment_sort_index == 0 else 'newest'))
+                        break
+
+                for entry in known_continuation_renderers[key](continuation_renderer):
+                    yield entry
+
+                continuation = YoutubeTabIE._extract_continuation(continuation_renderer)  # TODO
+                break
+
+    def _extract_comments(self, ytcfg, video_id, contents, webpage, xsrf_token):
+        """Entry for comment extraction"""
+        comments = []
+        known_entry_comment_renderers = (
+            'itemSectionRenderer',
+        )
+        estimated_total = 0
+        for entry in contents:
+            for key, renderer in entry.items():
+                if key not in known_entry_comment_renderers:
+                    continue
+
+                comment_iter = self._comment_entries(
+                    renderer,
+                    identity_token=self._extract_identity_token(webpage, item_id=video_id),
+                    account_syncid=self._extract_account_syncid(ytcfg),
+                    session_token_list=[xsrf_token])
+
+                for comment in comment_iter:
+                    if isinstance(comment, int):
+                        estimated_total = comment
+                        continue
+                    comments.append(comment)
+                break
+        self.to_screen("Downloaded %d/%d comments" % (len(comments), estimated_total))
+        return {
+            'comments': comments,
+            'comment_count': len(comments),
+        }
+
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
         video_id = self._match_id(url)
@@ -2024,156 +2315,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     errnote='Unable to download video annotations', fatal=False,
                     data=urlencode_postdata({xsrf_field_name: xsrf_token}))
 
-        # Get comments
-        # TODO: Refactor and move to seperate function
-        def extract_comments():
-            expected_video_comment_count = 0
-            video_comments = []
-            comment_xsrf = xsrf_token
-
-            def find_value(html, key, num_chars=2, separator='"'):
-                pos_begin = html.find(key) + len(key) + num_chars
-                pos_end = html.find(separator, pos_begin)
-                return html[pos_begin: pos_end]
-
-            def search_dict(partial, key):
-                if isinstance(partial, dict):
-                    for k, v in partial.items():
-                        if k == key:
-                            yield v
-                        else:
-                            for o in search_dict(v, key):
-                                yield o
-                elif isinstance(partial, list):
-                    for i in partial:
-                        for o in search_dict(i, key):
-                            yield o
-
-            continuations = []
-            if initial_data:
-                try:
-                    ncd = next(search_dict(initial_data, 'nextContinuationData'))
-                    continuations = [ncd['continuation']]
-                # Handle videos where comments have been disabled entirely
-                except StopIteration:
-                    pass
-
-            def get_continuation(continuation, session_token, replies=False):
-                query = {
-                    'pbj': 1,
-                    'ctoken': continuation,
-                }
-                if replies:
-                    query['action_get_comment_replies'] = 1
-                else:
-                    query['action_get_comments'] = 1
-
-                while True:
-                    content, handle = self._download_webpage_handle(
-                        'https://www.youtube.com/comment_service_ajax',
-                        video_id,
-                        note=False,
-                        expected_status=[413],
-                        data=urlencode_postdata({
-                            'session_token': session_token
-                        }),
-                        query=query,
-                        headers={
-                            'Accept': '*/*',
-                            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:76.0) Gecko/20100101 Firefox/76.0',
-                            'X-YouTube-Client-Name': '1',
-                            'X-YouTube-Client-Version': '2.20201202.06.01'
-                        }
-                    )
-
-                    response_code = handle.getcode()
-                    if (response_code == 200):
-                        return self._parse_json(content, video_id)
-                    if (response_code == 413):
-                        return None
-                    raise ExtractorError('Unexpected HTTP error code: %s' % response_code)
-
-            first_continuation = True
-            chain_msg = ''
-            self.to_screen('Downloading comments')
-            while continuations:
-                continuation = continuations.pop()
-                comment_response = get_continuation(continuation, comment_xsrf)
-                if not comment_response:
-                    continue
-                if list(search_dict(comment_response, 'externalErrorMessage')):
-                    raise ExtractorError('Error returned from server: ' + next(search_dict(comment_response, 'externalErrorMessage')))
-
-                if 'continuationContents' not in comment_response['response']:
-                    # Something is wrong here. Youtube won't accept this continuation token for some reason and responds with a user satisfaction dialog (error?)
-                    continue
-                # not sure if this actually helps
-                if 'xsrf_token' in comment_response:
-                    comment_xsrf = comment_response['xsrf_token']
-
-                item_section = comment_response['response']['continuationContents']['itemSectionContinuation']
-                if first_continuation:
-                    expected_video_comment_count = int(item_section['header']['commentsHeaderRenderer']['countText']['runs'][0]['text'].replace(' Comments', '').replace('1 Comment', '1').replace(',', ''))
-                    first_continuation = False
-                if 'contents' not in item_section:
-                    # continuation returned no comments?
-                    # set an empty array as to not break the for loop
-                    item_section['contents'] = []
-
-                for meta_comment in item_section['contents']:
-                    comment = meta_comment['commentThreadRenderer']['comment']['commentRenderer']
-                    video_comments.append({
-                        'id': comment['commentId'],
-                        'text': ''.join([c['text'] for c in try_get(comment, lambda x: x['contentText']['runs'], list) or []]),
-                        'time_text': ''.join([c['text'] for c in comment['publishedTimeText']['runs']]),
-                        'author': comment.get('authorText', {}).get('simpleText', ''),
-                        'votes': comment.get('voteCount', {}).get('simpleText', '0'),
-                        'author_thumbnail': comment['authorThumbnail']['thumbnails'][-1]['url'],
-                        'parent': 'root'
-                    })
-                    if 'replies' not in meta_comment['commentThreadRenderer']:
-                        continue
-
-                    reply_continuations = [rcn['nextContinuationData']['continuation'] for rcn in meta_comment['commentThreadRenderer']['replies']['commentRepliesRenderer']['continuations']]
-                    while reply_continuations:
-                        time.sleep(1)
-                        continuation = reply_continuations.pop()
-                        replies_data = get_continuation(continuation, comment_xsrf, True)
-                        if not replies_data or 'continuationContents' not in replies_data[1]['response']:
-                            continue
-
-                        if self._downloader.params.get('verbose', False):
-                            chain_msg = ' (chain %s)' % comment['commentId']
-                        self.to_screen('Comments downloaded: %d of ~%d%s' % (len(video_comments), expected_video_comment_count, chain_msg))
-                        reply_comment_meta = replies_data[1]['response']['continuationContents']['commentRepliesContinuation']
-                        for reply_meta in reply_comment_meta.get('contents', {}):
-                            reply_comment = reply_meta['commentRenderer']
-                            video_comments.append({
-                                'id': reply_comment['commentId'],
-                                'text': ''.join([c['text'] for c in reply_comment['contentText']['runs']]),
-                                'time_text': ''.join([c['text'] for c in reply_comment['publishedTimeText']['runs']]),
-                                'author': reply_comment.get('authorText', {}).get('simpleText', ''),
-                                'votes': reply_comment.get('voteCount', {}).get('simpleText', '0'),
-                                'author_thumbnail': reply_comment['authorThumbnail']['thumbnails'][-1]['url'],
-                                'parent': comment['commentId']
-                            })
-                        if 'continuations' not in reply_comment_meta or len(reply_comment_meta['continuations']) == 0:
-                            continue
-                        reply_continuations += [rcn['nextContinuationData']['continuation'] for rcn in reply_comment_meta['continuations']]
-
-                self.to_screen('Comments downloaded: %d of ~%d' % (len(video_comments), expected_video_comment_count))
-                if 'continuations' in item_section:
-                    continuations += [ncd['nextContinuationData']['continuation'] for ncd in item_section['continuations']]
-                time.sleep(1)
-
-            self.to_screen('Total comments downloaded: %d of ~%d' % (len(video_comments), expected_video_comment_count))
-            return {
-                'comments': video_comments,
-                'comment_count': expected_video_comment_count
-            }
-
         if get_comments:
-            info['__post_extractor'] = extract_comments
+            info['__post_extractor'] = lambda: self._extract_comments(ytcfg, video_id, contents, webpage, xsrf_token)
 
