Release 2021.07.24

Based on: https://github.com/ytdl-org/youtube-dl/pull/29487 by glenn-slayden

.gitattributes

@@ -1 +1,4 @@
+* text=auto
+
 Makefile* text whitespace=-tab-in-indent
+*.sh text eol=lf

.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 2021.06.08. 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 Issues with outdated version will be REJECTED.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 2021.07.21. 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 Issues with outdated version will be REJECTED.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 **2021.06.08**
+- [ ] I've verified that I'm running yt-dlp version **2021.07.21**
 - [ ] 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
@@ -42,9 +42,9 @@ Provide the complete verbose output of yt-dlp that clearly demonstrates the prob
 Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your command line>`), copy the WHOLE output and insert it below. It should look similar to this:
  [debug] System config: []
  [debug] User config: []
- [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
+ [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKc']
  [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
- [debug] yt-dlp version 2021.06.08
+ [debug] yt-dlp version 2021.07.21
  [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 2021.06.08. 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 2021.07.21. 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 **2021.06.08**
+- [ ] I've verified that I'm running yt-dlp version **2021.07.21**
 - [ ] 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 2021.06.08. 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 2021.07.21. 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 **2021.06.08**
+- [ ] I've verified that I'm running yt-dlp version **2021.07.21**
 - [ ] 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 2021.06.08. 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 2021.07.21. 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 **2021.06.08**
+- [ ] I've verified that I'm running yt-dlp version **2021.07.21**
 - [ ] 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
@@ -44,9 +44,9 @@ Provide the complete verbose output of yt-dlp that clearly demonstrates the prob
 Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your command line>`), copy the WHOLE output and insert it below. It should look similar to this:
  [debug] System config: []
  [debug] User config: []
- [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
+ [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKc']
  [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
- [debug] yt-dlp version 2021.06.08
+ [debug] yt-dlp version 2021.07.21
  [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 2021.06.08. 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 2021.07.21. 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 **2021.06.08**
+- [ ] I've verified that I'm running yt-dlp version **2021.07.21**
 - [ ] I've searched the bugtracker for similar feature requests including closed ones
 
 

.github/ISSUE_TEMPLATE_tmpl/1_broken_site.md

@@ -42,7 +42,7 @@ Provide the complete verbose output of yt-dlp that clearly demonstrates the prob
 Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your command line>`), copy the WHOLE output and insert it below. It should look similar to this:
  [debug] System config: []
  [debug] User config: []
- [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
+ [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKc']
  [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
  [debug] yt-dlp version %(version)s
  [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2

.github/ISSUE_TEMPLATE_tmpl/4_bug_report.md

@@ -44,7 +44,7 @@ Provide the complete verbose output of yt-dlp that clearly demonstrates the prob
 Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your command line>`), copy the WHOLE output and insert it below. It should look similar to this:
  [debug] System config: []
  [debug] User config: []
- [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
+ [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKc']
  [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
  [debug] yt-dlp version %(version)s
  [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2

.github/workflows/build.yml

@@ -95,14 +95,15 @@ jobs:
 
     steps:
     - uses: actions/checkout@v2
-    - name: Set up Python
+    # 3.8 is used for Win7 support
+    - name: Set up Python 3.8
       uses: actions/setup-python@v2
       with:
           python-version: '3.8'
     - name: Upgrade pip and enable wheel support
       run: python -m pip install --upgrade pip setuptools wheel
     - name: Install Requirements
-      run: pip install pyinstaller mutagen pycryptodome
+      run: pip install pyinstaller mutagen pycryptodome websockets
     - name: Bump version
       id: bump_version
       run: python devscripts/update-version.py
@@ -137,15 +138,16 @@ jobs:
 
     steps:
     - uses: actions/checkout@v2
-    - name: Set up Python 3.4.4 32-Bit
+    # 3.7 is used for Vista support. See https://github.com/yt-dlp/yt-dlp/issues/390
+    - name: Set up Python 3.7 32-Bit
       uses: actions/setup-python@v2
       with:
-          python-version: '3.4.4'
+          python-version: '3.7'
           architecture: 'x86'
     - name: Upgrade pip and enable wheel support
-      run: python -m pip install pip==19.1.1 setuptools==43.0.0 wheel==0.33.6
-    - name: Install Requirements for 32 Bit
-      run: pip install pyinstaller==3.5 mutagen==1.42.0 pycryptodome==3.9.4 pefile==2019.4.18
+      run: python -m pip install --upgrade pip setuptools wheel
+    - name: Install Requirements
+      run: pip install pyinstaller mutagen pycryptodome websockets
     - name: Bump version
       id: bump_version
       run: python devscripts/update-version.py

.github/workflows/core.yml

@@ -9,11 +9,13 @@ jobs:
       fail-fast: false
       matrix:
         os: [ubuntu-18.04]
-        python-version: [3.6, 3.7, 3.8, 3.9, pypy-3.6, pypy-3.7]
+        # py3.9 is in quick-test
+        python-version: [3.7, 3.8, 3.10-dev, pypy-3.6, pypy-3.7]
         run-tests-ext: [sh]
         include:
+        # atleast one of the tests must be in windows
         - os: windows-latest
-          python-version: 3.4  # Windows x86 build is still in 3.4
+          python-version: 3.6
           run-tests-ext: bat
     steps:
     - uses: actions/checkout@v2
@@ -21,11 +23,9 @@ jobs:
       uses: actions/setup-python@v2
       with:
         python-version: ${{ matrix.python-version }}
-    - name: Install nose
-      run: pip install nose
+    - name: Install pytest
+      run: pip install pytest
     - name: Run tests
       continue-on-error: False
-      env:
-        YTDL_TEST_SET: core
-      run: ./devscripts/run_tests.${{ matrix.run-tests-ext }}
+      run: ./devscripts/run_tests.${{ matrix.run-tests-ext }} core
   # Linter is in quick-test

.github/workflows/download.yml

@@ -9,11 +9,11 @@ jobs:
       fail-fast: true
       matrix:
         os: [ubuntu-18.04]
-        python-version: [3.6, 3.7, 3.8, 3.9, pypy-3.6, pypy-3.7]
+        python-version: [3.7, 3.8, 3.9, 3.10-dev, pypy-3.6, pypy-3.7]
         run-tests-ext: [sh]
         include:
         - os: windows-latest
-          python-version: 3.4  # Windows x86 build is still in 3.4
+          python-version: 3.6
           run-tests-ext: bat
     steps:
     - uses: actions/checkout@v2
@@ -21,10 +21,8 @@ jobs:
       uses: actions/setup-python@v2
       with:
         python-version: ${{ matrix.python-version }}
-    - name: Install nose
-      run: pip install nose
+    - name: Install pytest
+      run: pip install pytest
     - name: Run tests
       continue-on-error: true
-      env:
-        YTDL_TEST_SET: download
-      run: ./devscripts/run_tests.${{ matrix.run-tests-ext }}
+      run: ./devscripts/run_tests.${{ matrix.run-tests-ext }} download

.github/workflows/quick-test.yml

@@ -11,12 +11,10 @@ jobs:
       uses: actions/setup-python@v2
       with:
         python-version: 3.9
-    - name: Install nose
-      run: pip install nose
+    - name: Install test requirements
+      run: pip install pytest pycryptodome
     - name: Run tests
-      env:
-        YTDL_TEST_SET: core
-      run: ./devscripts/run_tests.sh
+      run: ./devscripts/run_tests.sh core
   flake8:
     name: Linter
     if: "!contains(github.event.head_commit.message, 'ci skip all')"
@@ -30,4 +28,4 @@ jobs:
     - name: Install flake8
       run: pip install flake8
     - name: Run flake8
-      run: flake8 .
+      run: flake8 .

.gitignore

@@ -33,6 +33,7 @@ cookies.txt
 *.info.json
 *.live_chat.json
 *.jpg
+*.jpeg
 *.png
 *.webp
 *.annotations.xml
@@ -44,6 +45,7 @@ cookies.txt
 # Python
 *.pyc
 *.pyo
+.pytest_cache
 wine-py2exe/
 py2exe.log
 build/
@@ -78,6 +80,7 @@ README.txt
 *.tar.gz
 *.zsh
 *.spec
+test/testdata/player-*.js
 
 # Binary
 /youtube-dl

CONTRIBUTING.md

@@ -3,7 +3,7 @@
 $ youtube-dl -v <your command line>
 [debug] System config: []
 [debug] User config: []
-[debug] Command-line args: [u'-v', u'https://www.youtube.com/watch?v=BaW_jenozKcj']
+[debug] Command-line args: [u'-v', u'https://www.youtube.com/watch?v=BaW_jenozKc']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
 [debug] youtube-dl version 2015.12.06
 [debug] Git HEAD: 135392e
@@ -81,16 +81,17 @@ To run the test, simply invoke your favorite test runner, or execute a test file
     python -m unittest discover
     python test/test_download.py
     nosetests
+    pytest
 
 See item 6 of [new extractor tutorial](#adding-support-for-a-new-site) for how to run extractor specific test cases.
 
 If you want to create a build of youtube-dl yourself, you'll need
 
-* python
+* python3
 * make (only GNU make is supported)
 * pandoc
 * zip
-* nosetests
+* pytest
 
 ### Adding support for a new site
 

CONTRIBUTORS

@@ -52,5 +52,14 @@ hhirtz
 louie-github
 MinePlayersPE
 olifre
-rhsmachine
+rhsmachine/zenerdi0de
 nihil-admirari
+krichbanana
+ohmybahgosh
+nyuszika7h
+blackjack4494
+pyx
+TpmKranz
+mzbaulhaque
+zackmark29
+mbway

Changelog.md

@@ -19,6 +19,169 @@
 -->
 
 
+### 2021.07.24
+
+* [youtube:tab] Extract video duration early
+* [downloader] Pass `info_dict` to `progress_hook`s
+* [youtube] Fix age-gated videos for API clients when cookies are supplied by [colethedj](https://github.com/colethedj)
+* [youtube] Disable `get_video_info` age-gate workaround - This endpoint seems to be completely dead
+* [youtube] Try all clients even if age-gated
+* [youtube] Fix subtitles only being extracted from the first client
+* [youtube] Simplify `_get_text`
+* [cookies] bugfix for microsoft edge on macOS
+* [cookies] Handle `sqlite` `ImportError` gracefully by [mbway](https://github.com/mbway)
+* [cookies] Handle errors when importing `keyring`
+
+### 2021.07.21
+
+* **Add option `--cookies-from-browser`** to load cookies from a browser by [mbway](https://github.com/mbway)
+    * Usage: `--cookies-from-browser BROWSER[:PROFILE_NAME_OR_PATH]`
+    * Also added `--no-cookies-from-browser`
+    * To decrypt chromium cookies, `keyring` is needed for UNIX and `pycryptodome` for Windows
+* Add option `--exec-before-download`
+* Add field `live_status`
+* [FFmpegMetadata] Add language of each stream and some refactoring
+* [douyin] Add extractor by [pukkandan](https://github.com/pukkandan), [pyx](https://github.com/pyx)
+* [pornflip] Add extractor by [mzbaulhaque](https://github.com/mzbaulhaque)
+* **[youtube] Extract data from multiple clients** by [pukkandan](https://github.com/pukkandan), [colethedj](https://github.com/colethedj)
+    * `player_client` now accepts multiple clients
+    * Default `player_client` = `android,web`
+        * This uses twice as many requests, but avoids throttling for most videos while also not losing any formats
+    * Music clients can be specifically requested and is enabled by default if `music.youtube.com`
+    * Added `player_client=ios` (Known issue: formats from ios are not sorted correctly)
+    * Add age-gate bypass for android and ios clients
+* [youtube] Extract more thumbnails
+    * The thumbnail URLs are hard-coded and their actual existence is tested lazily
+    * Added option `--no-check-formats` to not test them
+* [youtube] Misc fixes
+    * Improve extraction of livestream metadata by [pukkandan](https://github.com/pukkandan), [krichbanana](https://github.com/krichbanana)
+    * Hide live dash formats since they can't be downloaded anyway
+    * Fix authentication when using multiple accounts by [colethedj](https://github.com/colethedj)
+    * Fix controversial videos when requested via API by [colethedj](https://github.com/colethedj)
+    * Fix session index extraction and headers for non-web player clients by [colethedj](https://github.com/colethedj)
+    * Make `--extractor-retries` work for more errors
+    * Fix sorting of 3gp format
+    * Sanity check `chapters` (and refactor related code)
+    * Make `parse_time_text` and `_extract_chapters` non-fatal
+    * Misc cleanup and bug fixes by [colethedj](https://github.com/colethedj)
+* [youtube:tab] Fix channels tab
+* [youtube:tab] Extract playlist availability by [colethedj](https://github.com/colethedj)
+* **[youtube:comments] Move comment extraction to new API** by [colethedj](https://github.com/colethedj)
+    * Adds extractor-args `comment_sort` (`top`/`new`), `max_comments`, `max_comment_depth`
+* [youtube:comments] Fix `is_favorited`, improve `like_count` parsing by [colethedj](https://github.com/colethedj)
+* [BravoTV] Improve metadata extraction by [kevinoconnor7](https://github.com/kevinoconnor7)
+* [crunchyroll:playlist] Force http
+* [yahoo:gyao:player] Relax `_VALID_URL` by [nao20010128nao](https://github.com/nao20010128nao)
+* [nebula] Authentication via tokens from cookie jar by [hheimbuerger](https://github.com/hheimbuerger), [TpmKranz](https://github.com/TpmKranz)
+* [RTP] Fix extraction and add subtitles by [fstirlitz](https://github.com/fstirlitz)
+* [viki] Rewrite extractors and add extractor-arg `video_types` to `vikichannel` by [zackmark29](https://github.com/zackmark29), [pukkandan](https://github.com/pukkandan)
+* [vlive] Extract thumbnail directly in addition to the one from Naver
+* [generic] Extract previously missed subtitles by [fstirlitz](https://github.com/fstirlitz)
+* [generic] Extract everything in the SMIL manifest and detect discarded subtitles by [fstirlitz](https://github.com/fstirlitz)
+* [embedthumbnail] Fix `_get_thumbnail_resolution`
+* [metadatafromfield] Do not detect numbers as field names
+* Fix selectors `all`, `mergeall` and add tests
+* Errors in playlist extraction should obey `--ignore-errors`
+* Fix bug where `original_url` was not propagated when `_type`=`url`
+* Revert "Merge webm formats into mkv if thumbnails are to be embedded (#173)"
+    * This was wrongly checking for `write_thumbnail`
+* Improve `extractor_args` parsing
+* Rename `NOTE` in `-F` to `MORE INFO` since it's often confused to be the same as `format_note`
+* Add `only_once` param for `write_debug` and `report_warning`
+* [extractor] Allow extracting multiple groups in `_search_regex` by [fstirlitz](https://github.com/fstirlitz)
+* [utils] Improve `traverse_obj`
+* [utils] Add `variadic`
+* [utils] Improve `js_to_json` comment regex by [fstirlitz](https://github.com/fstirlitz)
+* [webtt] Fix timestamps
+* [compat] Remove unnecessary code
+* [doc] fix default of multistreams
+
+
+### 2021.07.07
+
+* Merge youtube-dl: Upto [commit/a803582](https://github.com/ytdl-org/youtube-dl/commit/a8035827177d6b59aca03bd717acb6a9bdd75ada)
+* Add `--extractor-args` to pass some extractor-specific arguments. See [readme](https://github.com/yt-dlp/yt-dlp#extractor-arguments)
+    * Add extractor option `skip` for `youtube`. Eg: `--extractor-args youtube:skip=hls,dash`
+    * Deprecates `--youtube-skip-dash-manifest`, `--youtube-skip-hls-manifest`, `--youtube-include-dash-manifest`, `--youtube-include-hls-manifest`
+* Allow `--list...` options to work with `--print`, `--quiet` and other `--list...` options
+* [youtube] Use `player` API for additional video extraction requests by [colethedj](https://github.com/colethedj)
+    * **Fixes youtube premium music** (format 141) extraction
+    * Adds extractor option `player_client` = `web`/`android`
+        * **`--extractor-args youtube:player_client=android` works around the throttling** for the time-being
+    * Adds extractor option `player_skip=config`
+    * Adds age-gate fallback using embedded client
+* [youtube] Choose correct Live chat API for upcoming streams by [krichbanana](https://github.com/krichbanana)
+* [youtube] Fix subtitle names for age-gated videos
+* [youtube:comments] Fix error handling and add `itct` to params by [colethedj](https://github.com/colethedj)
+* [youtube_live_chat] Fix download with cookies by [siikamiika](https://github.com/siikamiika)
+* [youtube_live_chat] use `clickTrackingParams` by [siikamiika](https://github.com/siikamiika)
+* [Funimation] Rewrite extractor
+    * Add `FunimationShowIE` by [Mevious](https://github.com/Mevious)
+    * **Treat the different versions of an episode as different formats of a single video**
+        * This changes the video `id` and will break break existing archives
+        * Compat option `seperate-video-versions` to fall back to old behavior including using the old video ids
+    * Support direct `/player/` URL
+    * Extractor options `language` and `version` to pre-select them during extraction
+        * These options may be removed in the future if we can extract all formats without additional network requests
+        * Do not rely on these for format selection and use `-f` filters instead
+* [AdobePass] Add Spectrum MSO by [kevinoconnor7](https://github.com/kevinoconnor7), [ohmybahgosh](https://github.com/ohmybahgosh)
+* [facebook] Extract description and fix title
+* [fancode] Fix extraction, support live and allow login with refresh token by [zenerdi0de](https://github.com/zenerdi0de)
+* [plutotv] Improve `_VALID_URL`
+* [RCTIPlus] Add extractor by [MinePlayersPE](https://github.com/MinePlayersPE)
+* [Soundcloud] Allow login using oauth token by [blackjack4494](https://github.com/blackjack4494)
+* [TBS] Support livestreams by [llacb47](https://github.com/llacb47)
+* [videa] Fix extraction by [nyuszika7h](https://github.com/nyuszika7h)
+* [yahoo] Fix extraction by [llacb47](https://github.com/llacb47), [pukkandan](https://github.com/pukkandan)
+* Process videos when using `--ignore-no-formats-error` by [krichbanana](https://github.com/krichbanana)
+* Fix `--throttled-rate` when using `--load-info-json`
+* Fix `--flat-playlist` when entry has no `ie_key`
+* Fix `check_formats` catching `ExtractorError` instead of `DownloadError`
+* Fix deprecated option `--list-formats-old`
+* [downloader/ffmpeg] Fix `--ppa` when using simultaneous download
+* [extractor] Prevent unnecessary download of hls manifests and refactor `hls_split_discontinuity`
+* [fragment] Handle status of download and errors in threads correctly; and minor refactoring
+* [thumbnailsconvertor] Treat `jpeg` as `jpg`
+* [utils] Fix issues with `LazyList` reversal
+* [extractor] Allow extractors to set their own login hint
+* [cleanup] Simplify format selector code with `LazyList` and `yield from`
+* [cleanup] Clean `extractor.common._merge_subtitles` signature
+* [cleanup] Fix some typos
+
+
+### 2021.06.23
+
+* Merge youtube-dl: Upto [commit/379f52a](https://github.com/ytdl-org/youtube-dl/commit/379f52a4954013767219d25099cce9e0f9401961)
+* **Add option `--throttled-rate`** below which video data is re-extracted
+* [fragment] **Merge during download for `-N`**, and refactor `hls`/`dash`
+* [websockets] Add `WebSocketFragmentFD` by [nao20010128nao](https://github.com/nao20010128nao), [pukkandan](https://github.com/pukkandan)
+* Allow `images` formats in addition to video/audio
+* [downloader/mhtml] Add new downloader for slideshows/storyboards by [fstirlitz](https://github.com/fstirlitz)
+* [youtube] Temporary **fix for age-gate**
+* [youtube] Support ongoing live chat by [siikamiika](https://github.com/siikamiika)
+* [youtube] Improve SAPISID cookie handling by [colethedj](https://github.com/colethedj)
+* [youtube] Login is not needed for `:ytrec`
+* [youtube] Non-fatal alert reporting for unavailable videos page by [colethedj](https://github.com/colethedj)
+* [twitcasting] Websocket support by [nao20010128nao](https://github.com/nao20010128nao)
+* [mediasite] Extract slides by [fstirlitz](https://github.com/fstirlitz)
+* [funimation] Extract subtitles 
+* [pornhub] Extract `cast`
+* [hotstar] Use server time for authentication instead of local time
+* [EmbedThumbnail] Fix for already downloaded thumbnail
+* [EmbedThumbnail] Add compat-option `embed-thumbnail-atomicparsley`
+* Expand `--check-formats` to thumbnails
+* Fix id sanitization in filenames
+* Skip fixup of existing files and add `--fixup force` to force it
+* Better error handling of syntax errors in `-f`
+* Use `NamedTemporaryFile` for `--check-formats`
+* [aria2c] Lower `--min-split-size` for HTTP downloads
+* [options] Rename `--add-metadata` to `--embed-metadata`
+* [utils] Improve `LazyList` and add tests
+* [build] Build Windows x86 version with py3.7 and remove redundant tests by [pukkandan](https://github.com/pukkandan), [shirt](https://github.com/shirt-dev)
+* [docs] Clarify that `--embed-metadata` embeds chapter markers
+* [cleanup] Refactor fixup
+
+
 ### 2021.06.09
 