         self.mark_watched(video_id, player_response)
 
@@ -2520,17 +2663,22 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             channel_url, 'channel id')
 
     @staticmethod
-    def _extract_grid_item_renderer(item):
+    def _extract_basic_item_renderer(item):
-        for item_kind in ('Playlist', 'Video', 'Channel'):
+        # Modified from _extract_grid_item_renderer
-            renderer = item.get('grid%sRenderer' % item_kind)
+        known_renderers = (
-            if renderer:
+            'playlistRenderer', 'videoRenderer', 'channelRenderer'
-                return renderer
+            'gridPlaylistRenderer', 'gridVideoRenderer', 'gridChannelRenderer'
+        )
+        for key, renderer in item.items():
+            if key not in known_renderers:
+                continue
+            return renderer
 
     def _grid_entries(self, grid_renderer):
         for item in grid_renderer['items']:
             if not isinstance(item, dict):
                 continue
-            renderer = self._extract_grid_item_renderer(item)
+            renderer = self._extract_basic_item_renderer(item)
             if not isinstance(renderer, dict):
                 continue
             title = try_get(
@@ -2559,7 +2707,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
         content = shelf_renderer.get('content')
         if not isinstance(content, dict):
             return
-        renderer = content.get('gridRenderer')
+        renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
         if renderer:
             # TODO: add support for nested playlists so each shelf is processed
             # as separate playlist
@@ -2601,20 +2749,6 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 continue
             yield self._extract_video(renderer)
 
-    r""" # Not needed in the new implementation
-    def _itemSection_entries(self, item_sect_renderer):
-        for content in item_sect_renderer['contents']:
-            if not isinstance(content, dict):
-                continue
-            renderer = content.get('videoRenderer', {})
-            if not isinstance(renderer, dict):
-                continue
-            video_id = renderer.get('videoId')
-            if not video_id:
-                continue
-            yield self._extract_video(renderer)
-    """
-
     def _rich_entries(self, rich_grid_renderer):
         renderer = try_get(
             rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
@@ -2713,7 +2847,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             ctp = continuation_ep.get('clickTrackingParams')
             return YoutubeTabIE._build_continuation_query(continuation, ctp)
 
-    def _entries(self, tab, identity_token, item_id):
+    def _entries(self, tab, item_id, identity_token, account_syncid):
 
         def extract_entries(parent_renderer):  # this needs to called again for continuation to work with feeds
             contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
@@ -2773,6 +2907,10 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
         if identity_token:
             headers['x-youtube-identity-token'] = identity_token
 
+        if account_syncid:
+            headers['X-Goog-PageId'] = account_syncid
+            headers['X-Goog-AuthUser'] = 0
+
         for page_num in itertools.count(1):
             if not continuation:
                 break
@@ -2803,9 +2941,14 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 else:
                     # Youtube sometimes sends incomplete data
                     # See: https://github.com/ytdl-org/youtube-dl/issues/28194
-                    if response.get('continuationContents') or response.get('onResponseReceivedActions'):
+                    if dict_get(response,
+                                ('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints')):
                         break
-                    last_error = 'Incomplete data recieved'
+
+                    # Youtube may send alerts if there was an issue with the continuation page
+                    self._extract_alerts(response, expected=False)
+
+                    last_error = 'Incomplete data received'
                     if count >= retries:
                         self._downloader.report_error(last_error)
 
@@ -2837,11 +2980,13 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 'gridPlaylistRenderer': (self._grid_entries, 'items'),
                 'gridVideoRenderer': (self._grid_entries, 'items'),
                 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
-                'itemSectionRenderer': (self._playlist_entries, 'contents'),
+                'itemSectionRenderer': (extract_entries, 'contents'),  # for feeds
                 'richItemRenderer': (extract_entries, 'contents'),  # for hashtag
+                'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
             }
             continuation_items = try_get(
-                response, lambda x: x['onResponseReceivedActions'][0]['appendContinuationItemsAction']['continuationItems'], list)
+                response,
+                lambda x: dict_get(x, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))[0]['appendContinuationItemsAction']['continuationItems'], list)
             continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
             video_items_renderer = None
             for key, value in continuation_item.items():
@@ -2888,7 +3033,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                         try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
         return {k: v for k, v in uploader.items() if v is not None}
 
-    def _extract_from_tabs(self, item_id, webpage, data, tabs, identity_token):
+    def _extract_from_tabs(self, item_id, webpage, data, tabs):
         playlist_id = title = description = channel_url = channel_name = channel_id = None
         thumbnails_list = tags = []
 
@@ -2952,16 +3097,41 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'channel_id': metadata['uploader_id'],
             'channel_url': metadata['uploader_url']})
         return self.playlist_result(
-            self._entries(selected_tab, identity_token, playlist_id),
+            self._entries(
+                selected_tab, playlist_id,
+                self._extract_identity_token(webpage, item_id),
+                self._extract_account_syncid(data)),
             **metadata)
 
+    def _extract_mix_playlist(self, playlist, playlist_id):
+        first_id = last_id = None
+        for page_num in itertools.count(1):
+            videos = list(self._playlist_entries(playlist))
+            if not videos:
+                return
+            start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
+            if start >= len(videos):
+                return
+            for video in videos[start:]:
+                if video['id'] == first_id:
+                    self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
+                    return
+                yield video
+            first_id = first_id or videos[0]['id']
+            last_id = videos[-1]['id']
+
+            _, data = self._extract_webpage(
+                'https://www.youtube.com/watch?list=%s&v=%s' % (playlist_id, last_id),
+                '%s page %d' % (playlist_id, page_num))
+            playlist = try_get(
+                data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
+
     def _extract_from_playlist(self, item_id, url, data, playlist):
         title = playlist.get('title') or try_get(
             data, lambda x: x['titleText']['simpleText'], compat_str)
         playlist_id = playlist.get('playlistId') or item_id
-        # Inline playlist rendition continuation does not always work
+
-        # at Youtube side, so delegating regular tab-based playlist URL
+        # Delegating everything except mix playlists to regular tab-based playlist URL
-        # processing whenever possible.
         playlist_url = urljoin(url, try_get(
             playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
             compat_str))
@@ -2969,67 +3139,42 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             return self.url_result(
                 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
                 video_title=title)
-        return self.playlist_result(
-            self._playlist_entries(playlist), playlist_id=playlist_id,
-            playlist_title=title)
 
-    @staticmethod
+        return self.playlist_result(
-    def _extract_alerts(data):
+            self._extract_mix_playlist(playlist, playlist_id),
-        for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
+            playlist_id=playlist_id, playlist_title=title)
-            if not isinstance(alert_dict, dict):
+
-                continue
+    def _extract_alerts(self, data, expected=False):
-            for renderer in alert_dict:
+
-                alert = alert_dict[renderer]
+        def _real_extract_alerts():
-                alert_type = alert.get('type')
+            for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
-                if not alert_type:
+                if not isinstance(alert_dict, dict):
                     continue
-                message = try_get(alert, lambda x: x['text']['simpleText'], compat_str)
+                for alert in alert_dict.values():
-                if message:
+                    alert_type = alert.get('type')
-                    yield alert_type, message
+                    if not alert_type:
-                for run in try_get(alert, lambda x: x['text']['runs'], list) or []:
+                        continue
-                    message = try_get(run, lambda x: x['text'], compat_str)
+                    message = try_get(alert, lambda x: x['text']['simpleText'], compat_str)
                     if message:
                         yield alert_type, message
+                    for run in try_get(alert, lambda x: x['text']['runs'], list) or []:
+                        message = try_get(run, lambda x: x['text'], compat_str)
+                        if message:
+                            yield alert_type, message
 
-    def _extract_identity_token(self, webpage, item_id):
+        err_msg = None
-        ytcfg = self._extract_ytcfg(item_id, webpage)
+        for alert_type, alert_message in _real_extract_alerts():
-        if ytcfg:
+            if alert_type.lower() == 'error':
-            token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
+                if err_msg:
-            if token:
+                    self._downloader.report_warning('YouTube said: %s - %s' % ('ERROR', err_msg))
-                return token
+                err_msg = alert_message
-        return self._search_regex(
-            r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
-            'identity token', default=None)
-
-    def _real_extract(self, url):
-        item_id = self._match_id(url)
-        url = compat_urlparse.urlunparse(
-            compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
-        is_home = re.match(r'(?P<pre>%s)(?P<post>/?(?![^#?]).*$)' % self._VALID_URL, url)
-        if is_home is not None and is_home.group('not_channel') is None and item_id != 'feed':
-            self._downloader.report_warning(
-                'A channel/user page was given. All the channel\'s videos will be downloaded. '
-                'To download only the videos in the home page, add a "/featured" to the URL')
-            url = '%s/videos%s' % (is_home.group('pre'), is_home.group('post') or '')
-
-        # Handle both video/playlist URLs
-        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
-        video_id = qs.get('v', [None])[0]
-        playlist_id = qs.get('list', [None])[0]
-
-        if is_home is not None and is_home.group('not_channel') is not None and is_home.group('not_channel').startswith('watch') and not video_id:
-            if playlist_id:
-                self._downloader.report_warning('%s is not a valid Youtube URL. Trying to download playlist %s' % (url, playlist_id))
-                url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
-                # return self.url_result(playlist_id, ie=YoutubePlaylistIE.ie_key())
             else:
-                raise ExtractorError('Unable to recognize tab page')
+                self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
-        if video_id and playlist_id:
-            if self._downloader.params.get('noplaylist'):
-                self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
-                return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
-            self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
 