 * Fix bug where `%(field)d` in filename template throws error
@@ -34,7 +197,7 @@
 * [extractor] Fix FourCC fallback when parsing ISM by [fstirlitz](https://github.com/fstirlitz)
 * [twitcasting] Add TwitCastingUserIE, TwitCastingLiveIE by [pukkandan](https://github.com/pukkandan), [nao20010128nao](https://github.com/nao20010128nao)
 * [vidio] Add VidioPremierIE and VidioLiveIE by [MinePlayersPE](Https://github.com/MinePlayersPE)
-* [viki] Fix extraction from by [ytdl-org/youtube-dl@59e583f](https://github.com/ytdl-org/youtube-dl/commit/59e583f7e8530ca92776c866897d895c072e2a82)
+* [viki] Fix extraction from [ytdl-org/youtube-dl@59e583f](https://github.com/ytdl-org/youtube-dl/commit/59e583f7e8530ca92776c866897d895c072e2a82)
 * [youtube] Support shorts URL
 * [zoom] Extract transcripts as subtitles
 * Add field `original_url` with the user-inputted URL

Makefile

@@ -13,7 +13,7 @@ pypi-files: AUTHORS Changelog.md LICENSE README.md README.txt supportedsites com
 .PHONY: all clean install test tar pypi-files completions ot offlinetest codetest supportedsites
 
 clean-test:
-	rm -rf *.dump *.part* *.ytdl *.info.json *.mp4 *.m4a *.flv *.mp3 *.avi *.mkv *.webm *.3gp *.wav *.ape *.swf *.jpg *.png *.frag *.frag.urls *.frag.aria2
+	rm -rf *.dump *.part* *.ytdl *.info.json *.mp4 *.m4a *.flv *.mp3 *.avi *.mkv *.webm *.3gp *.wav *.ape *.swf *.jpg *.png *.frag *.frag.urls *.frag.aria2 test/testdata/player-*.js
 clean-dist:
 	rm -rf yt-dlp.1.temp.md yt-dlp.1 README.txt MANIFEST build/ dist/ .coverage cover/ yt-dlp.tar.gz completions/ yt_dlp/extractor/lazy_extractors.py *.spec CONTRIBUTING.md.tmp yt-dlp yt-dlp.exe yt_dlp.egg-info/ AUTHORS .mailmap
 clean-cache:
@@ -49,23 +49,11 @@ codetest:
 	flake8 .
 
 test:
-	#nosetests --with-coverage --cover-package=yt_dlp --cover-html --verbose --processes 4 test
-	nosetests --verbose test
+	$(PYTHON) -m pytest
 	$(MAKE) codetest
 
-# Keep this list in sync with devscripts/run_tests.sh
 offlinetest: codetest
-	$(PYTHON) -m nose --verbose test \
-		--exclude test_age_restriction.py \
-		--exclude test_download.py \
-		--exclude test_iqiyi_sdk_interpreter.py \
-		--exclude test_overwrites.py \
-		--exclude test_socks.py \
-		--exclude test_subtitles.py \
-		--exclude test_write_annotations.py \
-		--exclude test_youtube_lists.py \
-		--exclude test_youtube_signature.py \
-		--exclude test_post_hooks.py
+	$(PYTHON) -m pytest -k "not download"
 
 yt-dlp: yt_dlp/*.py yt_dlp/*/*.py
 	mkdir -p zip

README.md

@@ -53,6 +53,7 @@ yt-dlp is a [youtube-dl](https://github.com/ytdl-org/youtube-dl) fork based on t
     * [Format Selection examples](#format-selection-examples)
 * [MODIFYING METADATA](#modifying-metadata)
     * [Modifying metadata examples](#modifying-metadata-examples)
+* [EXTRACTOR ARGUMENTS](#extractor-arguments)
 * [PLUGINS](#plugins)
 * [DEPRECATED OPTIONS](#deprecated-options)
 * [MORE](#more)
@@ -66,7 +67,7 @@ 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 [commit/c2350ca](https://github.com/ytdl-org/youtube-dl/commit/c2350cac243ba1ec1586fe85b0d62d1b700047a2)**: (v2021.06.06) 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 [commit/379f52a](https://github.com/ytdl-org/youtube-dl/commit/379f52a4954013767219d25099cce9e0f9401961)**: (v2021.06.06) 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 `--write-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.
 
@@ -74,19 +75,22 @@ The major new features from the latest release of [blackjack4494/yt-dlc](https:/
     * All Feeds (`:ytfav`, `:ytwatchlater`, `:ytsubs`, `:ythistory`, `:ytrec`) supports downloading multiple pages of content
     * Search (`ytsearch:`, `ytsearchdate:`), search URLs and in-channel search works
     * Mixes supports downloading multiple pages of content
+    * Partial workaround for throttling issue
     * Redirect channel's home URL automatically to `/video` to preserve the old behaviour
     * `255kbps` audio is extracted from youtube music if premium cookies are given
     * Youtube music Albums, channels etc can be downloaded
 
+* **Cookies from browser**: Cookies can be automatically extracted from all major web browsers using `--cookies-from-browser BROWSER[:PROFILE]`
+
 * **Split video by chapters**: Videos can be split into multiple files based on chapters using `--split-chapters`
 
 * **Multi-threaded fragment downloads**: Download multiple fragments of m3u8/mpd videos in parallel. 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
 
-* **New extractors**: AnimeLab, Philo MSO, Rcs, Gedi, bitwave.tv, mildom, audius, zee5, mtv.it, wimtv, pluto.tv, niconico users, discoveryplus.in, mediathek, NFHSNetwork, nebula, ukcolumn, whowatch, MxplayerShow, parlview (au), YoutubeWebArchive, fancode, Saitosan, ShemarooMe, telemundo, VootSeries, SonyLIVSeries, HotstarSeries, VidioPremier, VidioLive
+* **New extractors**: AnimeLab, Philo MSO, Spectrum MSO, Rcs, Gedi, bitwave.tv, mildom, audius, zee5, mtv.it, wimtv, pluto.tv, niconico users, discoveryplus.in, mediathek, NFHSNetwork, nebula, ukcolumn, whowatch, MxplayerShow, parlview (au), YoutubeWebArchive, fancode, Saitosan, ShemarooMe, telemundo, VootSeries, SonyLIVSeries, HotstarSeries, VidioPremier, VidioLive, RCTIPlus, TBS Live, douyin, pornflip
 
-* **Fixed extractors**: archive.org, roosterteeth.com, skyit, instagram, itv, SouthparkDe, spreaker, Vlive, akamai, ina, rumble, tennistv, amcnetworks, la7 podcasts, linuxacadamy, nitter, twitcasting, viu, crackle, curiositystream, mediasite, rmcdecouverte, sonyliv, tubi, tenplay, patreon
+* **Fixed extractors**: archive.org, roosterteeth.com, skyit, instagram, itv, SouthparkDe, spreaker, Vlive, akamai, ina, rumble, tennistv, amcnetworks, la7 podcasts, linuxacadamy, nitter, twitcasting, viu, crackle, curiositystream, mediasite, rmcdecouverte, sonyliv, tubi, tenplay, patreon, videa, yahoo, BravoTV, crunchyroll playlist, RTP, viki
 
 * **Subtitle extraction from manifests**: Subtitles can be extracted from streaming media manifests. See [commit/be6202f](https://github.com/yt-dlp/yt-dlp/commit/be6202f12b97858b9d716e608394b51065d0419f) for details
 
@@ -127,10 +131,12 @@ Some of yt-dlp's default options are different from that of youtube-dl and youtu
 * `--add-metadata` attaches the `infojson` to `mkv` files in addition to writing the metadata when used with `--write-infojson`. Use `--compat-options no-attach-info-json` to revert this
 * `playlist_index` behaves differently when used with options like `--playlist-reverse` and `--playlist-items`. See [#302](https://github.com/yt-dlp/yt-dlp/issues/302) for details. You can use `--compat-options playlist-index` if you want to keep the earlier behavior
 * The output of `-F` is listed in a new format. Use `--compat-options list-formats` to revert this
+* All *experiences* of a funimation episode are considered as a single video. This behavior breaks existing archives. Use `--compat-options seperate-video-versions` to extract information from only the default player
 * Youtube live chat (if available) is considered as a subtitle. Use `--sub-langs all,-live_chat` to download all subtitles except live chat. You can also use `--compat-options no-live-chat` to prevent live chat from downloading
 * Youtube channel URLs are automatically redirected to `/video`. Append a `/featured` to the URL to download only the videos in the home page. If the channel does not have a videos tab, we try to download the equivalent `UU` playlist instead. Also, `/live` URLs raise an error if there are no live videos instead of silently downloading the entire channel. You may use `--compat-options no-youtube-channel-redirect` to revert all these redirections
 * Unavailable videos are also listed for youtube playlists. Use `--compat-options no-youtube-unavailable-videos` to remove this
 * If `ffmpeg` is used as the downloader, the downloading and merging of formats happen in a single step when possible. Use `--compat-options no-direct-merge` to revert this
+* Thumbnail embedding in `mp4` is done with mutagen if possible. Use `--compat-options embed-thumbnail-atomicparsley` to force the use of AtomicParsley instead
 
 For ease of use, a few more compat options are available:
 * `--compat-options all`: Use all compat options
@@ -181,6 +187,8 @@ While all the other dependancies are optional, `ffmpeg` and `ffprobe` are highly
 * [**sponskrub**](https://github.com/faissaloo/SponSkrub) - For using the [sponskrub options](#sponskrub-sponsorblock-options). Licenced under [GPLv3+](https://github.com/faissaloo/SponSkrub/blob/master/LICENCE.md)
 * [**mutagen**](https://github.com/quodlibet/mutagen) - For embedding thumbnail in certain formats. Licenced under [GPLv2+](https://github.com/quodlibet/mutagen/blob/master/COPYING)
 * [**pycryptodome**](https://github.com/Legrandin/pycryptodome) - For decrypting various data. Licenced under [BSD2](https://github.com/Legrandin/pycryptodome/blob/master/LICENSE.rst)
+* [**websockets**](https://github.com/aaugustin/websockets) - For downloading over websocket. Licenced under [BSD3](https://github.com/aaugustin/websockets/blob/main/LICENSE)
+* [**keyring**](https://github.com/jaraco/keyring) - For decrypting cookies of chromium-based browsers on Linux. Licenced under [MIT](https://github.com/jaraco/keyring/blob/main/LICENSE)
 * [**AtomicParsley**](https://github.com/wez/atomicparsley) - For embedding thumbnail in mp4/m4a if mutagen is not present. Licenced under [GPLv2+](https://github.com/wez/atomicparsley/blob/master/COPYING)
 * [**rtmpdump**](http://rtmpdump.mplayerhq.hu) - For downloading `rtmp` streams. ffmpeg will be used as a fallback. Licenced under [GPLv2+](http://rtmpdump.mplayerhq.hu)
 * [**mplayer**](http://mplayerhq.hu/design7/info.html) or [**mpv**](https://mpv.io) - For downloading `rstp` streams. ffmpeg will be used as a fallback. Licenced under [GPLv2+](https://github.com/mpv-player/mpv/blob/master/Copyright)
@@ -189,14 +197,14 @@ While all the other dependancies are optional, `ffmpeg` and `ffprobe` are highly
 
 To use or redistribute the dependencies, you must agree to their respective licensing terms.
 
-Note that the windows releases are already built with the python interpreter, mutagen and pycryptodome included.
+Note that the windows releases are already built with the python interpreter, mutagen, pycryptodome and websockets included.
 
 ### COMPILE
 
 **For Windows**:
-To build the Windows executable, you must have pyinstaller (and optionally mutagen and pycryptodome)
+To build the Windows executable, you must have pyinstaller (and optionally mutagen, pycryptodome, websockets)
 
-    python3 -m pip install --upgrade pyinstaller mutagen pycryptodome
+    python3 -m pip install --upgrade pyinstaller mutagen pycryptodome websockets
 
 Once you have all the necessary dependencies installed, just run `py pyinst.py`. The executable will be built for the same architecture (32/64 bit) as the python used to build it.
 