+        if err_msg:
+            raise ExtractorError('YouTube said: %s' % err_msg, expected=expected)
+
+    def _extract_webpage(self, url, item_id):
         retries = self._downloader.params.get('extractor_retries', 3)
         count = -1
         last_error = 'Incomplete yt initial data recieved'
@@ -3041,40 +3186,67 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 self.report_warning('%s. Retrying ...' % last_error)
             webpage = self._download_webpage(
                 url, item_id,
-                'Downloading webpage%s' % ' (retry #%d)' % count if count else '')
+                'Downloading webpage%s' % (' (retry #%d)' % count if count else ''))
-            identity_token = self._extract_identity_token(webpage, item_id)
             data = self._extract_yt_initial_data(item_id, webpage)
-            err_msg = None
+            self._extract_alerts(data, expected=True)
-            for alert_type, alert_message in self._extract_alerts(data):
-                if alert_type.lower() == 'error':
-                    if err_msg:
-                        self._downloader.report_warning('YouTube said: %s - %s' % ('ERROR', err_msg))
-                    err_msg = alert_message
-                else:
-                    self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
-            if err_msg:
-                raise ExtractorError('YouTube said: %s' % err_msg, expected=True)
             if data.get('contents') or data.get('currentVideoEndpoint'):
                 break
             if count >= retries:
                 self._downloader.report_error(last_error)
+        return webpage, data
+
+    def _real_extract(self, url):
+        item_id = self._match_id(url)
+        url = compat_urlparse.urlunparse(
+            compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
+
+        # This is not matched in a channel page with a tab selected
+        mobj = re.match(r'(?P<pre>%s)(?P<post>/?(?![^#?]).*$)' % self._VALID_URL, url)
+        mobj = mobj.groupdict() if mobj else {}
+        if mobj and not mobj.get('not_channel'):
+            self._downloader.report_warning(
+                'A channel/user page was given. All the channel\'s videos will be downloaded. '
+                'To download only the videos in the home page, add a "/featured" to the URL')
+            url = '%s/videos%s' % (mobj.get('pre'), mobj.get('post') or '')
+
+        # Handle both video/playlist URLs
+        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
+        video_id = qs.get('v', [None])[0]
+        playlist_id = qs.get('list', [None])[0]
+
+        if not video_id and (mobj.get('not_channel') or '').startswith('watch'):
+            if not playlist_id:
+                # If there is neither video or playlist ids,
+                # youtube redirects to home page, which is undesirable
+                raise ExtractorError('Unable to recognize tab page')
+            self._downloader.report_warning('A video URL was given without video ID. Trying to download playlist %s' % playlist_id)
+            url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
+
+        if video_id and playlist_id:
+            if self._downloader.params.get('noplaylist'):
+                self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
+                return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
+            self.to_screen('Downloading playlist %s; add --no-playlist to just download video %s' % (playlist_id, video_id))
+
+        webpage, data = self._extract_webpage(url, item_id)
 
         tabs = try_get(
             data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
         if tabs:
-            return self._extract_from_tabs(item_id, webpage, data, tabs, identity_token)
+            return self._extract_from_tabs(item_id, webpage, data, tabs)
+
         playlist = try_get(
             data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
         if playlist:
             return self._extract_from_playlist(item_id, url, data, playlist)
-        # Fallback to video extraction if no playlist alike page is recognized.
+
-        # First check for the current video then try the v attribute of URL query.
         video_id = try_get(
             data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'],
             compat_str) or video_id
         if video_id:
+            self._downloader.report_warning('Unable to recognize playlist. Downloading just video %s' % video_id)
             return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
-        # Failed to recognize
+
         raise ExtractorError('Unable to recognize tab page')
 
 
@@ -3338,7 +3510,6 @@ class YoutubeFeedsInfoExtractor(YoutubeTabIE):
     Subclasses must define the _FEED_NAME property.
     """
     _LOGIN_REQUIRED = True
-    # _MAX_PAGES = 5
     _TESTS = []
 
     @property

yt_dlp/extractor/zee5.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     int_or_none,
     parse_age_limit,
@@ -16,24 +17,34 @@ from ..utils import (
 