@@ -207,7 +215,7 @@ You can also build the executable without any version info or metadata by using:
 Note that pyinstaller [does not support](https://github.com/pyinstaller/pyinstaller#requirements-and-tested-platforms) Python installed from the Windows store without using a virtual environment
 
 **For Unix**:
-You will need the required build tools: `python`, `make` (GNU), `pandoc`, `zip`, `nosetests`  
+You will need the required build tools: `python`, `make` (GNU), `pandoc`, `zip`, `pytest`  
 Then simply run `make`. You can also run `make yt-dlp` instead to compile only the binary without updating any of the additional files
 
 **Note**: In either platform, `devscripts\update-version.py` can be used to automatically update the version number
@@ -372,6 +380,9 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
                                      (default is 1)
     -r, --limit-rate RATE            Maximum download rate in bytes per second
                                      (e.g. 50K or 4.2M)
+    --throttled-rate RATE            Minimum download rate in bytes per second
+                                     below which throttling is assumed and the
+                                     video data is re-extracted (e.g. 100K)
     -R, --retries RETRIES            Number of retries (default is 10), or
                                      "infinite"
     --fragment-retries RETRIES       Number of retries for a fragment (default
@@ -428,7 +439,8 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
     --downloader-args NAME:ARGS      Give these arguments to the external
                                      downloader. Specify the downloader name and
                                      the arguments separated by a colon ":". You
-                                     can use this option multiple times
+                                     can use this option multiple times to give
+                                     different arguments to different downloaders
                                      (Alias: --external-downloader-args)
 
 ## Filesystem Options:
@@ -512,7 +524,19 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
                                      option)
     --cookies FILE                   File to read cookies from and dump cookie
                                      jar in
-    --no-cookies                     Do not read/dump cookies (default)
+    --no-cookies                     Do not read/dump cookies from/to file
+                                     (default)
+    --cookies-from-browser BROWSER[:PROFILE]
+                                     Load cookies from a user profile of the
+                                     given web browser. Currently supported
+                                     browsers are: brave|chrome|chromium|edge|fi
+                                     refox|opera|safari|vivaldi. You can specify
+                                     the user profile name or directory using
+                                     "BROWSER:PROFILE_NAME" or
+                                     "BROWSER:PROFILE_PATH". If no profile is
+                                     given, the most recently accessed one is
+                                     used
+    --no-cookies-from-browser        Do not load cookies from browser (default)
     --cache-dir DIR                  Location in the filesystem where youtube-dl
                                      can store some downloaded information (such
                                      as client ids and signatures) permanently.
@@ -630,7 +654,9 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
     --no-prefer-free-formats         Don't give any special preference to free
                                      containers (default)
     --check-formats                  Check that the formats selected are
-                                     actually downloadable (Experimental)
+                                     actually downloadable
+    --no-check-formats               Do not check that the formats selected are
+                                     actually downloadable
     -F, --list-formats               List all available formats of requested
                                      videos
     --merge-output-format FORMAT     If a merge is required (e.g.
@@ -710,7 +736,8 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
                                      Metadata, EmbedSubtitle, EmbedThumbnail,
                                      SubtitlesConvertor, ThumbnailsConvertor,
                                      VideoRemuxer, VideoConvertor, SponSkrub,
-                                     FixupStretched, FixupM4a and FixupM3u8. The
+                                     FixupStretched, FixupM4a, FixupM3u8,
+                                     FixupTimestamp and FixupDuration. The
                                      supported executables are: AtomicParsley,
                                      FFmpeg, FFprobe, and SponSkrub. You can
                                      also specify "PP+EXE:ARGS" to give the
@@ -734,10 +761,13 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
     --embed-subs                     Embed subtitles in the video (only for mp4,
                                      webm and mkv videos)
     --no-embed-subs                  Do not embed subtitles (default)
-    --embed-thumbnail                Embed thumbnail in the audio as cover art
+    --embed-thumbnail                Embed thumbnail in the video as cover art
     --no-embed-thumbnail             Do not embed thumbnail (default)
-    --add-metadata                   Write metadata to the video file
-    --no-add-metadata                Do not write metadata (default)
+    --embed-metadata                 Embed metadata including chapter markers
+                                     (if supported by the format) to the video
+                                     file (Alias: --add-metadata)
+    --no-embed-metadata              Do not write metadata (default)
+                                     (Alias: --no-add-metadata)
     --parse-metadata FROM:TO         Parse additional metadata like title/artist
                                      from other fields; see "MODIFYING METADATA"
                                      for details
@@ -747,7 +777,8 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
                                      file. One of never (do nothing), warn (only
                                      emit a warning), detect_or_warn (the
                                      default; fix file if we can, warn
-                                     otherwise)
+                                     otherwise), force (try fixing even if file
+                                     already exists
     --ffmpeg-location PATH           Location of the ffmpeg binary; either the
                                      path to the binary or its containing
                                      directory
@@ -760,6 +791,8 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
                                      downloaded file is also available. If no
                                      fields are passed, "%(filepath)s" is
                                      appended to the end of the command
+    --exec-before-download CMD       Execute a command before the actual
+                                     download. The syntax is the same as --exec
     --convert-subs FORMAT            Convert the subtitles to another format
                                      (currently supported: srt|vtt|ass|lrc)
                                      (Alias: --convert-subtitles)
@@ -806,18 +839,10 @@ Then simply run `make`. You can also run `make yt-dlp` instead to compile only t
     --no-hls-split-discontinuity     Do not split HLS playlists to different
                                      formats at discontinuities such as ad
                                      breaks (default)
-    --youtube-include-dash-manifest  Download the DASH manifests and related
-                                     data on YouTube videos (default)
-                                     (Alias: --no-youtube-skip-dash-manifest)
-    --youtube-skip-dash-manifest     Do not download the DASH manifests and
-                                     related data on YouTube videos
-                                     (Alias: --no-youtube-include-dash-manifest)
-    --youtube-include-hls-manifest   Download the HLS manifests and related data
-                                     on YouTube videos (default)
-                                     (Alias: --no-youtube-skip-hls-manifest)
-    --youtube-skip-hls-manifest      Do not download the HLS manifests and
-                                     related data on YouTube videos
-                                     (Alias: --no-youtube-include-hls-manifest)
+    --extractor-args KEY:ARGS        Pass these arguments to the extractor. See
+                                     "EXTRACTOR ARGUMENTS" for details. You can
+                                     use this option multiple times to give
+                                     arguments for different extractors
 
 # CONFIGURATION
 
@@ -932,6 +957,7 @@ The available fields are:
  - `average_rating` (numeric): Average rating give by users, the scale used depends on the webpage
  - `comment_count` (numeric): Number of comments on the video (For some extractors, comments are only downloaded at the end, and so this field cannot be used)
  - `age_limit` (numeric): Age restriction for the video (years)
+ - `live_status` (string): One of 'is_live', 'was_live', 'upcoming', 'not_live'
  - `is_live` (boolean): Whether this video is a live stream or a fixed-length video
  - `was_live` (boolean): Whether this video was originally a live stream
  - `playable_in_embed` (string): Whether this video is allowed to play in embedded players on other sites
@@ -1011,7 +1037,7 @@ Available only when used in `--print`:
 
 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.
+For example for `-o %(title)s-%(id)s.%(ext)s` and an mp4 video with title `yt-dlp test video` and id `BaW_jenozKc`, this will result in a `yt-dlp test video-BaW_jenozKc.mp4` file created in the current directory.
 
 For numeric sequences you can use numeric related formatting, for example, `%(view_count)05d` will result in a string with view count padded with zeros up to 5 characters, like in `00042`.
 
@@ -1095,7 +1121,7 @@ If you want to download multiple videos and they don't have the same formats ava
 
 If you want to download several formats of the same video use a comma as a separator, e.g. `-f 22,17,18` will download all these three formats, of course if they are available. Or a more sophisticated example combined with the precedence feature: `-f 136/137/mp4/bestvideo,140/m4a/bestaudio`.
 
-You can merge the video and audio of multiple formats into a single file using `-f <format1>+<format2>+...` (requires ffmpeg installed), for example `-f bestvideo+bestaudio` will download the best video-only format, the best audio-only format and mux them together with ffmpeg. If `--no-video-multistreams` is used, all formats with a video stream except the first one are ignored. Similarly, if `--no-audio-multistreams` is used, all formats with an audio stream except the first one are ignored. For example, `-f bestvideo+best+bestaudio` will download and merge all 3 given formats. The resulting file will have 2 video streams and 2 audio streams. But `-f bestvideo+best+bestaudio --no-video-multistreams` will download and merge only `bestvideo` and `bestaudio`. `best` is ignored since another format containing a video stream (`bestvideo`) has already been selected. The order of the formats is therefore important. `-f best+bestaudio --no-audio-multistreams` will download and merge both formats while `-f bestaudio+best --no-audio-multistreams` will ignore `best` and download only `bestaudio`.
+You can merge the video and audio of multiple formats into a single file using `-f <format1>+<format2>+...` (requires ffmpeg installed), for example `-f bestvideo+bestaudio` will download the best video-only format, the best audio-only format and mux them together with ffmpeg. Unless `--video-multistreams` is used, all formats with a video stream except the first one are ignored. Similarly, unless `--audio-multistreams` is used, all formats with an audio stream except the first one are ignored. For example, `-f bestvideo+best+bestaudio --video-multistreams --audio-multistreams` will download and merge all 3 given formats. The resulting file will have 2 video streams and 2 audio streams. But `-f bestvideo+best+bestaudio --no-video-multistreams` will download and merge only `bestvideo` and `bestaudio`. `best` is ignored since another format containing a video stream (`bestvideo`) has already been selected. The order of the formats is therefore important. `-f best+bestaudio --no-audio-multistreams` will download and merge both formats while `-f bestaudio+best --no-audio-multistreams` will ignore `best` and download only `bestaudio`.
 
 ## Filtering Formats
 
@@ -1140,7 +1166,7 @@ You can change the criteria for being considered the `best` by using `-S` (`--fo
  - `lang`: Language preference as given by the extractor
  - `quality`: The quality of the format as given by the extractor
  - `source`: Preference of the source as given by the extractor
- - `proto`: Protocol used for download (`https`/`ftps` > `http`/`ftp` > `m3u8_native` > `m3u8` > `http_dash_segments` > other > `mms`/`rtsp` > unknown > `f4f`/`f4m`)
+ - `proto`: Protocol used for download (`https`/`ftps` > `http`/`ftp` > `m3u8_native`/`m3u8` > `http_dash_segments`> `websocket_frag` > other > `mms`/`rtsp` > unknown > `f4f`/`f4m`)
  - `vcodec`: Video Codec (`av01` > `vp9.2` > `vp9` > `h265` > `h264` > `vp8` > `h263` > `theora` > other > unknown)
  - `acodec`: Audio Codec (`opus` > `vorbis` > `aac` > `mp4a` > `mp3` > `ac3` > `dts` > other > unknown)
  - `codec`: Equivalent to `vcodec,acodec`
@@ -1321,6 +1347,29 @@ $ yt-dlp --parse-metadata 'description:(?s)(?P<meta_comment>.+)' --add-metadata
 
 ```
 
+# EXTRACTOR ARGUMENTS
+
+Some extractors accept additional arguments which can be passed using `--extractor-args KEY:ARGS`. `ARGS` is a `;` (semicolon) seperated string of `ARG=VAL1,VAL2`. Eg: `--extractor-args "youtube:skip=dash,hls;player_client=android" --extractor-args "funimation:version=uncut"`
+
+The following extractors use this feature:
+* **youtube**
+    * `skip`: `hls` or `dash` (or both) to skip download of the respective manifests
+    * `player_client`: Clients to extract video data from - one or more of `web`, `android`, `ios`, `web_music`, `android_music`, `ios_music`. By default, `android,web` is used. If the URL is from `music.youtube.com`, `android,web,android_music,web_music` is used
+    * `player_skip`: `configs` - skip any requests for client configs and use defaults
+    * `comment_sort`: `top` or `new` (default) - choose comment sorting mode (on YouTube's side).
+    * `max_comments`: maximum amount of comments to download (default all).
+    * `max_comment_depth`: maximum depth for nested comments. YouTube supports depths 1 or 2 (default). 
+
+* **funimation**
+    * `language`: Languages to extract. Eg: `funimation:language=english,japanese`
+    * `version`: The video version to extract - `uncut` or `simulcast`
+
+* **vikiChannel**
+    * `video_types`: Types of videos to download - one or more of `episodes`, `movies`, `clips`, `trailers`
+
+NOTE: These options may be changed/removed in the future without concern for backward compatibility
+
+
 # PLUGINS
 
 Plugins are loaded from `<root-dir>/ytdlp_plugins/<type>/__init__.py`. Currently only `extractor` plugins are supported. Support for `downloader` and `postprocessor` plugins may be added in the future. See [ytdlp_plugins](ytdlp_plugins) for example.
@@ -1352,6 +1401,10 @@ While these options still work, their use is not recommended since there are oth
     --list-formats-old               --compat-options list-formats (Alias: --no-list-formats-as-table)
     --list-formats-as-table          --compat-options -list-formats [Default] (Alias: --no-list-formats-old)
     --sponskrub-args ARGS            --ppa "sponskrub:ARGS"
+    --youtube-skip-dash-manifest     --extractor-args "youtube:skip=dash" (Alias: --no-youtube-include-dash-manifest)
+    --youtube-skip-hls-manifest      --extractor-args "youtube:skip=hls" (Alias: --no-youtube-include-hls-manifest)
+    --youtube-include-dash-manifest  Default (Alias: --no-youtube-skip-dash-manifest)
+    --youtube-include-hls-manifest   Default (Alias: --no-youtube-skip-hls-manifest)
     --test                           Used by developers for testing extractors. Not intended for the end user
     --youtube-print-sig-code         Used for testing youtube signatures
 

devscripts/run_tests.bat

@@ -1,17 +1,16 @@
+@setlocal
 @echo off
+cd /d %~dp0..
 
-rem Keep this list in sync with the `offlinetest` target in Makefile
-set DOWNLOAD_TESTS="age_restriction^|download^|iqiyi_sdk_interpreter^|socks^|subtitles^|write_annotations^|youtube_lists^|youtube_signature^|post_hooks"
-
-if "%YTDL_TEST_SET%" == "core" (
-    set test_set="-I test_("%DOWNLOAD_TESTS%")\.py"
-    set multiprocess_args=""
-) else if "%YTDL_TEST_SET%" == "download" (
-    set test_set="-I test_(?!"%DOWNLOAD_TESTS%").+\.py"
-    set multiprocess_args="--processes=4 --process-timeout=540"
+if ["%~1"]==[""] (
+    set "test_set="
+) else if ["%~1"]==["core"] (
+    set "test_set=-k "not download""
+) else if ["%~1"]==["download"] (
+    set "test_set=-k download"
 ) else (
-    echo YTDL_TEST_SET is not set or invalid
+    echo.Invalid test type "%~1". Use "core" ^| "download"
     exit /b 1
 )
 
-nosetests test --verbose %test_set:"=% %multiprocess_args:"=%
+pytest %test_set%

devscripts/run_tests.sh

@@ -1,22 +1,15 @@
-#!/bin/bash
+#!/bin/sh
 
-# Keep this list in sync with the `offlinetest` target in Makefile
-DOWNLOAD_TESTS="age_restriction|download|iqiyi_sdk_interpreter|overwrites|socks|subtitles|write_annotations|youtube_lists|youtube_signature|post_hooks"
+if [ -z $1 ]; then
+    test_set='test'
+elif [ $1 = 'core' ]; then
+    test_set='not download'
+elif [ $1 = 'download' ]; then
+    test_set='download'
+else
+    echo 'Invalid test type "'$1'". Use "core" | "download"'
+    exit 1
+fi
 
-test_set=""
-multiprocess_args=""
-
-case "$YTDL_TEST_SET" in
-    core)
-        test_set="-I test_($DOWNLOAD_TESTS)\.py"
-    ;;
-    download)
-        test_set="-I test_(?!$DOWNLOAD_TESTS).+\.py"
-        multiprocess_args="--processes=4 --process-timeout=540"
-    ;;
-    *)
-        break
-    ;;
-esac
-
-nosetests test --verbose $test_set $multiprocess_args
+echo python3 -m pytest -k $test_set
+python3 -m pytest -k "$test_set"

pyinst.py

@@ -6,6 +6,7 @@ import sys
 # import os
 import platform
 
+from PyInstaller.utils.hooks import collect_submodules
 from PyInstaller.utils.win32.versioninfo import (
     VarStruct, VarFileInfo, StringStruct, StringTable,
     StringFileInfo, FixedFileInfo, VSVersionInfo, SetVersion,
@@ -66,16 +67,15 @@ VERSION_FILE = VSVersionInfo(
     ]
 )
 
+dependancies = ['Crypto', 'mutagen'] + collect_submodules('websockets')
+excluded_modules = ['test', 'ytdlp_plugins', 'youtube-dl', 'youtube-dlc']
+
 PyInstaller.__main__.run([
     '--name=yt-dlp%s' % _x86,
     '--onefile',
     '--icon=devscripts/cloud.ico',
-    '--exclude-module=youtube_dl',
-    '--exclude-module=youtube_dlc',
-    '--exclude-module=test',
-    '--exclude-module=ytdlp_plugins',
-    '--hidden-import=mutagen',
-    '--hidden-import=Crypto',
+    *[f'--exclude-module={module}' for module in excluded_modules],
+    *[f'--hidden-import={module}' for module in dependancies],
     '--upx-exclude=vcruntime140.dll',
     'yt_dlp/__main__.py',
 ])

pytest.ini

@@ -0,0 +1,4 @@
+[pytest]
+addopts = -ra -v --strict-markers
+markers =
+    download

requirements.txt

@@ -1,2 +1,3 @@
 mutagen
 pycryptodome
+websockets

setup.py

@@ -19,7 +19,7 @@ LONG_DESCRIPTION = '\n\n'.join((
     '**PS**: Some links in this document will not work since this is a copy of the README.md from Github',
     open('README.md', 'r', encoding='utf-8').read()))
 
-REQUIREMENTS = ['mutagen', 'pycryptodome']
+REQUIREMENTS = ['mutagen', 'pycryptodome', 'websockets']
 
 if sys.argv[1:2] == ['py2exe']:
     raise NotImplementedError('py2exe is not currently supported; instead, use "pyinst.py" to build with pyinstaller')

supportedsites.md

@@ -225,8 +225,7 @@
  - **Culturebox**
  - **CultureUnplugged**
  - **curiositystream**
- - **curiositystream:collections**
- - **curiositystream:series**
+ - **curiositystream:collection**
  - **CWTV**
  - **DagelijkseKost**: dagelijksekost.een.be
  - **DailyMail**
@@ -260,6 +259,7 @@
  - **dlive:vod**
  - **DoodStream**
  - **Dotsub**
+ - **Douyin**
  - **DouyuShow**
  - **DouyuTV**: 斗鱼
  - **DPlay**
@@ -307,6 +307,7 @@
  - **EyedoTV**
  - **facebook**
  - **FacebookPluginsVideo**
+ - **fancode:live**
  - **fancode:vod**
  - **faz.net**
  - **fc2**
@@ -344,6 +345,8 @@
  - **FrontendMastersLesson**
  - **FujiTVFODPlus7**
  - **Funimation**
+ - **funimation:page**
+ - **funimation:show**
  - **Funk**
  - **Fusion**
  - **Fux**
@@ -497,8 +500,6 @@
  - **LinuxAcademy**
  - **LiTV**
  - **LiveJournal**
- - **LiveLeak**
- - **LiveLeakEmbed**
  - **livestream**
  - **livestream:original**
  - **LnkGo**
@@ -769,6 +770,7 @@
  - **PopcornTV**
  - **PornCom**
  - **PornerBros**
+ - **PornFlip**
  - **PornHd**
  - **PornHub**: PornHub and Thumbzilla
  - **PornHubPagedVideoList**
@@ -811,6 +813,8 @@
  - **RCS**
  - **RCSEmbeds**
  - **RCSVarious**
+ - **RCTIPlus**
+ - **RCTIPlusSeries**
  - **RDS**: RDS.ca
  - **RedBull**
  - **RedBullEmbed**

test/helper.py

@@ -22,6 +22,14 @@ from yt_dlp.utils import (
 )
 
 
+if "pytest" in sys.modules:
+    import pytest
+    is_download_test = pytest.mark.download
+else:
+    def is_download_test(testClass):
+        return testClass
+
+
 def get_params(override=None):
     PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
                                    "parameters.json")

test/parameters.json

@@ -1,4 +1,5 @@
 {
+    "check_formats": false,
     "consoletitle": false,
     "continuedl": true,
     "forcedescription": false,

test/test_InfoExtractor.py

@@ -35,13 +35,13 @@ class InfoExtractorTestRequestHandler(compat_http_server.BaseHTTPRequestHandler)
             assert False
 
 
-class TestIE(InfoExtractor):
+class DummyIE(InfoExtractor):
     pass
 
 
 class TestInfoExtractor(unittest.TestCase):
     def setUp(self):
-        self.ie = TestIE(FakeYDL())
+        self.ie = DummyIE(FakeYDL())
 
     def test_ie_key(self):
         self.assertEqual(get_info_extractor(YoutubeIE.ie_key()), YoutubeIE)

test/test_YoutubeDL.py

@@ -35,6 +35,9 @@ class YDL(FakeYDL):
     def to_screen(self, msg):
         self.msgs.append(msg)
 
+    def dl(self, *args, **kwargs):
+        assert False, 'Downloader must not be invoked for test_YoutubeDL'
+
 
 def _make_result(formats, **kwargs):
     res = {
@@ -117,35 +120,24 @@ class TestFormatSelection(unittest.TestCase):
         ]
         info_dict = _make_result(formats)
 
-        ydl = YDL({'format': '20/47'})
-        ydl.process_ie_result(info_dict.copy())
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], '47')
+        def test(inp, *expected, multi=False):
+            ydl = YDL({
+                'format': inp,
+                'allow_multiple_video_streams': multi,
+                'allow_multiple_audio_streams': multi,
+            })
+            ydl.process_ie_result(info_dict.copy())
+            downloaded = map(lambda x: x['format_id'], ydl.downloaded_info_dicts)
+            self.assertEqual(list(downloaded), list(expected))
 
-        ydl = YDL({'format': '20/71/worst'})
-        ydl.process_ie_result(info_dict.copy())
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], '35')
-
-        ydl = YDL()
-        ydl.process_ie_result(info_dict.copy())
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], '2')
-
-        ydl = YDL({'format': 'webm/mp4'})
-        ydl.process_ie_result(info_dict.copy())
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], '47')
-
-        ydl = YDL({'format': '3gp/40/mp4'})
-        ydl.process_ie_result(info_dict.copy())
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], '35')
-
-        ydl = YDL({'format': 'example-with-dashes'})
-        ydl.process_ie_result(info_dict.copy())
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], 'example-with-dashes')
+        test('20/47', '47')
+        test('20/71/worst', '35')
+        test(None, '2')
+        test('webm/mp4', '47')
+        test('3gp/40/mp4', '35')
+        test('example-with-dashes', 'example-with-dashes')
+        test('all', '35', 'example-with-dashes', '45', '47', '2')  # Order doesn't actually matter for this
+        test('mergeall', '2+47+45+example-with-dashes+35', multi=True)
 
     def test_format_selection_audio(self):
         formats = [
@@ -461,14 +453,13 @@ class TestFormatSelection(unittest.TestCase):
 
     def test_invalid_format_specs(self):
         def assert_syntax_error(format_spec):
-            ydl = YDL({'format': format_spec})
-            info_dict = _make_result([{'format_id': 'foo', 'url': TEST_URL}])
-            self.assertRaises(SyntaxError, ydl.process_ie_result, info_dict)
+            self.assertRaises(SyntaxError, YDL, {'format': format_spec})
 
         assert_syntax_error('bestvideo,,best')
         assert_syntax_error('+bestaudio')
         assert_syntax_error('bestvideo+')
         assert_syntax_error('/')
+        assert_syntax_error('[720<height]')
 
     def test_format_filtering(self):
         formats = [
@@ -665,15 +656,15 @@ class TestYoutubeDL(unittest.TestCase):
     }
 
     def test_prepare_outtmpl_and_filename(self):
-        def test(tmpl, expected, **params):
+        def test(tmpl, expected, *, info=None, **params):
             params['outtmpl'] = tmpl
             ydl = YoutubeDL(params)
             ydl._num_downloads = 1
             self.assertEqual(ydl.validate_outtmpl(tmpl), None)
 
-            outtmpl, tmpl_dict = ydl.prepare_outtmpl(tmpl, self.outtmpl_info)
+            outtmpl, tmpl_dict = ydl.prepare_outtmpl(tmpl, info or self.outtmpl_info)
             out = outtmpl % tmpl_dict
-            fname = ydl.prepare_filename(self.outtmpl_info)
+            fname = ydl.prepare_filename(info or self.outtmpl_info)
 
             if callable(expected):
                 self.assertTrue(expected(out))
@@ -701,6 +692,15 @@ class TestYoutubeDL(unittest.TestCase):
         test('%(width)06d.%%(ext)s', 'NA.%(ext)s')
         test('%%(width)06d.%(ext)s', '%(width)06d.mp4')
 
+        # ID sanitization
+        test('%(id)s', '_abcd', info={'id': '_abcd'})
+        test('%(some_id)s', '_abcd', info={'some_id': '_abcd'})
+        test('%(formats.0.id)s', '_abcd', info={'formats': [{'id': '_abcd'}]})
+        test('%(id)s', '-abcd', info={'id': '-abcd'})
+        test('%(id)s', '.abcd', info={'id': '.abcd'})
+        test('%(id)s', 'ab__cd', info={'id': 'ab__cd'})
+        test('%(id)s', ('ab:cd', 'ab -cd'), info={'id': 'ab:cd'})
+
         # Invalid templates
         self.assertTrue(isinstance(YoutubeDL.validate_outtmpl('%'), ValueError))
         self.assertTrue(isinstance(YoutubeDL.validate_outtmpl('%(title)'), ValueError))

test/test_age_restriction.py

@@ -7,8 +7,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import try_rm
-
+from test.helper import try_rm, is_download_test
 
 from yt_dlp import YoutubeDL
 
@@ -32,6 +31,7 @@ def _download_restricted(url, filename, age):
     return res
 
 
+@is_download_test
 class TestAgeRestriction(unittest.TestCase):
     def _assert_restricted(self, url, filename, age, old_age=None):
         self.assertTrue(_download_restricted(url, filename, old_age))

test/test_cookies.py

@@ -0,0 +1,96 @@
+import unittest
+from datetime import datetime, timezone
+
+from yt_dlp import cookies
+from yt_dlp.cookies import (
+    CRYPTO_AVAILABLE,
+    LinuxChromeCookieDecryptor,
+    MacChromeCookieDecryptor,
+    WindowsChromeCookieDecryptor,
+    YDLLogger,
+    parse_safari_cookies,
+    pbkdf2_sha1,
+)
+
+
+class MonkeyPatch:
+    def __init__(self, module, temporary_values):
+        self._module = module
+        self._temporary_values = temporary_values
+        self._backup_values = {}
+
+    def __enter__(self):
+        for name, temp_value in self._temporary_values.items():
+            self._backup_values[name] = getattr(self._module, name)
+            setattr(self._module, name, temp_value)
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        for name, backup_value in self._backup_values.items():
+            setattr(self._module, name, backup_value)
+
+
+class TestCookies(unittest.TestCase):
+    def test_chrome_cookie_decryptor_linux_derive_key(self):
+        key = LinuxChromeCookieDecryptor.derive_key(b'abc')
+        self.assertEqual(key, b'7\xa1\xec\xd4m\xfcA\xc7\xb19Z\xd0\x19\xdcM\x17')
+
+    def test_chrome_cookie_decryptor_mac_derive_key(self):
+        key = MacChromeCookieDecryptor.derive_key(b'abc')
+        self.assertEqual(key, b'Y\xe2\xc0\xd0P\xf6\xf4\xe1l\xc1\x8cQ\xcb|\xcdY')
+
+    def test_chrome_cookie_decryptor_linux_v10(self):
+        with MonkeyPatch(cookies, {'_get_linux_keyring_password': lambda *args, **kwargs: b''}):
+            encrypted_value = b'v10\xccW%\xcd\xe6\xe6\x9fM" \xa7\xb0\xca\xe4\x07\xd6'
+            value = 'USD'
+            decryptor = LinuxChromeCookieDecryptor('Chrome', YDLLogger())
+            self.assertEqual(decryptor.decrypt(encrypted_value), value)
+
+    def test_chrome_cookie_decryptor_linux_v11(self):
+        with MonkeyPatch(cookies, {'_get_linux_keyring_password': lambda *args, **kwargs: b'',
+                                   'KEYRING_AVAILABLE': True}):
+            encrypted_value = b'v11#\x81\x10>`w\x8f)\xc0\xb2\xc1\r\xf4\x1al\xdd\x93\xfd\xf8\xf8N\xf2\xa9\x83\xf1\xe9o\x0elVQd'
+            value = 'tz=Europe.London'
+            decryptor = LinuxChromeCookieDecryptor('Chrome', YDLLogger())
+            self.assertEqual(decryptor.decrypt(encrypted_value), value)
+
+    @unittest.skipIf(not CRYPTO_AVAILABLE, 'cryptography library not available')
+    def test_chrome_cookie_decryptor_windows_v10(self):
+        with MonkeyPatch(cookies, {
+            '_get_windows_v10_key': lambda *args, **kwargs: b'Y\xef\xad\xad\xeerp\xf0Y\xe6\x9b\x12\xc2<z\x16]\n\xbb\xb8\xcb\xd7\x9bA\xc3\x14e\x99{\xd6\xf4&'
+        }):
+            encrypted_value = b'v10T\xb8\xf3\xb8\x01\xa7TtcV\xfc\x88\xb8\xb8\xef\x05\xb5\xfd\x18\xc90\x009\xab\xb1\x893\x85)\x87\xe1\xa9-\xa3\xad='
+            value = '32101439'
+            decryptor = WindowsChromeCookieDecryptor('', YDLLogger())
+            self.assertEqual(decryptor.decrypt(encrypted_value), value)
+
+    def test_chrome_cookie_decryptor_mac_v10(self):
+        with MonkeyPatch(cookies, {'_get_mac_keyring_password': lambda *args, **kwargs: b'6eIDUdtKAacvlHwBVwvg/Q=='}):
+            encrypted_value = b'v10\xb3\xbe\xad\xa1[\x9fC\xa1\x98\xe0\x9a\x01\xd9\xcf\xbfc'
+            value = '2021-06-01-22'
+            decryptor = MacChromeCookieDecryptor('', YDLLogger())
+            self.assertEqual(decryptor.decrypt(encrypted_value), value)
+
+    def test_safari_cookie_parsing(self):
+        cookies = \
+            b'cook\x00\x00\x00\x01\x00\x00\x00i\x00\x00\x01\x00\x01\x00\x00\x00\x10\x00\x00\x00\x00\x00\x00\x00Y' \
+            b'\x00\x00\x00\x00\x00\x00\x00 \x00\x00\x00\x00\x00\x00\x008\x00\x00\x00B\x00\x00\x00F\x00\x00\x00H' \
+            b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x03\xa5>\xc3A\x00\x00\x80\xc3\x07:\xc3A' \
+            b'localhost\x00foo\x00/\x00test%20%3Bcookie\x00\x00\x00\x054\x07\x17 \x05\x00\x00\x00Kbplist00\xd1\x01' \
+            b'\x02_\x10\x18NSHTTPCookieAcceptPolicy\x10\x02\x08\x0b&\x00\x00\x00\x00\x00\x00\x01\x01\x00\x00\x00' \
+            b'\x00\x00\x00\x00\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00('
+
+        jar = parse_safari_cookies(cookies)
+        self.assertEqual(len(jar), 1)
+        cookie = list(jar)[0]
+        self.assertEqual(cookie.domain, 'localhost')
+        self.assertEqual(cookie.port, None)