 
 class Zee5IE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?zee5\.com/[^#?]*/(?P<display_id>[-\w]+)/(?P<id>[-\d]+)'
+    _VALID_URL = r'''(?x)
+                     (?:
+                        zee5:|
+                        (?:https?://)(?:www\.)?zee5\.com/(?:[^#?]+/)?
+                        (?:
+                            (?:tvshows|kids|zee5originals)(?:/[^#/?]+){3}
+                            |movies/[^#/?]+
+                        )/(?P<display_id>[^#/?]+)/
+                     )
+                     (?P<id>[^#/?]+)/?(?:$|[?#])
+                     '''
     _TESTS = [{
         'url': 'https://www.zee5.com/movies/details/krishna-the-birth/0-0-63098',
         'info_dict': {
-            "id": "0-0-63098",
+            'id': '0-0-63098',
-            "ext": "m3u8",
+            'ext': 'mp4',
-            "display_id": "krishna-the-birth",
+            'display_id': 'krishna-the-birth',
-            "title": "Krishna - The Birth",
+            'title': 'Krishna - The Birth',
-            "duration": 4368,
+            'duration': 4368,
-            "average_rating": 4,
+            'average_rating': 4,
-            "description": str,
+            'description': str,
-            "alt_title": "Krishna - The Birth",
+            'alt_title': 'Krishna - The Birth',
-            "uploader": "Zee Entertainment Enterprises Ltd",
+            'uploader': 'Zee Entertainment Enterprises Ltd',
-            "release_date": "20060101",
+            'release_date': '20060101',
-            "upload_date": "20060101",
+            'upload_date': '20060101',
-            "timestamp": 1136073600,
+            'timestamp': 1136073600,
-            "thumbnail": "https://akamaividz.zee5.com/resources/0-0-63098/list/270x152/0063098_list_80888170.jpg",
+            'thumbnail': 'https://akamaividz.zee5.com/resources/0-0-63098/list/270x152/0063098_list_80888170.jpg',
-            "tags": list
+            'tags': list
         },
         'params': {
             'format': 'bv',
@@ -41,37 +52,43 @@ class Zee5IE(InfoExtractor):
     }, {
         'url': 'https://zee5.com/tvshows/details/krishna-balram/0-6-1871/episode-1-the-test-of-bramha/0-1-233402',
         'info_dict': {
-            "id": "0-1-233402",
+            'id': '0-1-233402',
-            'ext': 'm3u8',
+            'ext': 'mp4',
-            "display_id": "episode-1-the-test-of-bramha",
+            'display_id': 'episode-1-the-test-of-bramha',
-            "title": "Episode 1 - The Test Of Bramha",
+            'title': 'Episode 1 - The Test Of Bramha',
-            "duration": 1336,
+            'duration': 1336,
-            "average_rating": 4,
+            'average_rating': 4,
-            "description": str,
+            'description': str,
-            "alt_title": "Episode 1 - The Test Of Bramha",
+            'alt_title': 'Episode 1 - The Test Of Bramha',
-            "uploader": "Green Gold",
+            'uploader': 'Green Gold',
-            "release_date": "20090101",
+            'release_date': '20090101',
-            "upload_date": "20090101",
+            'upload_date': '20090101',
-            "timestamp": 1230768000,
+            'timestamp': 1230768000,
-            "thumbnail": "https://akamaividz.zee5.com/resources/0-1-233402/list/270x152/01233402_list.jpg",
+            'thumbnail': 'https://akamaividz.zee5.com/resources/0-1-233402/list/270x152/01233402_list.jpg',
-            "series": "Krishna Balram",
+            'series': 'Krishna Balram',
-            "season_number": 1,
+            'season_number': 1,
-            "episode_number": 1,
+            'episode_number': 1,
-            "tags": list,
+            'tags': list,
         },
         'params': {
             'format': 'bv',
         },
+    }, {
+        'url': 'https://www.zee5.com/hi/tvshows/details/kundali-bhagya/0-6-366/kundali-bhagya-march-08-2021/0-1-manual_7g9jv1os7730?country=IN',
+        'only_matching': True
+    }, {
+        'url': 'https://www.zee5.com/global/hi/tvshows/details/kundali-bhagya/0-6-366/kundali-bhagya-march-08-2021/0-1-manual_7g9jv1os7730',
+        'only_matching': True
     }]
 
     def _real_extract(self, url):
         video_id, display_id = re.match(self._VALID_URL, url).group('id', 'display_id')
         access_token_request = self._download_json(
             'https://useraction.zee5.com/token/platform_tokens.php?platform_name=web_app',
-            video_id, note="Downloading access token")
+            video_id, note='Downloading access token')
         token_request = self._download_json(
             'https://useraction.zee5.com/tokennd',
-            video_id, note="Downloading video token")
+            video_id, note='Downloading video token')
         json_data = self._download_json(
             'https://gwapi.zee5.com/content/details/{}?translation=en&country=IN'.format(video_id),
             video_id, headers={'X-Access-Token': access_token_request['token']})
@@ -111,3 +128,78 @@ class Zee5IE(InfoExtractor):
             'episode_number': int_or_none(try_get(json_data, lambda x: x['index'])),
             'tags': try_get(json_data, lambda x: x['tags'], list)
         }
+
+
+class Zee5SeriesIE(InfoExtractor):
+    IE_NAME = 'zee5:series'
+    _VALID_URL = r'''(?x)
+                     (?:
+                        zee5:series:|
+                        (?:https?://)(?:www\.)?zee5\.com/(?:[^#?]+/)?
+                        (?:tvshows|kids|zee5originals)(?:/[^#/?]+){2}/
+                     )
+                     (?P<id>[^#/?]+)/?(?:$|[?#])
+                     '''
+    _TESTS = [{
+        'url': 'https://www.zee5.com/kids/kids-shows/krishna-balram/0-6-1871',
+        'playlist_mincount': 43,
+        'info_dict': {
+            'id': '0-6-1871',
+        },
+    }, {
+        'url': 'https://www.zee5.com/tvshows/details/bhabi-ji-ghar-par-hai/0-6-199',
+        'playlist_mincount': 1500,
+        'info_dict': {
+            'id': '0-6-199',
+        },
+    }, {
+        'url': 'https://www.zee5.com/tvshows/details/agent-raghav-crime-branch/0-6-965',
+        'playlist_mincount': 25,
+        'info_dict': {
+            'id': '0-6-965',
+        },
+    }, {
+        'url': 'https://www.zee5.com/ta/tvshows/details/nagabhairavi/0-6-3201',
+        'playlist_mincount': 3,
+        'info_dict': {
+            'id': '0-6-3201',
+        },
+    }, {
+        'url': 'https://www.zee5.com/global/hi/tvshows/details/khwaabon-ki-zamin-par/0-6-270',
+        'playlist_mincount': 150,
+        'info_dict': {
+            'id': '0-6-270',
+        },
+    }
+    ]
+
+    def _entries(self, show_id):
+        access_token_request = self._download_json(
+            'https://useraction.zee5.com/token/platform_tokens.php?platform_name=web_app',
+            show_id, note='Downloading access token')
+        headers = {
+            'X-Access-Token': access_token_request['token'],
+            'Referer': 'https://www.zee5.com/',
+        }
+        show_url = 'https://gwapi.zee5.com/content/tvshow/{}?translation=en&country=IN'.format(show_id)
+
+        page_num = 0
+        show_json = self._download_json(show_url, video_id=show_id, headers=headers)
+        for season in show_json.get('seasons') or []:
+            season_id = try_get(season, lambda x: x['id'], compat_str)
+            next_url = 'https://gwapi.zee5.com/content/tvshow/?season_id={}&type=episode&translation=en&country=IN&on_air=false&asset_subtype=tvshow&page=1&limit=100'.format(season_id)
+            while next_url:
+                page_num += 1
+                episodes_json = self._download_json(
+                    next_url, video_id=show_id, headers=headers,
+                    note='Downloading JSON metadata page %d' % page_num)
+                for episode in try_get(episodes_json, lambda x: x['episode'], list) or []:
+                    video_id = episode.get('id')
+                    yield self.url_result(
+                        'zee5:%s' % video_id,
+                        ie=Zee5IE.ie_key(), video_id=video_id)
+                next_url = url_or_none(episodes_json.get('next_episode_api'))
+
+    def _real_extract(self, url):
+        show_id = self._match_id(url)
+        return self.playlist_result(self._entries(show_id), playlist_id=show_id)