+        self.assertEqual(cookie.path, '/')
+        self.assertEqual(cookie.name, 'foo')
+        self.assertEqual(cookie.value, 'test%20%3Bcookie')
+        self.assertFalse(cookie.secure)
+        expected_expiration = datetime(2021, 6, 18, 21, 39, 19, tzinfo=timezone.utc)
+        self.assertEqual(cookie.expires, int(expected_expiration.timestamp()))
+
+    def test_pbkdf2_sha1(self):
+        key = pbkdf2_sha1(b'peanuts', b' ' * 16, 1, 16)
+        self.assertEqual(key, b'g\xe1\x8e\x0fQ\x1c\x9b\xf3\xc9`!\xaa\x90\xd9\xd34')

test/test_download.py

@@ -10,12 +10,13 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import (
     assertGreaterEqual,
+    expect_info_dict,
     expect_warnings,
     get_params,
     gettestcases,
-    expect_info_dict,
-    try_rm,
+    is_download_test,
     report_warning,
+    try_rm,
 )
 
 
@@ -64,6 +65,7 @@ def _file_md5(fn):
 defs = gettestcases()
 
 
+@is_download_test
 class TestDownload(unittest.TestCase):
     # Parallel testing in nosetests. See
     # http://nose.readthedocs.org/en/latest/doc_tests/test_multiprocess/multiprocess.html

test/test_iqiyi_sdk_interpreter.py

@@ -8,7 +8,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL
+from test.helper import FakeYDL, is_download_test
 from yt_dlp.extractor import IqiyiIE
 
 
@@ -31,6 +31,7 @@ class WarningLogger(object):
         pass
 
 
+@is_download_test
 class TestIqiyiSDKInterpreter(unittest.TestCase):
     def test_iqiyi_sdk_interpreter(self):
         '''

test/test_post_hooks.py

@@ -7,7 +7,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import get_params, try_rm
+from test.helper import get_params, try_rm, is_download_test
 import yt_dlp.YoutubeDL
 from yt_dlp.utils import DownloadError
 
@@ -22,6 +22,7 @@ TEST_ID = 'gr51aVj-mLg'
 EXPECTED_NAME = 'gr51aVj-mLg'
 
 
+@is_download_test
 class TestPostHooks(unittest.TestCase):
     def setUp(self):
         self.stored_name_1 = None

test/test_socks.py

@@ -14,6 +14,7 @@ import subprocess
 from test.helper import (
     FakeYDL,
     get_params,
+    is_download_test,
 )
 from yt_dlp.compat import (
     compat_str,
@@ -21,6 +22,7 @@ from yt_dlp.compat import (
 )
 
 
+@is_download_test
 class TestMultipleSocks(unittest.TestCase):
     @staticmethod
     def _check_params(attrs):
@@ -76,6 +78,7 @@ class TestMultipleSocks(unittest.TestCase):
             params['secondary_server_ip'])
 
 
+@is_download_test
 class TestSocks(unittest.TestCase):
     _SKIP_SOCKS_TEST = True
 

test/test_subtitles.py

@@ -7,7 +7,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL, md5
+from test.helper import FakeYDL, md5, is_download_test
 
 
 from yt_dlp.extractor import (
@@ -30,6 +30,7 @@ from yt_dlp.extractor import (
 )
 
 
+@is_download_test
 class BaseTestSubtitles(unittest.TestCase):
     url = None
     IE = None
@@ -55,6 +56,7 @@ class BaseTestSubtitles(unittest.TestCase):
         return dict((l, sub_info['data']) for l, sub_info in subtitles.items())
 
 
+@is_download_test
 class TestYoutubeSubtitles(BaseTestSubtitles):
     url = 'QRS8MkLhQmM'
     IE = YoutubeIE
@@ -111,6 +113,7 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.assertFalse(subtitles)
 
 
+@is_download_test
 class TestDailymotionSubtitles(BaseTestSubtitles):
     url = 'http://www.dailymotion.com/video/xczg00'
     IE = DailymotionIE
@@ -134,6 +137,7 @@ class TestDailymotionSubtitles(BaseTestSubtitles):
         self.assertFalse(subtitles)
 
 
+@is_download_test
 class TestTedSubtitles(BaseTestSubtitles):
     url = 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html'
     IE = TEDIE
@@ -149,6 +153,7 @@ class TestTedSubtitles(BaseTestSubtitles):
             self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
 
+@is_download_test
 class TestVimeoSubtitles(BaseTestSubtitles):
     url = 'http://vimeo.com/76979871'
     IE = VimeoIE
@@ -170,6 +175,7 @@ class TestVimeoSubtitles(BaseTestSubtitles):
         self.assertFalse(subtitles)
 
 
+@is_download_test
 class TestWallaSubtitles(BaseTestSubtitles):
     url = 'http://vod.walla.co.il/movie/2705958/the-yes-men'
     IE = WallaIE
@@ -191,6 +197,7 @@ class TestWallaSubtitles(BaseTestSubtitles):
         self.assertFalse(subtitles)
 
 
+@is_download_test
 class TestCeskaTelevizeSubtitles(BaseTestSubtitles):
     url = 'http://www.ceskatelevize.cz/ivysilani/10600540290-u6-uzasny-svet-techniky'
     IE = CeskaTelevizeIE
@@ -212,6 +219,7 @@ class TestCeskaTelevizeSubtitles(BaseTestSubtitles):
         self.assertFalse(subtitles)
 
 
+@is_download_test
 class TestLyndaSubtitles(BaseTestSubtitles):
     url = 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html'
     IE = LyndaIE
@@ -224,6 +232,7 @@ class TestLyndaSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), '09bbe67222259bed60deaa26997d73a7')
 
 
+@is_download_test
 class TestNPOSubtitles(BaseTestSubtitles):
     url = 'http://www.npo.nl/nos-journaal/28-08-2014/POW_00722860'
     IE = NPOIE
@@ -236,6 +245,7 @@ class TestNPOSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['nl']), 'fc6435027572b63fb4ab143abd5ad3f4')
 
 
+@is_download_test
 class TestMTVSubtitles(BaseTestSubtitles):
     url = 'http://www.cc.com/video-clips/p63lk0/adam-devine-s-house-party-chasing-white-swans'
     IE = ComedyCentralIE
@@ -251,6 +261,7 @@ class TestMTVSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), '78206b8d8a0cfa9da64dc026eea48961')
 
 
+@is_download_test
 class TestNRKSubtitles(BaseTestSubtitles):
     url = 'http://tv.nrk.no/serie/ikke-gjoer-dette-hjemme/DMPV73000411/sesong-2/episode-1'
     IE = NRKTVIE
@@ -263,6 +274,7 @@ class TestNRKSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['no']), '544fa917d3197fcbee64634559221cc2')
 
 
+@is_download_test
 class TestRaiPlaySubtitles(BaseTestSubtitles):
     IE = RaiPlayIE
 
@@ -283,6 +295,7 @@ class TestRaiPlaySubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['it']), '4b3264186fbb103508abe5311cfcb9cd')
 
 
+@is_download_test
 class TestVikiSubtitles(BaseTestSubtitles):
     url = 'http://www.viki.com/videos/1060846v-punch-episode-18'
     IE = VikiIE
@@ -295,6 +308,7 @@ class TestVikiSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), '53cb083a5914b2d84ef1ab67b880d18a')
 
 
+@is_download_test
 class TestThePlatformSubtitles(BaseTestSubtitles):
     # from http://www.3playmedia.com/services-features/tools/integrations/theplatform/
     # (see http://theplatform.com/about/partners/type/subtitles-closed-captioning/)
@@ -309,6 +323,7 @@ class TestThePlatformSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), '97e7670cbae3c4d26ae8bcc7fdd78d4b')
 
 
+@is_download_test
 class TestThePlatformFeedSubtitles(BaseTestSubtitles):
     url = 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207'
     IE = ThePlatformFeedIE
@@ -321,6 +336,7 @@ class TestThePlatformFeedSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['en']), '48649a22e82b2da21c9a67a395eedade')
 
 
+@is_download_test
 class TestRtveSubtitles(BaseTestSubtitles):
     url = 'http://www.rtve.es/alacarta/videos/los-misterios-de-laura/misterios-laura-capitulo-32-misterio-del-numero-17-2-parte/2428621/'
     IE = RTVEALaCartaIE
@@ -335,6 +351,7 @@ class TestRtveSubtitles(BaseTestSubtitles):
         self.assertEqual(md5(subtitles['es']), '69e70cae2d40574fb7316f31d6eb7fca')
 
 
+@is_download_test
 class TestDemocracynowSubtitles(BaseTestSubtitles):
     url = 'http://www.democracynow.org/shows/2015/7/3'
     IE = DemocracynowIE

test/test_utils.py

@@ -12,6 +12,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 # Various small unit tests
 import io
+import itertools
 import json
 import xml.etree.ElementTree
 
@@ -108,6 +109,7 @@ from yt_dlp.utils import (
     cli_bool_option,
     parse_codecs,
     iri_to_uri,
+    LazyList,
 )
 from yt_dlp.compat import (
     compat_chr,
@@ -1052,6 +1054,9 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('{ "040": "040" }')
         self.assertEqual(json.loads(on), {'040': '040'})
 
+        on = js_to_json('[1,//{},\n2]')
+        self.assertEqual(json.loads(on), [1, 2])
+
     def test_js_to_json_malformed(self):
         self.assertEqual(js_to_json('42a1'), '42"a1"')
         self.assertEqual(js_to_json('42a-1'), '42"a"-1')
@@ -1525,6 +1530,47 @@ Line 1
         self.assertEqual(clean_podcast_url('https://www.podtrac.com/pts/redirect.mp3/chtbl.com/track/5899E/traffic.megaphone.fm/HSW7835899191.mp3'), 'https://traffic.megaphone.fm/HSW7835899191.mp3')
         self.assertEqual(clean_podcast_url('https://play.podtrac.com/npr-344098539/edge1.pod.npr.org/anon.npr-podcasts/podcast/npr/waitwait/2020/10/20201003_waitwait_wwdtmpodcast201003-015621a5-f035-4eca-a9a1-7c118d90bc3c.mp3'), 'https://edge1.pod.npr.org/anon.npr-podcasts/podcast/npr/waitwait/2020/10/20201003_waitwait_wwdtmpodcast201003-015621a5-f035-4eca-a9a1-7c118d90bc3c.mp3')
 
+    def test_LazyList(self):
+        it = list(range(10))
+
+        self.assertEqual(list(LazyList(it)), it)
+        self.assertEqual(LazyList(it).exhaust(), it)
+        self.assertEqual(LazyList(it)[5], it[5])
+
+        self.assertEqual(LazyList(it)[::2], it[::2])
+        self.assertEqual(LazyList(it)[1::2], it[1::2])
+        self.assertEqual(LazyList(it)[6:2:-2], it[6:2:-2])
+        self.assertEqual(LazyList(it)[::-1], it[::-1])
+
+        self.assertTrue(LazyList(it))
+        self.assertFalse(LazyList(range(0)))
+        self.assertEqual(len(LazyList(it)), len(it))
+        self.assertEqual(repr(LazyList(it)), repr(it))
+        self.assertEqual(str(LazyList(it)), str(it))
+
+        self.assertEqual(list(LazyList(it).reverse()), it[::-1])
+        self.assertEqual(list(LazyList(it).reverse()[1:3:7]), it[::-1][1:3:7])
+
+    def test_LazyList_laziness(self):
+
+        def test(ll, idx, val, cache):
+            self.assertEqual(ll[idx], val)
+            self.assertEqual(getattr(ll, '_LazyList__cache'), list(cache))
+
+        ll = LazyList(range(10))
+        test(ll, 0, 0, range(1))
+        test(ll, 5, 5, range(6))
+        test(ll, -3, 7, range(10))
+
+        ll = LazyList(range(10)).reverse()
+        test(ll, -1, 0, range(1))
+        test(ll, 3, 6, range(10))
+
+        ll = LazyList(itertools.count())
+        test(ll, 10, 10, range(11))
+        ll.reverse()
+        test(ll, -15, 14, range(15))
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_write_annotations.py

@@ -8,7 +8,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import get_params, try_rm
+from test.helper import get_params, try_rm, is_download_test
 
 
 import io
@@ -38,6 +38,7 @@ ANNOTATIONS_FILE = TEST_ID + '.annotations.xml'
 EXPECTED_ANNOTATIONS = ['Speech bubble', 'Note', 'Title', 'Spotlight', 'Label']
 
 
+@is_download_test
 class TestAnnotations(unittest.TestCase):
     def setUp(self):
         # Clear old files

test/test_youtube_lists.py

@@ -7,7 +7,7 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL
+from test.helper import FakeYDL, is_download_test
 
 
 from yt_dlp.extractor import (
@@ -17,6 +17,7 @@ from yt_dlp.extractor import (
 )
 
 
+@is_download_test
 class TestYoutubeLists(unittest.TestCase):
     def assertIsPlaylist(self, info):
         """Make sure the info has '_type' set to 'playlist'"""

test/test_youtube_signature.py

@@ -12,7 +12,7 @@ import io
 import re
 import string
 
-from test.helper import FakeYDL
+from test.helper import FakeYDL, is_download_test
 from yt_dlp.extractor import YoutubeIE
 from yt_dlp.compat import compat_str, compat_urlretrieve
 
@@ -65,6 +65,7 @@ _TESTS = [
 ]
 
 
+@is_download_test
 class TestPlayerInfo(unittest.TestCase):
     def test_youtube_extract_player_info(self):
         PLAYER_URLS = (
@@ -87,6 +88,7 @@ class TestPlayerInfo(unittest.TestCase):
             self.assertEqual(player_id, expected_player_id)
 
 
+@is_download_test
 class TestSignature(unittest.TestCase):
     def setUp(self):
         TEST_DIR = os.path.dirname(os.path.abspath(__file__))

tox.ini

@@ -1,5 +1,7 @@
 [tox]
 envlist = py26,py27,py33,py34,py35
+
+# Needed?
 [testenv]
 deps =
    nose

yt_dlp/__init__.py

@@ -20,6 +20,7 @@ from .compat import (
     compat_getpass,
     workaround_optparse_bug9161,
 )
+from .cookies import SUPPORTED_BROWSERS
 from .utils import (
     DateRange,
     decodeOption,
@@ -151,6 +152,11 @@ def _real_main(argv=None):
         if numeric_limit is None:
             parser.error('invalid rate limit specified')
         opts.ratelimit = numeric_limit
+    if opts.throttledratelimit is not None:
+        numeric_limit = FileDownloader.parse_bytes(opts.throttledratelimit)
+        if numeric_limit is None:
+            parser.error('invalid rate limit specified')
+        opts.throttledratelimit = numeric_limit
     if opts.min_filesize is not None:
         numeric_limit = FileDownloader.parse_bytes(opts.min_filesize)
         if numeric_limit is None:
@@ -237,6 +243,12 @@ def _real_main(argv=None):
         if opts.convertthumbnails not in FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS:
             parser.error('invalid thumbnail format specified')
 
+    if opts.cookiesfrombrowser is not None:
+        opts.cookiesfrombrowser = [
+            part.strip() or None for part in opts.cookiesfrombrowser.split(':', 1)]
+        if opts.cookiesfrombrowser[0] not in SUPPORTED_BROWSERS:
+            parser.error('unsupported browser specified for cookies')
+
     if opts.date is not None:
         date = DateRange.day(opts.date)
     else:
@@ -268,6 +280,7 @@ def _real_main(argv=None):
         'filename', 'format-sort', 'abort-on-error', 'format-spec', 'no-playlist-metafiles',
         'multistreams', 'no-live-chat', 'playlist-index', 'list-formats', 'no-direct-merge',
         'no-youtube-channel-redirect', 'no-youtube-unavailable-videos', 'no-attach-info-json',
+        'embed-thumbnail-atomicparsley', 'seperate-video-versions',
     ]
     compat_opts = parse_compat_opts()
 
@@ -409,6 +422,13 @@ def _real_main(argv=None):
             # Run this before the actual video download
             'when': 'before_dl'
         })
+    # Must be after all other before_dl
+    if opts.exec_before_dl_cmd:
+        postprocessors.append({
+            'key': 'ExecAfterDownload',
+            'exec_cmd': opts.exec_before_dl_cmd,
+            'when': 'before_dl'
+        })
     if opts.extractaudio:
         postprocessors.append({
             'key': 'FFmpegExtractAudio',
@@ -551,6 +571,7 @@ def _real_main(argv=None):
         'ignoreerrors': opts.ignoreerrors,
         'force_generic_extractor': opts.force_generic_extractor,
         'ratelimit': opts.ratelimit,
+        'throttledratelimit': opts.throttledratelimit,
         'overwrites': opts.overwrites,
         'retries': opts.retries,
         'fragment_retries': opts.fragment_retries,
@@ -614,6 +635,7 @@ def _real_main(argv=None):
         'break_on_reject': opts.break_on_reject,
         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
         'cookiefile': opts.cookiefile,
+        'cookiesfrombrowser': opts.cookiesfrombrowser,
         'nocheckcertificate': opts.no_check_certificate,
         'prefer_insecure': opts.prefer_insecure,
         'proxy': opts.proxy,
@@ -624,6 +646,7 @@ def _real_main(argv=None):
         'include_ads': opts.include_ads,
         'default_search': opts.default_search,
         'dynamic_mpd': opts.dynamic_mpd,
+        'extractor_args': opts.extractor_args,
         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
         'encoding': opts.encoding,

yt_dlp/cookies.py

@@ -0,0 +1,755 @@
+import ctypes
+import json
+import os
+import shutil
+import struct
+import subprocess
+import sys
+import tempfile
+from datetime import datetime, timedelta, timezone
+from hashlib import pbkdf2_hmac
+
+from yt_dlp.aes import aes_cbc_decrypt
+from yt_dlp.compat import (
+    compat_b64decode,
+    compat_cookiejar_Cookie,
+)
+from yt_dlp.utils import (
+    bug_reports_message,
+    bytes_to_intlist,
+    expand_path,
+    intlist_to_bytes,
+    process_communicate_or_kill,
+    YoutubeDLCookieJar,
+)
+
+try:
+    import sqlite3
+    SQLITE_AVAILABLE = True
+except ImportError:
+    # although sqlite3 is part of the standard library, it is possible to compile python without
+    # sqlite support. See: https://github.com/yt-dlp/yt-dlp/issues/544
+    SQLITE_AVAILABLE = False
+
+
+try:
+    from Crypto.Cipher import AES
+    CRYPTO_AVAILABLE = True
+except ImportError:
+    CRYPTO_AVAILABLE = False
+
+try:
+    import keyring
+    KEYRING_AVAILABLE = True
+    KEYRING_UNAVAILABLE_REASON = f'due to unknown reasons{bug_reports_message()}'
+except ImportError:
+    KEYRING_AVAILABLE = False
+    KEYRING_UNAVAILABLE_REASON = (
+        'as the `keyring` module is not installed. '
+        'Please install by running `python3 -m pip install keyring`. '
+        'Depending on your platform, additional packages may be required '
+        'to access the keyring; see  https://pypi.org/project/keyring')
+except Exception as _err:
+    KEYRING_AVAILABLE = False
+    KEYRING_UNAVAILABLE_REASON = 'as the `keyring` module could not be initialized: %s' % _err
+
+
+CHROMIUM_BASED_BROWSERS = {'brave', 'chrome', 'chromium', 'edge', 'opera', 'vivaldi'}
+SUPPORTED_BROWSERS = CHROMIUM_BASED_BROWSERS | {'firefox', 'safari'}
+
+
+class YDLLogger:
+    def __init__(self, ydl=None):
+        self._ydl = ydl
+
+    def debug(self, message):
+        if self._ydl:
+            self._ydl.write_debug(message)
+
+    def info(self, message):
+        if self._ydl:
+            self._ydl.to_screen(f'[Cookies] {message}')
+
+    def warning(self, message, only_once=False):
+        if self._ydl:
+            self._ydl.report_warning(message, only_once)
+
+    def error(self, message):
+        if self._ydl:
+            self._ydl.report_error(message)
+
+
+def load_cookies(cookie_file, browser_specification, ydl):
+    cookie_jars = []
+    if browser_specification is not None:
+        browser_name, profile = _parse_browser_specification(*browser_specification)
+        cookie_jars.append(extract_cookies_from_browser(browser_name, profile, YDLLogger(ydl)))
+
+    if cookie_file is not None:
+        cookie_file = expand_path(cookie_file)
+        jar = YoutubeDLCookieJar(cookie_file)
+        if os.access(cookie_file, os.R_OK):
+            jar.load(ignore_discard=True, ignore_expires=True)
+        cookie_jars.append(jar)
+
+    return _merge_cookie_jars(cookie_jars)
+
+
+def extract_cookies_from_browser(browser_name, profile=None, logger=YDLLogger()):
+    if browser_name == 'firefox':
+        return _extract_firefox_cookies(profile, logger)
+    elif browser_name == 'safari':
+        return _extract_safari_cookies(profile, logger)
+    elif browser_name in CHROMIUM_BASED_BROWSERS:
+        return _extract_chrome_cookies(browser_name, profile, logger)
+    else:
+        raise ValueError('unknown browser: {}'.format(browser_name))
+
+
+def _extract_firefox_cookies(profile, logger):
+    logger.info('Extracting cookies from firefox')
+    if not SQLITE_AVAILABLE:
+        logger.warning('Cannot extract cookies from firefox without sqlite3 support. '
+                       'Please use a python interpreter compiled with sqlite3 support')
+        return YoutubeDLCookieJar()
+
+    if profile is None:
+        search_root = _firefox_browser_dir()
+    elif _is_path(profile):
+        search_root = profile
+    else:
+        search_root = os.path.join(_firefox_browser_dir(), profile)
+
+    cookie_database_path = _find_most_recently_used_file(search_root, 'cookies.sqlite')
+    if cookie_database_path is None:
+        raise FileNotFoundError('could not find firefox cookies database in {}'.format(search_root))
+    logger.debug('extracting from: "{}"'.format(cookie_database_path))
+
+    with tempfile.TemporaryDirectory(prefix='youtube_dl') as tmpdir:
+        cursor = None
+        try:
+            cursor = _open_database_copy(cookie_database_path, tmpdir)
+            cursor.execute('SELECT host, name, value, path, expiry, isSecure FROM moz_cookies')
+            jar = YoutubeDLCookieJar()
+            for host, name, value, path, expiry, is_secure in cursor.fetchall():
+                cookie = compat_cookiejar_Cookie(
+                    version=0, name=name, value=value, port=None, port_specified=False,
+                    domain=host, domain_specified=bool(host), domain_initial_dot=host.startswith('.'),
+                    path=path, path_specified=bool(path), secure=is_secure, expires=expiry, discard=False,
+                    comment=None, comment_url=None, rest={})
+                jar.set_cookie(cookie)
+            logger.info('Extracted {} cookies from firefox'.format(len(jar)))
+            return jar
+        finally:
+            if cursor is not None:
+                cursor.connection.close()
+
+
+def _firefox_browser_dir():
+    if sys.platform in ('linux', 'linux2'):
+        return os.path.expanduser('~/.mozilla/firefox')
+    elif sys.platform == 'win32':
+        return os.path.expandvars(r'%APPDATA%\Mozilla\Firefox\Profiles')
+    elif sys.platform == 'darwin':
+        return os.path.expanduser('~/Library/Application Support/Firefox')
+    else:
+        raise ValueError('unsupported platform: {}'.format(sys.platform))
+
+
+def _get_chromium_based_browser_settings(browser_name):
+    # https://chromium.googlesource.com/chromium/src/+/HEAD/docs/user_data_dir.md
+    if sys.platform in ('linux', 'linux2'):
+        config = _config_home()
+        browser_dir = {
+            'brave': os.path.join(config, 'BraveSoftware/Brave-Browser'),
+            'chrome': os.path.join(config, 'google-chrome'),
+            'chromium': os.path.join(config, 'chromium'),
+            'edge': os.path.join(config, 'microsoft-edge'),
+            'opera': os.path.join(config, 'opera'),
+            'vivaldi': os.path.join(config, 'vivaldi'),
+        }[browser_name]
+
+    elif sys.platform == 'win32':
+        appdata_local = os.path.expandvars('%LOCALAPPDATA%')
+        appdata_roaming = os.path.expandvars('%APPDATA%')
+        browser_dir = {
+            'brave': os.path.join(appdata_local, r'BraveSoftware\Brave-Browser\User Data'),
+            'chrome': os.path.join(appdata_local, r'Google\Chrome\User Data'),
+            'chromium': os.path.join(appdata_local, r'Chromium\User Data'),
+            'edge': os.path.join(appdata_local, r'Microsoft\Edge\User Data'),
+            'opera': os.path.join(appdata_roaming, r'Opera Software\Opera Stable'),
+            'vivaldi': os.path.join(appdata_local, r'Vivaldi\User Data'),
+        }[browser_name]
+
+    elif sys.platform == 'darwin':
+        appdata = os.path.expanduser('~/Library/Application Support')
+        browser_dir = {
+            'brave': os.path.join(appdata, 'BraveSoftware/Brave-Browser'),
+            'chrome': os.path.join(appdata, 'Google/Chrome'),
+            'chromium': os.path.join(appdata, 'Chromium'),
+            'edge': os.path.join(appdata, 'Microsoft Edge'),
+            'opera': os.path.join(appdata, 'com.operasoftware.Opera'),
+            'vivaldi': os.path.join(appdata, 'Vivaldi'),
+        }[browser_name]
+
+    else:
+        raise ValueError('unsupported platform: {}'.format(sys.platform))
+
+    # Linux keyring names can be determined by snooping on dbus while opening the browser in KDE:
+    # dbus-monitor "interface='org.kde.KWallet'" "type=method_return"
+    keyring_name = {
+        'brave': 'Brave',
+        'chrome': 'Chrome',
+        'chromium': 'Chromium',
+        'edge': 'Microsoft Edge' if sys.platform == 'darwin' else 'Chromium',
+        'opera': 'Opera' if sys.platform == 'darwin' else 'Chromium',
+        'vivaldi': 'Vivaldi' if sys.platform == 'darwin' else 'Chrome',
+    }[browser_name]
+
+    browsers_without_profiles = {'opera'}
+
+    return {
+        'browser_dir': browser_dir,
+        'keyring_name': keyring_name,
+        'supports_profiles': browser_name not in browsers_without_profiles
+    }
+
+
+def _extract_chrome_cookies(browser_name, profile, logger):
+    logger.info('Extracting cookies from {}'.format(browser_name))
+
+    if not SQLITE_AVAILABLE:
+        logger.warning(('Cannot extract cookies from {} without sqlite3 support. '
+                        'Please use a python interpreter compiled with sqlite3 support').format(browser_name))
+        return YoutubeDLCookieJar()
+
+    config = _get_chromium_based_browser_settings(browser_name)
+
+    if profile is None:
+        search_root = config['browser_dir']
+    elif _is_path(profile):
+        search_root = profile
+        config['browser_dir'] = os.path.dirname(profile) if config['supports_profiles'] else profile
+    else:
+        if config['supports_profiles']:
+            search_root = os.path.join(config['browser_dir'], profile)
+        else:
+            logger.error('{} does not support profiles'.format(browser_name))
+            search_root = config['browser_dir']
+
+    cookie_database_path = _find_most_recently_used_file(search_root, 'Cookies')
+    if cookie_database_path is None:
+        raise FileNotFoundError('could not find {} cookies database in "{}"'.format(browser_name, search_root))
+    logger.debug('extracting from: "{}"'.format(cookie_database_path))
+
+    decryptor = get_cookie_decryptor(config['browser_dir'], config['keyring_name'], logger)
+
+    with tempfile.TemporaryDirectory(prefix='youtube_dl') as tmpdir:
+        cursor = None
+        try:
+            cursor = _open_database_copy(cookie_database_path, tmpdir)
+            cursor.connection.text_factory = bytes
+            column_names = _get_column_names(cursor, 'cookies')
+            secure_column = 'is_secure' if 'is_secure' in column_names else 'secure'
+            cursor.execute('SELECT host_key, name, value, encrypted_value, path, '
+                           'expires_utc, {} FROM cookies'.format(secure_column))
+            jar = YoutubeDLCookieJar()
+            failed_cookies = 0
+            for host_key, name, value, encrypted_value, path, expires_utc, is_secure in cursor.fetchall():
+                host_key = host_key.decode('utf-8')
+                name = name.decode('utf-8')
+                value = value.decode('utf-8')
+                path = path.decode('utf-8')
+
+                if not value and encrypted_value:
+                    value = decryptor.decrypt(encrypted_value)
+                    if value is None:
+                        failed_cookies += 1
+                        continue
+
+                cookie = compat_cookiejar_Cookie(
+                    version=0, name=name, value=value, port=None, port_specified=False,
+                    domain=host_key, domain_specified=bool(host_key), domain_initial_dot=host_key.startswith('.'),
+                    path=path, path_specified=bool(path), secure=is_secure, expires=expires_utc, discard=False,
+                    comment=None, comment_url=None, rest={})
+                jar.set_cookie(cookie)
+            if failed_cookies > 0:
+                failed_message = ' ({} could not be decrypted)'.format(failed_cookies)
+            else:
+                failed_message = ''
+            logger.info('Extracted {} cookies from {}{}'.format(len(jar), browser_name, failed_message))
+            return jar
+        finally:
+            if cursor is not None:
+                cursor.connection.close()
+
+
+class ChromeCookieDecryptor:
+    """
+    Overview:
+
+        Linux:
+        - cookies are either v10 or v11
+            - v10: AES-CBC encrypted with a fixed key
+            - v11: AES-CBC encrypted with an OS protected key (keyring)
+            - v11 keys can be stored in various places depending on the activate desktop environment [2]
+
+        Mac:
+        - cookies are either v10 or not v10
+            - v10: AES-CBC encrypted with an OS protected key (keyring) and more key derivation iterations than linux
+            - not v10: 'old data' stored as plaintext
+
+        Windows:
+        - cookies are either v10 or not v10
+            - v10: AES-GCM encrypted with a key which is encrypted with DPAPI
+            - not v10: encrypted with DPAPI
+
+    Sources:
+    - [1] https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/
+    - [2] https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/key_storage_linux.cc
+        - KeyStorageLinux::CreateService
+    """
+
+    def decrypt(self, encrypted_value):
+        raise NotImplementedError
+
+
+def get_cookie_decryptor(browser_root, browser_keyring_name, logger):
+    if sys.platform in ('linux', 'linux2'):
+        return LinuxChromeCookieDecryptor(browser_keyring_name, logger)
+    elif sys.platform == 'darwin':
+        return MacChromeCookieDecryptor(browser_keyring_name, logger)
+    elif sys.platform == 'win32':
+        return WindowsChromeCookieDecryptor(browser_root, logger)
+    else:
+        raise NotImplementedError('Chrome cookie decryption is not supported '
+                                  'on this platform: {}'.format(sys.platform))
+
+
+class LinuxChromeCookieDecryptor(ChromeCookieDecryptor):
+    def __init__(self, browser_keyring_name, logger):
+        self._logger = logger
+        self._v10_key = self.derive_key(b'peanuts')
+        if KEYRING_AVAILABLE:
+            self._v11_key = self.derive_key(_get_linux_keyring_password(browser_keyring_name))
+        else:
+            self._v11_key = None
+
+    @staticmethod
+    def derive_key(password):
+        # values from
+        # https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_linux.cc
+        return pbkdf2_sha1(password, salt=b'saltysalt', iterations=1, key_length=16)
+
+    def decrypt(self, encrypted_value):
+        version = encrypted_value[:3]
+        ciphertext = encrypted_value[3:]
+
+        if version == b'v10':
+            return _decrypt_aes_cbc(ciphertext, self._v10_key, self._logger)
+
+        elif version == b'v11':
+            if self._v11_key is None:
+                self._logger.warning(f'cannot decrypt cookie {KEYRING_UNAVAILABLE_REASON}', only_once=True)
+                return None
+            return _decrypt_aes_cbc(ciphertext, self._v11_key, self._logger)
+
+        else:
+            return None
+
+
+class MacChromeCookieDecryptor(ChromeCookieDecryptor):
+    def __init__(self, browser_keyring_name, logger):
+        self._logger = logger
+        password = _get_mac_keyring_password(browser_keyring_name)
+        self._v10_key = None if password is None else self.derive_key(password)
+
+    @staticmethod
+    def derive_key(password):
+        # values from
+        # https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_mac.mm
+        return pbkdf2_sha1(password, salt=b'saltysalt', iterations=1003, key_length=16)
+
+    def decrypt(self, encrypted_value):
+        version = encrypted_value[:3]
+        ciphertext = encrypted_value[3:]
+
+        if version == b'v10':
+            if self._v10_key is None:
+                self._logger.warning('cannot decrypt v10 cookies: no key found', only_once=True)
+                return None
+
+            return _decrypt_aes_cbc(ciphertext, self._v10_key, self._logger)
+
+        else:
+            # other prefixes are considered 'old data' which were stored as plaintext
+            # https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_mac.mm
+            return encrypted_value
+
+
+class WindowsChromeCookieDecryptor(ChromeCookieDecryptor):
+    def __init__(self, browser_root, logger):
+        self._logger = logger
+        self._v10_key = _get_windows_v10_key(browser_root, logger)
+
+    def decrypt(self, encrypted_value):
+        version = encrypted_value[:3]
+        ciphertext = encrypted_value[3:]
+
+        if version == b'v10':
+            if self._v10_key is None:
+                self._logger.warning('cannot decrypt v10 cookies: no key found', only_once=True)
+                return None
+            elif not CRYPTO_AVAILABLE:
+                self._logger.warning('cannot decrypt cookie as the `pycryptodome` module is not installed. '
+                                     'Please install by running `python3 -m pip install pycryptodome`',
+                                     only_once=True)
+                return None
+
+            # https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_win.cc
+            #   kNonceLength
+            nonce_length = 96 // 8
+            # boringssl
+            #   EVP_AEAD_AES_GCM_TAG_LEN
+            authentication_tag_length = 16
+
+            raw_ciphertext = ciphertext
+            nonce = raw_ciphertext[:nonce_length]
+            ciphertext = raw_ciphertext[nonce_length:-authentication_tag_length]
+            authentication_tag = raw_ciphertext[-authentication_tag_length:]
+
+            return _decrypt_aes_gcm(ciphertext, self._v10_key, nonce, authentication_tag, self._logger)
+
+        else:
+            # any other prefix means the data is DPAPI encrypted
+            # https://chromium.googlesource.com/chromium/src/+/refs/heads/main/components/os_crypt/os_crypt_win.cc
+            return _decrypt_windows_dpapi(encrypted_value, self._logger).decode('utf-8')
+
+
+def _extract_safari_cookies(profile, logger):
+    if profile is not None:
+        logger.error('safari does not support profiles')
+    if sys.platform != 'darwin':
+        raise ValueError('unsupported platform: {}'.format(sys.platform))
+
+    cookies_path = os.path.expanduser('~/Library/Cookies/Cookies.binarycookies')
+
+    if not os.path.isfile(cookies_path):
+        raise FileNotFoundError('could not find safari cookies database')
+
+    with open(cookies_path, 'rb') as f:
+        cookies_data = f.read()
+
+    jar = parse_safari_cookies(cookies_data, logger=logger)
+    logger.info('Extracted {} cookies from safari'.format(len(jar)))
+    return jar
+
+
+class ParserError(Exception):
+    pass
+
+
+class DataParser:
+    def __init__(self, data, logger):
+        self._data = data
+        self.cursor = 0
+        self._logger = logger
+
+    def read_bytes(self, num_bytes):
+        if num_bytes < 0:
+            raise ParserError('invalid read of {} bytes'.format(num_bytes))
+        end = self.cursor + num_bytes
+        if end > len(self._data):
+            raise ParserError('reached end of input')
+        data = self._data[self.cursor:end]
+        self.cursor = end
+        return data
+
+    def expect_bytes(self, expected_value, message):
+        value = self.read_bytes(len(expected_value))
+        if value != expected_value:
+            raise ParserError('unexpected value: {} != {} ({})'.format(value, expected_value, message))
+
+    def read_uint(self, big_endian=False):
+        data_format = '>I' if big_endian else '<I'
+        return struct.unpack(data_format, self.read_bytes(4))[0]
+
+    def read_double(self, big_endian=False):
+        data_format = '>d' if big_endian else '<d'
+        return struct.unpack(data_format, self.read_bytes(8))[0]
+
+    def read_cstring(self):
+        buffer = []
+        while True:
+            c = self.read_bytes(1)
+            if c == b'\x00':
+                return b''.join(buffer).decode('utf-8')
+            else:
+                buffer.append(c)
+
+    def skip(self, num_bytes, description='unknown'):
+        if num_bytes > 0:
+            self._logger.debug('skipping {} bytes ({}): {}'.format(
+                num_bytes, description, self.read_bytes(num_bytes)))
+        elif num_bytes < 0:
+            raise ParserError('invalid skip of {} bytes'.format(num_bytes))
+
+    def skip_to(self, offset, description='unknown'):
+        self.skip(offset - self.cursor, description)
+
+    def skip_to_end(self, description='unknown'):
+        self.skip_to(len(self._data), description)
+
+
+def _mac_absolute_time_to_posix(timestamp):
+    return int((datetime(2001, 1, 1, 0, 0, tzinfo=timezone.utc) + timedelta(seconds=timestamp)).timestamp())
+
+
+def _parse_safari_cookies_header(data, logger):
+    p = DataParser(data, logger)
+    p.expect_bytes(b'cook', 'database signature')
+    number_of_pages = p.read_uint(big_endian=True)
+    page_sizes = [p.read_uint(big_endian=True) for _ in range(number_of_pages)]
+    return page_sizes, p.cursor
+
+
+def _parse_safari_cookies_page(data, jar, logger):
+    p = DataParser(data, logger)
+    p.expect_bytes(b'\x00\x00\x01\x00', 'page signature')
+    number_of_cookies = p.read_uint()
+    record_offsets = [p.read_uint() for _ in range(number_of_cookies)]
+    if number_of_cookies == 0:
+        logger.debug('a cookies page of size {} has no cookies'.format(len(data)))
+        return
+
+    p.skip_to(record_offsets[0], 'unknown page header field')
+
+    for record_offset in record_offsets:
+        p.skip_to(record_offset, 'space between records')
+        record_length = _parse_safari_cookies_record(data[record_offset:], jar, logger)
+        p.read_bytes(record_length)
+    p.skip_to_end('space in between pages')
+
+
+def _parse_safari_cookies_record(data, jar, logger):
+    p = DataParser(data, logger)
+    record_size = p.read_uint()
+    p.skip(4, 'unknown record field 1')
+    flags = p.read_uint()
+    is_secure = bool(flags & 0x0001)
+    p.skip(4, 'unknown record field 2')
+    domain_offset = p.read_uint()
+    name_offset = p.read_uint()
+    path_offset = p.read_uint()
+    value_offset = p.read_uint()
+    p.skip(8, 'unknown record field 3')
+    expiration_date = _mac_absolute_time_to_posix(p.read_double())
+    _creation_date = _mac_absolute_time_to_posix(p.read_double())  # noqa: F841
+
+    try:
+        p.skip_to(domain_offset)
+        domain = p.read_cstring()
+
+        p.skip_to(name_offset)
+        name = p.read_cstring()
+
+        p.skip_to(path_offset)
+        path = p.read_cstring()
+
+        p.skip_to(value_offset)
+        value = p.read_cstring()
+    except UnicodeDecodeError:
+        logger.warning('failed to parse cookie because UTF-8 decoding failed')
+        return record_size
+
+    p.skip_to(record_size, 'space at the end of the record')
+
+    cookie = compat_cookiejar_Cookie(
+        version=0, name=name, value=value, port=None, port_specified=False,
+        domain=domain, domain_specified=bool(domain), domain_initial_dot=domain.startswith('.'),
+        path=path, path_specified=bool(path), secure=is_secure, expires=expiration_date, discard=False,
+        comment=None, comment_url=None, rest={})
+    jar.set_cookie(cookie)
+    return record_size
+
+
+def parse_safari_cookies(data, jar=None, logger=YDLLogger()):
+    """
+    References:
+        - https://github.com/libyal/dtformats/blob/main/documentation/Safari%20Cookies.asciidoc
+            - this data appears to be out of date but the important parts of the database structure is the same
+            - there are a few bytes here and there which are skipped during parsing
+    """
+    if jar is None:
+        jar = YoutubeDLCookieJar()
+    page_sizes, body_start = _parse_safari_cookies_header(data, logger)
+    p = DataParser(data[body_start:], logger)
+    for page_size in page_sizes:
+        _parse_safari_cookies_page(p.read_bytes(page_size), jar, logger)
+    p.skip_to_end('footer')
+    return jar
+
+
+def _get_linux_keyring_password(browser_keyring_name):
+    password = keyring.get_password('{} Keys'.format(browser_keyring_name),
+                                    '{} Safe Storage'.format(browser_keyring_name))
+    if password is None:
+        # this sometimes occurs in KDE because chrome does not check hasEntry and instead
+        # just tries to read the value (which kwallet returns "") whereas keyring checks hasEntry
+        # to verify this:
+        # dbus-monitor "interface='org.kde.KWallet'" "type=method_return"
+        # while starting chrome.
+        # this may be a bug as the intended behaviour is to generate a random password and store
+        # it, but that doesn't matter here.
+        password = ''
+    return password.encode('utf-8')
+
+
+def _get_mac_keyring_password(browser_keyring_name):
+    if KEYRING_AVAILABLE:
+        password = keyring.get_password('{} Safe Storage'.format(browser_keyring_name), browser_keyring_name)
+        return password.encode('utf-8')
+    else:
+        proc = subprocess.Popen(['security', 'find-generic-password',
+                                 '-w',  # write password to stdout
+                                 '-a', browser_keyring_name,  # match 'account'
+                                 '-s', '{} Safe Storage'.format(browser_keyring_name)],  # match 'service'
+                                stdout=subprocess.PIPE,
+                                stderr=subprocess.DEVNULL)
+        try:
+            stdout, stderr = process_communicate_or_kill(proc)
+            return stdout
+        except BaseException:
+            return None
+
+
+def _get_windows_v10_key(browser_root, logger):
+    path = _find_most_recently_used_file(browser_root, 'Local State')
+    if path is None:
+        logger.error('could not find local state file')
+        return None
+    with open(path, 'r') as f:
+        data = json.load(f)
+    try:
+        base64_key = data['os_crypt']['encrypted_key']
+    except KeyError:
+        logger.error('no encrypted key in Local State')
+        return None
+    encrypted_key = compat_b64decode(base64_key)
+    prefix = b'DPAPI'
+    if not encrypted_key.startswith(prefix):
+        logger.error('invalid key')
+        return None
+    return _decrypt_windows_dpapi(encrypted_key[len(prefix):], logger)
+
+
+def pbkdf2_sha1(password, salt, iterations, key_length):
+    return pbkdf2_hmac('sha1', password, salt, iterations, key_length)
+
+
+def _decrypt_aes_cbc(ciphertext, key, logger, initialization_vector=b' ' * 16):
+    plaintext = aes_cbc_decrypt(bytes_to_intlist(ciphertext),
+                                bytes_to_intlist(key),
+                                bytes_to_intlist(initialization_vector))
+    padding_length = plaintext[-1]
+    try:
+        return intlist_to_bytes(plaintext[:-padding_length]).decode('utf-8')
+    except UnicodeDecodeError:
+        logger.warning('failed to decrypt cookie because UTF-8 decoding failed. Possibly the key is wrong?')
+        return None
+
+
+def _decrypt_aes_gcm(ciphertext, key, nonce, authentication_tag, logger):
+    cipher = AES.new(key, AES.MODE_GCM, nonce)
+    try:
+        plaintext = cipher.decrypt_and_verify(ciphertext, authentication_tag)
+    except ValueError:
+        logger.warning('failed to decrypt cookie because the MAC check failed. Possibly the key is wrong?')
+        return None
+
+    try:
+        return plaintext.decode('utf-8')
+    except UnicodeDecodeError:
+        logger.warning('failed to decrypt cookie because UTF-8 decoding failed. Possibly the key is wrong?')
+        return None
+
+
+def _decrypt_windows_dpapi(ciphertext, logger):
+    """
+    References:
+        - https://docs.microsoft.com/en-us/windows/win32/api/dpapi/nf-dpapi-cryptunprotectdata
+    """
+    from ctypes.wintypes import DWORD
+
+    class DATA_BLOB(ctypes.Structure):
+        _fields_ = [('cbData', DWORD),
+                    ('pbData', ctypes.POINTER(ctypes.c_char))]
+
+    buffer = ctypes.create_string_buffer(ciphertext)
+    blob_in = DATA_BLOB(ctypes.sizeof(buffer), buffer)
+    blob_out = DATA_BLOB()
+    ret = ctypes.windll.crypt32.CryptUnprotectData(
+        ctypes.byref(blob_in),  # pDataIn
+        None,  # ppszDataDescr: human readable description of pDataIn
+        None,  # pOptionalEntropy: salt?
+        None,  # pvReserved: must be NULL
+        None,  # pPromptStruct: information about prompts to display
+        0,  # dwFlags
+        ctypes.byref(blob_out)  # pDataOut
+    )
+    if not ret:
+        logger.warning('failed to decrypt with DPAPI')
+        return None
+
+    result = ctypes.string_at(blob_out.pbData, blob_out.cbData)
+    ctypes.windll.kernel32.LocalFree(blob_out.pbData)
+    return result
+
+
+def _config_home():
+    return os.environ.get('XDG_CONFIG_HOME', os.path.expanduser('~/.config'))
+
+
+def _open_database_copy(database_path, tmpdir):
+    # cannot open sqlite databases if they are already in use (e.g. by the browser)
+    database_copy_path = os.path.join(tmpdir, 'temporary.sqlite')
+    shutil.copy(database_path, database_copy_path)
+    conn = sqlite3.connect(database_copy_path)
+    return conn.cursor()
+
+
+def _get_column_names(cursor, table_name):
+    table_info = cursor.execute('PRAGMA table_info({})'.format(table_name)).fetchall()
+    return [row[1].decode('utf-8') for row in table_info]
+
+
+def _find_most_recently_used_file(root, filename):
+    # if there are multiple browser profiles, take the most recently used one
+    paths = []
+    for root, dirs, files in os.walk(root):
+        for file in files:
+            if file == filename:
+                paths.append(os.path.join(root, file))
+    return None if not paths else max(paths, key=lambda path: os.lstat(path).st_mtime)
+
+
+def _merge_cookie_jars(jars):
+    output_jar = YoutubeDLCookieJar()
+    for jar in jars:
+        for cookie in jar:
+            output_jar.set_cookie(cookie)
+        if jar.filename is not None:
+            output_jar.filename = jar.filename
+    return output_jar
+
+
+def _is_path(value):
+    return os.path.sep in value
+
+
+def _parse_browser_specification(browser_name, profile=None):
+    if browser_name not in SUPPORTED_BROWSERS:
+        raise ValueError(f'unsupported browser: "{browser_name}"')
+    if profile is not None and _is_path(profile):
+        profile = os.path.expanduser(profile)
+    return browser_name, profile

yt_dlp/downloader/__init__.py

@@ -22,8 +22,10 @@ from .http import HttpFD
 from .rtmp import RtmpFD
 from .rtsp import RtspFD
 from .ism import IsmFD
+from .mhtml import MhtmlFD
 from .niconico import NiconicoDmcFD
-from .youtube_live_chat import YoutubeLiveChatReplayFD
+from .websocket import WebSocketFragmentFD
+from .youtube_live_chat import YoutubeLiveChatFD
 from .external import (
     get_external_downloader,
     FFmpegFD,
@@ -39,8 +41,11 @@ PROTOCOL_MAP = {
     'f4m': F4mFD,
     'http_dash_segments': DashSegmentsFD,
     'ism': IsmFD,
+    'mhtml': MhtmlFD,
     'niconico_dmc': NiconicoDmcFD,
-    'youtube_live_chat_replay': YoutubeLiveChatReplayFD,
+    'websocket_frag': WebSocketFragmentFD,
+    'youtube_live_chat': YoutubeLiveChatFD,
+    'youtube_live_chat_replay': YoutubeLiveChatFD,
 }
 
 
@@ -50,6 +55,7 @@ def shorten_protocol_name(proto, simplify=False):
         'rtmp_ffmpeg': 'rtmp_f',
         'http_dash_segments': 'dash',
         'niconico_dmc': 'dmc',
+        'websocket_frag': 'WSfrag',
     }
     if simplify:
         short_protocol_names.update({

yt_dlp/downloader/common.py

@@ -1,5 +1,6 @@
 from __future__ import division, unicode_literals
 
+import copy
 import os
 import re
 import sys
@@ -32,6 +33,7 @@ class FileDownloader(object):
     verbose:            Print additional info to stdout.
     quiet:              Do not print messages to stdout.
     ratelimit:          Download speed limit, in bytes/sec.
+    throttledratelimit: Assume the download is being throttled below this speed (bytes/sec)
     retries:            Number of times to retry for HTTP error 5xx
     buffersize:         Size of download buffer in bytes.
     noresizebuffer:     Do not automatically resize the download buffer.
@@ -359,7 +361,7 @@ class FileDownloader(object):
                     'filename': filename,
                     'status': 'finished',
                     'total_bytes': os.path.getsize(encodeFilename(filename)),
-                })
+                }, info_dict)
                 return True, False
 
         if subtitle is False:
@@ -387,7 +389,16 @@ class FileDownloader(object):
         """Real download process. Redefine in subclasses."""
         raise NotImplementedError('This method must be implemented by subclasses')
 
-    def _hook_progress(self, status):
+    def _hook_progress(self, status, info_dict):
+        if not self._progress_hooks:
+            return
+        info_dict = dict(info_dict)
+        for key in ('__original_infodict', '__postprocessors'):
+            info_dict.pop(key, None)
+        # youtube-dl passes the same status object to all the hooks.
+        # Some third party scripts seems to be relying on this.
+        # So keep this behavior if possible
+        status['info_dict'] = copy.deepcopy(info_dict)
         for ph in self._progress_hooks:
             ph(status)
 

yt_dlp/downloader/dash.py

@@ -1,21 +1,9 @@
 from __future__ import unicode_literals
 
-import errno
-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,
-)
+from ..utils import urljoin
 
 
 class DashSegmentsFD(FragmentFD):
@@ -41,10 +29,7 @@ class DashSegmentsFD(FragmentFD):
         if real_downloader:
             self._prepare_external_frag_download(ctx)
         else:
-            self._prepare_and_start_frag_download(ctx)
-
-        fragment_retries = self.params.get('fragment_retries', 0)
-        skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
+            self._prepare_and_start_frag_download(ctx, info_dict)
 
         fragments_to_download = []
         frag_index = 0
@@ -72,120 +57,6 @@ class DashSegmentsFD(FragmentFD):
             # TODO: Make progress updates work without hooking twice
             # for ph in self._progress_hooks:
             #     fd.add_progress_hook(ph)
-            success = fd.real_download(filename, info_copy)
-            if not success:
-                return False
-        else:
-            def download_fragment(fragment):
-                i = fragment['index']
-                frag_index = fragment['frag_index']
-                fragment_url = fragment['url']
+            return fd.real_download(filename, info_copy)
 
-                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
-                    ctx['dest_stream'].close()
-                    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):
-                fatal = frag_index == 1 or not skip_unavailable_fragments
-                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 EnvironmentError as ose:
-                        if ose.errno != errno.ENOENT:
-                            raise
-                        # FileNotFoundError
-                        if not fatal:
-                            self.report_skip_fragment(frag_index)
-                            return True
-                        else:
-                            ctx['dest_stream'].close()
-                            self.report_error(
-                                'fragment %s not found, unable to continue' % frag_index)
-                            return False
-                else:
-                    if not fatal:
-                        self.report_skip_fragment(frag_index)
-                        return True
-                    else:
-                        ctx['dest_stream'].close()
-                        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')
-                _download_fragment = lambda f: (f, download_fragment(f)[1])
-                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
-
-                for fragment, frag_index in map(lambda x: x.result(), futures):
-                    fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], frag_index)
-                    down, frag_sanitized = sanitize_open(fragment_filename, 'rb')
-                    fragment['fragment_filename_sanitized'] = frag_sanitized
-                    frag_content = down.read()
-                    down.close()
-                    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
+        return self.download_and_append_fragments(ctx, fragments_to_download, info_dict)

yt_dlp/downloader/external.py

@@ -67,7 +67,7 @@ class ExternalFD(FileDownloader):
                     'downloaded_bytes': fsize,
                     'total_bytes': fsize,
                 })
-            self._hook_progress(status)
+            self._hook_progress(status, info_dict)
             return True
         else:
             self.to_stderr('\n')
@@ -280,6 +280,8 @@ class Aria2cFD(ExternalFD):
                '--file-allocation=none', '-x16', '-j16', '-s16']
         if 'fragments' in info_dict:
             cmd += ['--allow-overwrite=true', '--allow-piece-length-change=true']
+        else:
+            cmd += ['--min-split-size', '1M']
 
         if info_dict.get('http_headers') is not None:
             for key, val in info_dict['http_headers'].items():
@@ -345,6 +347,10 @@ class FFmpegFD(ExternalFD):
         # TODO: Fix path for ffmpeg
         return FFmpegPostProcessor().available
 
+    def on_process_started(self, proc, stdin):
+        """ Override this in subclasses  """
+        pass
+
     def _call_downloader(self, tmpfilename, info_dict):
         urls = [f['url'] for f in info_dict.get('requested_formats', [])] or [info_dict['url']]
         ffpp = FFmpegPostProcessor(downloader=self)
@@ -371,8 +377,6 @@ class FFmpegFD(ExternalFD):
             # http://trac.ffmpeg.org/ticket/6125#comment:10
             args += ['-seekable', '1' if seekable else '0']
 
-        args += self._configuration_args()
-
         # start_time = info_dict.get('start_time') or 0
         # if start_time:
         #     args += ['-ss', compat_str(start_time)]
@@ -440,7 +444,8 @@ class FFmpegFD(ExternalFD):
 
         for url in urls:
             args += ['-i', url]
-        args += ['-c', 'copy']
+
+        args += self._configuration_args() + ['-c', 'copy']
         if info_dict.get('requested_formats'):
             for (i, fmt) in enumerate(info_dict['requested_formats']):
                 if fmt.get('acodec') != 'none':
@@ -472,6 +477,8 @@ class FFmpegFD(ExternalFD):
         self._debug_cmd(args)
 
         proc = subprocess.Popen(args, stdin=subprocess.PIPE, env=env)
+        if url in ('-', 'pipe:'):
+            self.on_process_started(proc, proc.stdin)
         try:
             retval = proc.wait()
         except BaseException as e:
@@ -480,7 +487,7 @@ class FFmpegFD(ExternalFD):
             # produces a file that is playable (this is mostly useful for live
             # streams). Note that Windows is not affected and produces playable
             # files (see https://github.com/ytdl-org/youtube-dl/issues/8300).
-            if isinstance(e, KeyboardInterrupt) and sys.platform != 'win32':
+            if isinstance(e, KeyboardInterrupt) and sys.platform != 'win32' and url not in ('-', 'pipe:'):
                 process_communicate_or_kill(proc, b'q')
             else:
                 proc.kill()

yt_dlp/downloader/f4m.py

@@ -380,7 +380,7 @@ class F4mFD(FragmentFD):
 
         base_url_parsed = compat_urllib_parse_urlparse(base_url)
 
-        self._start_frag_download(ctx)
+        self._start_frag_download(ctx, info_dict)
 
         frag_index = 0
         while fragments_list:
@@ -434,6 +434,6 @@ class F4mFD(FragmentFD):
                     msg = 'Missed %d fragments' % (fragments_list[0][1] - (frag_i + 1))
                     self.report_warning(msg)
 
-        self._finish_frag_download(ctx)
+        self._finish_frag_download(ctx, info_dict)
 
         return True

yt_dlp/downloader/fragment.py

@@ -4,9 +4,26 @@ import os
 import time
 import json
 
+try:
+    from Crypto.Cipher import AES
+    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 .common import FileDownloader
 from .http import HttpFD
+from ..compat import (
+    compat_urllib_error,
+    compat_struct_pack,
+)
 from ..utils import (
+    DownloadError,
     error_to_compat_str,
     encodeFilename,
     sanitize_open,
@@ -56,7 +73,7 @@ class FragmentFD(FileDownloader):
 
     def report_retry_fragment(self, err, frag_index, count, retries):
         self.to_screen(
-            '[download] Got server HTTP error: %s. Retrying fragment %d (attempt %d of %s) ...'
+            '\r[download] Got server HTTP error: %s. Retrying fragment %d (attempt %d of %s) ...'
             % (error_to_compat_str(err), frag_index, count, self.format_retries(retries)))
 
     def report_skip_fragment(self, frag_index):
@@ -66,9 +83,9 @@ class FragmentFD(FileDownloader):
         headers = info_dict.get('http_headers')
         return sanitized_Request(url, None, headers) if headers else url
 
-    def _prepare_and_start_frag_download(self, ctx):
+    def _prepare_and_start_frag_download(self, ctx, info_dict):
         self._prepare_frag_download(ctx)
-        self._start_frag_download(ctx)
+        self._start_frag_download(ctx, info_dict)
 
     def __do_ytdl_file(self, ctx):
         return not ctx['live'] and not ctx['tmpfilename'] == '-' and not self.params.get('_no_ytdl_file')
@@ -112,11 +129,15 @@ class FragmentFD(FileDownloader):
             return False, None
         if fragment_info_dict.get('filetime'):
             ctx['fragment_filetime'] = fragment_info_dict.get('filetime')
-        down, frag_sanitized = sanitize_open(fragment_filename, 'rb')
+        ctx['fragment_filename_sanitized'] = fragment_filename
+        return True, self._read_fragment(ctx)
+
+    def _read_fragment(self, ctx):
+        down, frag_sanitized = sanitize_open(ctx['fragment_filename_sanitized'], 'rb')
         ctx['fragment_filename_sanitized'] = frag_sanitized
         frag_content = down.read()
         down.close()
-        return True, frag_content
+        return frag_content
 
     def _append_fragment(self, ctx, frag_content):
         try:
@@ -198,7 +219,7 @@ class FragmentFD(FileDownloader):
             'complete_frags_downloaded_bytes': resume_len,
         })
 
-    def _start_frag_download(self, ctx):
+    def _start_frag_download(self, ctx, info_dict):
         resume_len = ctx['complete_frags_downloaded_bytes']
         total_frags = ctx['total_frags']