yt_dlp/options.py

@@ -214,12 +214,11 @@ def parseOpts(overrideArguments=None):
         help='Mark videos watched (YouTube only)')
     general.add_option(
         '--no-mark-watched',
-        action='store_false', dest='mark_watched', default=False,
+        action='store_false', dest='mark_watched',
-        help='Do not mark videos watched')
+        help='Do not mark videos watched (default)')
     general.add_option(
         '--no-colors',
-        action='store_true', dest='no_color',
+        action='store_true', dest='no_color', default=False,
-        default=False,
         help='Do not emit color codes in output')
 
     network = optparse.OptionGroup(parser, 'Network Options')
@@ -558,6 +557,10 @@ def parseOpts(overrideArguments=None):
         help='Languages of the subtitles to download (optional) separated by commas, use --list-subs for available language tags')
 
     downloader = optparse.OptionGroup(parser, 'Download Options')
+    downloader.add_option(
+        '-N', '--concurrent-fragments',
+        dest='concurrent_fragment_downloads', metavar='N', default=1, type=int,
+        help='Number of fragments to download concurrently (default is %default)')
     downloader.add_option(
         '-r', '--limit-rate', '--rate-limit',
         dest='ratelimit', metavar='RATE',
@@ -1087,8 +1090,8 @@ def parseOpts(overrideArguments=None):
             'The supported executables are: SponSkrub, FFmpeg, FFprobe, and AtomicParsley. '
             'You can also specify "PP+EXE:ARGS" to give the arguments to the specified executable '
             'only when being used by the specified postprocessor. Additionally, for ffmpeg/ffprobe, '
-            'a number can be appended to the exe name seperated by "_i" to pass the argument '
+            '"_i"/"_o" can be appended to the prefix optionally followed by a number to pass the argument '
-            'before the specified input file. Eg: --ppa "Merger+ffmpeg_i1:-v quiet". '
+            'before the specified input/output file. Eg: --ppa "Merger+ffmpeg_i1:-v quiet". '
             'You can use this option multiple times to give different arguments to different '
             'postprocessors. (Alias: --ppa)'))
     postproc.add_option(
@@ -1154,7 +1157,7 @@ def parseOpts(overrideArguments=None):
         help='Write metadata to the video file\'s xattrs (using dublin core and xdg standards)')
     postproc.add_option(
         '--fixup',
-        metavar='POLICY', dest='fixup', default='detect_or_warn',
+        metavar='POLICY', dest='fixup', default=None,
         help=(
             'Automatically correct known faults of the file. '
             'One of never (do nothing), warn (only emit a warning), '
@@ -1179,6 +1182,17 @@ def parseOpts(overrideArguments=None):
         '--convert-subs', '--convert-subtitles',
         metavar='FORMAT', dest='convertsubtitles', default=None,
         help='Convert the subtitles to other format (currently supported: srt|ass|vtt|lrc)')
+    postproc.add_option(
+        '--split-chapters', '--split-tracks',
+        dest='split_chapters', action='store_true', default=False,
+        help=(
+            'Split video into multiple files based on internal chapters. '
+            'The "chapter:" prefix can be used with "--paths" and "--output" to '
+            'set the output filename for the split files. See "OUTPUT TEMPLATE" for details'))
+    postproc.add_option(
+        '--no-split-chapters', '--no-split-tracks',
+        dest='split_chapters', action='store_false',
+        help='Do not split video based on chapters (default)')
 
     sponskrub = optparse.OptionGroup(parser, 'SponSkrub (SponsorBlock) Options', description=(
         'SponSkrub (https://github.com/yt-dlp/SponSkrub) is a utility to mark/remove sponsor segments '

yt_dlp/postprocessor/__init__.py

@@ -13,6 +13,7 @@ from .ffmpeg import (
     FFmpegVideoConvertorPP,
     FFmpegVideoRemuxerPP,
     FFmpegSubtitlesConvertorPP,
+    FFmpegSplitChaptersPP,
 )
 from .xattrpp import XAttrMetadataPP
 from .execafterdownload import ExecAfterDownloadPP
@@ -31,6 +32,7 @@ __all__ = [
     'ExecAfterDownloadPP',
     'FFmpegEmbedSubtitlePP',
     'FFmpegExtractAudioPP',
+    'FFmpegSplitChaptersPP',
     'FFmpegFixupM3u8PP',
     'FFmpegFixupM4aPP',
     'FFmpegFixupStretchedPP',

yt_dlp/postprocessor/common.py

@@ -91,10 +91,18 @@ class PostProcessor(object):
         except Exception:
             self.report_warning(errnote)
 
-    def _configuration_args(self, *args, **kwargs):
+    def _configuration_args(self, exe, keys=None, default=[], use_compat=True):
+        pp_key = self.pp_key().lower()
+        exe = exe.lower()
+        root_key = exe if pp_key == exe else '%s+%s' % (pp_key, exe)
+        keys = ['%s%s' % (root_key, k) for k in (keys or [''])]
+        if root_key in keys:
+            keys += [root_key] + ([] if pp_key == exe else [(self.pp_key(), exe)]) + ['default']
+        else:
+            use_compat = False
         return cli_configuration_args(
             self._downloader.params.get('postprocessor_args'),
-            self.pp_key().lower(), *args, **kwargs)
+            keys, default, use_compat)
 
 
 class AudioConversionError(PostProcessingError):

yt_dlp/postprocessor/embedthumbnail.py

@@ -85,6 +85,8 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
             thumbnail_filename = thumbnail_jpg_filename
             thumbnail_ext = 'jpg'
 
+        mtime = os.stat(encodeFilename(filename)).st_mtime
+
         success = True
         if info['ext'] == 'mp3':
             options = [
@@ -139,7 +141,7 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
                        encodeFilename(thumbnail_filename, True),
                        encodeArgument('-o'),
                        encodeFilename(temp_filename, True)]
-                cmd += [encodeArgument(o) for o in self._configuration_args(exe='AtomicParsley')]
+                cmd += [encodeArgument(o) for o in self._configuration_args('AtomicParsley')]
 
                 self.to_screen('Adding thumbnail to "%s"' % filename)
                 self.write_debug('AtomicParsley command line: %s' % shell_quote(cmd))
@@ -187,6 +189,8 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
             os.remove(encodeFilename(filename))
             os.rename(encodeFilename(temp_filename), encodeFilename(filename))
 
+        self.try_utime(filename, mtime, mtime)
+
         files_to_delete = [thumbnail_filename]
         if self._already_have_thumbnail:
             info['__files_to_move'][original_thumbnail] = replace_extension(

yt_dlp/postprocessor/ffmpeg.py

@@ -10,6 +10,7 @@ import json
 
 from .common import AudioConversionError, PostProcessor
 
+from ..compat import compat_str
 from ..utils import (
     encodeArgument,
     encodeFilename,
@@ -234,25 +235,35 @@ class FFmpegPostProcessor(PostProcessor):
         return num, len(streams)
 
     def run_ffmpeg_multiple_files(self, input_paths, out_path, opts):
+        return self.real_run_ffmpeg(
+            [(path, []) for path in input_paths],
+            [(out_path, opts)])
+
+    def real_run_ffmpeg(self, input_path_opts, output_path_opts):
         self.check_version()
 
         oldest_mtime = min(
-            os.stat(encodeFilename(path)).st_mtime for path in input_paths)
+            os.stat(encodeFilename(path)).st_mtime for path, _ in input_path_opts)
 
         cmd = [encodeFilename(self.executable, True), encodeArgument('-y')]
         # avconv does not have repeat option
         if self.basename == 'ffmpeg':
             cmd += [encodeArgument('-loglevel'), encodeArgument('repeat+info')]
 
-        def make_args(file, pre=[], post=[], *args, **kwargs):
+        def make_args(file, args, name, number):
-            args = pre + self._configuration_args(*args, **kwargs) + post
+            keys = ['_%s%d' % (name, number), '_%s' % name]
+            if name == 'o' and number == 1:
+                keys.append('')
+            args += self._configuration_args(self.basename, keys)
+            if name == 'i':
+                args.append('-i')
             return (
-                [encodeArgument(o) for o in args]
+                [encodeArgument(arg) for arg in args]
                 + [encodeFilename(self._ffmpeg_filename_argument(file), True)])
 
-        for i, path in enumerate(input_paths):
+        for arg_type, path_opts in (('i', input_path_opts), ('o', output_path_opts)):
-            cmd += make_args(path, post=['-i'], exe='%s_i%d' % (self.basename, i + 1), use_default_arg=False)
+            cmd += [arg for i, o in enumerate(path_opts)
-        cmd += make_args(out_path, pre=opts, exe=self.basename)
+                    for arg in make_args(o[0], o[1], arg_type, i + 1)]
 
         self.write_debug('ffmpeg command line: %s' % shell_quote(cmd))
         p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE)
@@ -262,7 +273,8 @@ class FFmpegPostProcessor(PostProcessor):
             if self.get_param('verbose', False):
                 self.report_error(stderr)
             raise FFmpegPostProcessorError(stderr.split('\n')[-1])
-        self.try_utime(out_path, oldest_mtime, oldest_mtime)
+        for out_path, _ in output_path_opts:
+            self.try_utime(out_path, oldest_mtime, oldest_mtime)
         return stderr.decode('utf-8', 'replace')
 
     def run_ffmpeg(self, path, out_path, opts):
@@ -758,3 +770,40 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
                 }
 
         return sub_filenames, info
+
+
+class FFmpegSplitChaptersPP(FFmpegPostProcessor):
+
+    def _prepare_filename(self, number, chapter, info):
+        info = info.copy()
+        info.update({
+            'section_number': number,
+            'section_title': chapter.get('title'),
+            'section_start': chapter.get('start_time'),
+            'section_end': chapter.get('end_time'),
+        })
+        return self._downloader.prepare_filename(info, 'chapter')
+
+    def _ffmpeg_args_for_chapter(self, number, chapter, info):
+        destination = self._prepare_filename(number, chapter, info)
+        if not self._downloader._ensure_dir_exists(encodeFilename(destination)):
+            return
+
+        chapter['_filename'] = destination
+        self.to_screen('Chapter %03d; Destination: %s' % (number, destination))
+        return (
+            destination,
+            ['-ss', compat_str(chapter['start_time']),
+             '-to', compat_str(chapter['end_time'])])
+
+    def run(self, info):
+        chapters = info.get('chapters') or []
+        if not chapters:
+            self.report_warning('There are no tracks to extract')
+            return [], info
+
+        self.to_screen('Splitting video by chapters; %d chapters found' % len(chapters))
+        for idx, chapter in enumerate(chapters):
+            destination, opts = self._ffmpeg_args_for_chapter(idx + 1, chapter, info)
+            self.real_run_ffmpeg([(info['filepath'], opts)], [(destination, ['-c', 'copy'])])
+        return [], info

yt_dlp/postprocessor/sponskrub.py

@@ -71,7 +71,7 @@ class SponSkrubPP(PostProcessor):
         if not self.cutout:
             cmd += ['-chapter']
         cmd += compat_shlex_split(self.args)  # For backward compatibility
-        cmd += self._configuration_args(exe=self._exe_name, use_default_arg='no_compat')
+        cmd += self._configuration_args(self._exe_name, use_compat=False)
         cmd += ['--', information['id'], filename, temp_filename]
         cmd = [encodeArgument(i) for i in cmd]
 

yt_dlp/update.py

@@ -49,12 +49,16 @@ def update_self(to_screen, verbose, opener):
                 h.update(mv[:n])
         return h.hexdigest()
 
-    to_screen('Current Build Hash %s' % calc_sha256sum(sys.executable))
-
     if not isinstance(globals().get('__loader__'), zipimporter) and not hasattr(sys, 'frozen'):
         to_screen('It looks like you installed yt-dlp with a package manager, pip, setup.py or a tarball. Please use that to update.')
         return
 
+    # sys.executable is set to the full pathname of the exe-file for py2exe
+    # though symlinks are not followed so that we need to do this manually
+    # with help of realpath
+    filename = compat_realpath(sys.executable if hasattr(sys, 'frozen') else sys.argv[0])
+    to_screen('Current Build Hash %s' % calc_sha256sum(filename))
+
     # Download and check versions info
     try:
         version_info = opener.open(JSON_URL).read().decode('utf-8')
@@ -103,11 +107,6 @@ def update_self(to_screen, verbose, opener):
             (i[1] for i in hashes if i[0] == 'yt-dlp%s' % label),
             None)
 
-    # sys.executable is set to the full pathname of the exe-file for py2exe
-    # though symlinks are not followed so that we need to do this manually
-    # with help of realpath
-    filename = compat_realpath(sys.executable if hasattr(sys, 'frozen') else sys.argv[0])
-
     if not os.access(filename, os.W_OK):
         to_screen('ERROR: no write permissions on %s' % filename)
         return
@@ -198,28 +197,18 @@ def update_self(to_screen, verbose, opener):
             to_screen('Visit https://github.com/yt-dlp/yt-dlp/releases/latest')
             return
 
+        expected_sum = get_sha256sum('zip', py_ver)
+        if expected_sum and hashlib.sha256(newcontent).hexdigest() != expected_sum:
+            to_screen('ERROR: unable to verify the new zip')
+            to_screen('Visit https://github.com/yt-dlp/yt-dlp/releases/latest')
+            return
+
         try:
-            with open(filename + '.new', 'wb') as outf:
+            with open(filename, 'wb') as outf:
                 outf.write(newcontent)
         except (IOError, OSError):
             if verbose:
                 to_screen(encode_compat_str(traceback.format_exc()))
-            to_screen('ERROR: unable to write the new version')
-            return
-
-        expected_sum = get_sha256sum('zip', py_ver)
-        if expected_sum and calc_sha256sum(filename + '.new') != expected_sum:
-            to_screen('ERROR: unable to verify the new zip')
-            to_screen('Visit https://github.com/yt-dlp/yt-dlp/releases/latest')
-            try:
-                os.remove(filename + '.new')
-            except OSError:
-                to_screen('ERROR: unable to remove corrupt zip')
-            return
-
-        try:
-            os.rename(filename + '.new', filename)
-        except OSError:
             to_screen('ERROR: unable to overwrite current version')
             return
 

yt_dlp/utils.py

@@ -4182,8 +4182,10 @@ def qualities(quality_ids):
 
 DEFAULT_OUTTMPL = {
     'default': '%(title)s [%(id)s].%(ext)s',
+    'chapter': '%(title)s - %(section_number)03d %(section_title)s [%(id)s].%(ext)s',
 }
 OUTTMPL_TYPES = {
+    'chapter': None,
     'subtitle': None,
     'thumbnail': None,
     'description': 'description',
@@ -4692,36 +4694,26 @@ def cli_valueless_option(params, command_option, param, expected_value=True):
     return [command_option] if param == expected_value else []
 
 
-def cli_configuration_args(argdict, key, default=[], exe=None, use_default_arg=True):
+def cli_configuration_args(argdict, keys, default=[], use_compat=True):
-    # use_default_arg can be True, False, or 'no_compat'
     if isinstance(argdict, (list, tuple)):  # for backward compatibility
-        if use_default_arg is True:
+        if use_compat:
             return argdict
         else:
             argdict = None
-
     if argdict is None:
         return default
     assert isinstance(argdict, dict)
 
-    key = key.lower()
+    assert isinstance(keys, (list, tuple))
-    args = exe_args = None
+    for key_list in keys:
-    if exe is not None:
+        if isinstance(key_list, compat_str):
-        assert isinstance(exe, compat_str)
+            key_list = (key_list,)
-        exe = exe.lower()
+        arg_list = list(filter(
-        args = argdict.get('%s+%s' % (key, exe))
+            lambda x: x is not None,
-        if args is None:
+            [argdict.get(key.lower()) for key in key_list]))
-            exe_args = argdict.get(exe)
+        if arg_list:
-
+            return [arg for args in arg_list for arg in args]
-    if args is None:
+    return default
-        args = argdict.get(key) if key != exe else None
-    if args is None and exe_args is None:
-        args = argdict.get('default', default) if use_default_arg else default
-
-    args, exe_args = args or [], exe_args or []
-    assert isinstance(args, (list, tuple))
-    assert isinstance(exe_args, (list, tuple))
-    return args + exe_args
 
 
 class ISO639Utils(object):

yt_dlp/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2021.03.01'
+__version__ = '2021.03.07'