         # This dict stores the download progress, it's updated by the progress
@@ -227,6 +248,7 @@ class FragmentFD(FileDownloader):
             time_now = time.time()
             state['elapsed'] = time_now - start
             frag_total_bytes = s.get('total_bytes') or 0
+            s['fragment_info_dict'] = s.pop('info_dict', {})
             if not ctx['live']:
                 estimated_size = (
                     (ctx['complete_frags_downloaded_bytes'] + frag_total_bytes)
@@ -249,13 +271,13 @@ class FragmentFD(FileDownloader):
                 state['speed'] = s.get('speed') or ctx.get('speed')
                 ctx['speed'] = state['speed']
                 ctx['prev_frag_downloaded_bytes'] = frag_downloaded_bytes
-            self._hook_progress(state)
+            self._hook_progress(state, info_dict)
 
         ctx['dl'].add_progress_hook(frag_progress_hook)
 
         return start
 
-    def _finish_frag_download(self, ctx):
+    def _finish_frag_download(self, ctx, info_dict):
         ctx['dest_stream'].close()
         if self.__do_ytdl_file(ctx):
             ytdl_filename = encodeFilename(self.ytdl_filename(ctx['filename']))
@@ -282,7 +304,7 @@ class FragmentFD(FileDownloader):
             'filename': ctx['filename'],
             'status': 'finished',
             'elapsed': elapsed,
-        })
+        }, info_dict)
 
     def _prepare_external_frag_download(self, ctx):
         if 'live' not in ctx:
@@ -304,3 +326,101 @@ class FragmentFD(FileDownloader):
             'tmpfilename': tmpfilename,
             'fragment_index': 0,
         })
+
+    def download_and_append_fragments(self, ctx, fragments, info_dict, pack_func=None):
+        fragment_retries = self.params.get('fragment_retries', 0)
+        is_fatal = (lambda idx: idx == 0) if self.params.get('skip_unavailable_fragments', True) else (lambda _: True)
+        if not pack_func:
+            pack_func = lambda frag_content, _: frag_content
+
+        def download_fragment(fragment, ctx):
+            frag_index = ctx['fragment_index'] = fragment['frag_index']
+            headers = info_dict.get('http_headers', {})
+            byte_range = fragment.get('byte_range')
+            if byte_range:
+                headers['Range'] = 'bytes=%d-%d' % (byte_range['start'], byte_range['end'] - 1)
+
+            # Never skip the first fragment
+            fatal = is_fatal(fragment.get('index') or (frag_index - 1))
+            count, frag_content = 0, None
+            while count <= fragment_retries:
+                try:
+                    success, frag_content = self._download_fragment(ctx, fragment['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)
+                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
+                ctx['dest_stream'].close()
+                self.report_error('Giving up after %s fragment retries' % fragment_retries)
+                return False, frag_index
+            return frag_content, frag_index
+
+        def decrypt_fragment(fragment, frag_content):
+            decrypt_info = fragment.get('decrypt_info')
+            if not decrypt_info or decrypt_info['METHOD'] != 'AES-128':
+                return frag_content
+            iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', fragment['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 self.params.get('test', False):
+                return frag_content
+            return AES.new(decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
+
+        def append_fragment(frag_content, frag_index, ctx):
+            if not frag_content:
+                if not is_fatal(frag_index - 1):
+                    self.report_skip_fragment(frag_index)
+                    return True
+                else:
+                    ctx['dest_stream'].close()
+                    self.report_error(
+                        'fragment %s not found, unable to continue' % frag_index)
+                    return False
+            self._append_fragment(ctx, pack_func(frag_content, frag_index))
+            return True
+
+        max_workers = self.params.get('concurrent_fragment_downloads', 1)
+        if can_threaded_download and max_workers > 1:
+
+            def _download_fragment(fragment):
+                ctx_copy = ctx.copy()
+                frag_content, frag_index = download_fragment(fragment, ctx_copy)
+                return fragment, frag_content, frag_index, ctx_copy.get('fragment_filename_sanitized')
+
+            self.report_warning('The download speed shown is only of one thread. This is a known issue and patches are welcome')
+            with concurrent.futures.ThreadPoolExecutor(max_workers) as pool:
+                for fragment, frag_content, frag_index, frag_filename in pool.map(_download_fragment, fragments):
+                    ctx['fragment_filename_sanitized'] = frag_filename
+                    ctx['fragment_index'] = frag_index
+                    result = append_fragment(decrypt_fragment(fragment, frag_content), frag_index, ctx)
+                    if not result:
+                        return False
+        else:
+            for fragment in fragments:
+                frag_content, frag_index = download_fragment(fragment, ctx)
+                result = append_fragment(decrypt_fragment(fragment, frag_content), frag_index, ctx)
+                if not result:
+                    return False
+
+        self._finish_frag_download(ctx, info_dict)
+        return True

yt_dlp/downloader/hls.py

@@ -1,32 +1,18 @@
 from __future__ import unicode_literals
 
-import errno
 import re
 import io
 import binascii
-try:
-    from Crypto.Cipher import AES
-    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
+from .fragment import FragmentFD, can_decrypt_frag
 from .external import FFmpegFD
 
 from ..compat import (
-    compat_urllib_error,
     compat_urlparse,
-    compat_struct_pack,
 )
 from ..utils import (
     parse_m3u8_attributes,
-    sanitize_open,
     update_url_query,
     bug_reports_message,
 )
@@ -147,14 +133,10 @@ class HlsFD(FragmentFD):
         if real_downloader:
             self._prepare_external_frag_download(ctx)
         else:
-            self._prepare_and_start_frag_download(ctx)
+            self._prepare_and_start_frag_download(ctx, info_dict)
 
         extra_state = ctx.setdefault('extra_state', {})
 
-        fragment_retries = self.params.get('fragment_retries', 0)
-        skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
-        test = self.params.get('test', False)
-
         format_index = info_dict.get('format_index')
         extra_query = None
         extra_param_to_segment_url = info_dict.get('extra_param_to_segment_url')
@@ -258,7 +240,7 @@ class HlsFD(FragmentFD):
                 media_sequence += 1
 
         # We only download the first fragment during the test
-        if test:
+        if self.params.get('test', False):
             fragments = [fragments[0] if fragments else None]
 
         if real_downloader:
@@ -268,195 +250,75 @@ class HlsFD(FragmentFD):
             # TODO: Make progress updates work without hooking twice
             # for ph in self._progress_hooks:
             #     fd.add_progress_hook(ph)
-            success = fd.real_download(filename, info_copy)
-            if not success:
-                return False
-        else:
-            def decrypt_fragment(fragment, frag_content):
-                decrypt_info = fragment['decrypt_info']
-                if decrypt_info['METHOD'] != 'AES-128':
-                    return frag_content
-                iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', fragment['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 test:
-                    return frag_content
-                return AES.new(decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
+            return fd.real_download(filename, info_copy)
 
-            def download_fragment(fragment):
-                frag_index = fragment['frag_index']
-                frag_url = fragment['url']
-                byte_range = fragment['byte_range']
+        if is_webvtt:
+            def pack_fragment(frag_content, frag_index):
+                output = io.StringIO()
+                adjust = 0
+                for block in webvtt.parse_fragment(frag_content):
+                    if isinstance(block, webvtt.CueBlock):
+                        block.start += adjust
+                        block.end += adjust
 
-                ctx['fragment_index'] = frag_index
+                        dedup_window = extra_state.setdefault('webvtt_dedup_window', [])
+                        cue = block.as_json
 
-                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:
-                    ctx['dest_stream'].close()
-                    self.report_error('Giving up after %s fragment retries' % fragment_retries)
-                    return False, frag_index
-
-                return decrypt_fragment(fragment, frag_content), frag_index
-
-            pack_fragment = lambda frag_content, _: frag_content
-
-            if is_webvtt:
-                def pack_fragment(frag_content, frag_index):
-                    output = io.StringIO()
-                    adjust = 0
-                    for block in webvtt.parse_fragment(frag_content):
-                        if isinstance(block, webvtt.CueBlock):
-                            block.start += adjust
-                            block.end += adjust
-
-                            dedup_window = extra_state.setdefault('webvtt_dedup_window', [])
-                            cue = block.as_json
-
-                            # skip the cue if an identical one appears
-                            # in the window of potential duplicates
-                            # and prune the window of unviable candidates
-                            i = 0
-                            skip = True
-                            while i < len(dedup_window):
-                                window_cue = dedup_window[i]
-                                if window_cue == cue:
-                                    break
-                                if window_cue['end'] >= cue['start']:
-                                    i += 1
-                                    continue
-                                del dedup_window[i]
-                            else:
-                                skip = False
-
-                            if skip:
+                        # skip the cue if an identical one appears
+                        # in the window of potential duplicates
+                        # and prune the window of unviable candidates
+                        i = 0
+                        skip = True
+                        while i < len(dedup_window):
+                            window_cue = dedup_window[i]
+                            if window_cue == cue:
+                                break
+                            if window_cue['end'] >= cue['start']:
+                                i += 1
                                 continue
-
-                            # add the cue to the window
-                            dedup_window.append(cue)
-                        elif isinstance(block, webvtt.Magic):
-                            # take care of MPEG PES timestamp overflow
-                            if block.mpegts is None:
-                                block.mpegts = 0
-                            extra_state.setdefault('webvtt_mpegts_adjust', 0)
-                            block.mpegts += extra_state['webvtt_mpegts_adjust'] << 33
-                            if block.mpegts < extra_state.get('webvtt_mpegts_last', 0):
-                                extra_state['webvtt_mpegts_adjust'] += 1
-                                block.mpegts += 1 << 33
-                            extra_state['webvtt_mpegts_last'] = block.mpegts
-
-                            if frag_index == 1:
-                                extra_state['webvtt_mpegts'] = block.mpegts or 0
-                                extra_state['webvtt_local'] = block.local or 0
-                                # XXX: block.local = block.mpegts = None ?
-                            else:
-                                if block.mpegts is not None and block.local is not None:
-                                    adjust = (
-                                        (block.mpegts - extra_state.get('webvtt_mpegts', 0))
-                                        - (block.local - extra_state.get('webvtt_local', 0))
-                                    )
-                                continue
-                        elif isinstance(block, webvtt.HeaderBlock):
-                            if frag_index != 1:
-                                # XXX: this should probably be silent as well
-                                # or verify that all segments contain the same data
-                                self.report_warning(bug_reports_message(
-                                    'Discarding a %s block found in the middle of the stream; '
-                                    'if the subtitles display incorrectly,'
-                                    % (type(block).__name__)))
-                                continue
-                        block.write_into(output)
-
-                    return output.getvalue().encode('utf-8')
-
-            def append_fragment(frag_content, frag_index):
-                fatal = frag_index == 1 or not skip_unavailable_fragments
-                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()
-                        frag_content = pack_fragment(frag_content, frag_index)
-                        self._append_fragment(ctx, frag_content)
-                        return True
-                    except EnvironmentError as ose:
-                        if ose.errno != errno.ENOENT:
-                            raise
-                        # FileNotFoundError
-                        if not fatal:
-                            self.report_skip_fragment(frag_index)
-                            return True
+                            del dedup_window[i]
                         else:
-                            ctx['dest_stream'].close()
-                            self.report_error(
-                                'fragment %s not found, unable to continue' % frag_index)
-                            return False
-                else:
-                    if not fatal:
-                        self.report_skip_fragment(frag_index)
-                        return True
-                    else:
-                        ctx['dest_stream'].close()
-                        self.report_error(
-                            'fragment %s not found, unable to continue' % frag_index)
-                        return False
+                            skip = 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')
-                _download_fragment = lambda f: (f, download_fragment(f)[1])
-                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
+                        if skip:
+                            continue
 
-                for fragment, frag_index in map(lambda x: x.result(), futures):
-                    fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], frag_index)
-                    down, frag_sanitized = sanitize_open(fragment_filename, 'rb')
-                    fragment['fragment_filename_sanitized'] = frag_sanitized
-                    frag_content = down.read()
-                    down.close()
-                    result = append_fragment(decrypt_fragment(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
+                        # add the cue to the window
+                        dedup_window.append(cue)
+                    elif isinstance(block, webvtt.Magic):
+                        # take care of MPEG PES timestamp overflow
+                        if block.mpegts is None:
+                            block.mpegts = 0
+                        extra_state.setdefault('webvtt_mpegts_adjust', 0)
+                        block.mpegts += extra_state['webvtt_mpegts_adjust'] << 33
+                        if block.mpegts < extra_state.get('webvtt_mpegts_last', 0):
+                            extra_state['webvtt_mpegts_adjust'] += 1
+                            block.mpegts += 1 << 33
+                        extra_state['webvtt_mpegts_last'] = block.mpegts
 
-            self._finish_frag_download(ctx)
-        return True
+                        if frag_index == 1:
+                            extra_state['webvtt_mpegts'] = block.mpegts or 0
+                            extra_state['webvtt_local'] = block.local or 0
+                            # XXX: block.local = block.mpegts = None ?
+                        else:
+                            if block.mpegts is not None and block.local is not None:
+                                adjust = (
+                                    (block.mpegts - extra_state.get('webvtt_mpegts', 0))
+                                    - (block.local - extra_state.get('webvtt_local', 0))
+                                )
+                            continue
+                    elif isinstance(block, webvtt.HeaderBlock):
+                        if frag_index != 1:
+                            # XXX: this should probably be silent as well
+                            # or verify that all segments contain the same data
+                            self.report_warning(bug_reports_message(
+                                'Discarding a %s block found in the middle of the stream; '
+                                'if the subtitles display incorrectly,'
+                                % (type(block).__name__)))
+                            continue
+                    block.write_into(output)
+
+                return output.getvalue().encode('utf-8')
+        else:
+            pack_fragment = None
+        return self.download_and_append_fragments(ctx, fragments, info_dict, pack_fragment)

yt_dlp/downloader/http.py

@@ -18,6 +18,7 @@ from ..utils import (
     int_or_none,
     sanitize_open,
     sanitized_Request,
+    ThrottledDownload,
     write_xattr,
     XAttrMetadataError,
     XAttrUnavailableError,
@@ -176,7 +177,7 @@ class HttpFD(FileDownloader):
                                 'status': 'finished',
                                 'downloaded_bytes': ctx.resume_len,
                                 'total_bytes': ctx.resume_len,
-                            })
+                            }, info_dict)
                             raise SucceedDownload()
                         else:
                             # The length does not match, we start the download over
@@ -223,6 +224,7 @@ class HttpFD(FileDownloader):
             # measure time over whole while-loop, so slow_down() and best_block_size() work together properly
             now = None  # needed for slow_down() in the first loop run
             before = start  # start measuring
+            throttle_start = None
 
             def retry(e):
                 to_stdout = ctx.tmpfilename == '-'
@@ -308,11 +310,23 @@ class HttpFD(FileDownloader):
                     'eta': eta,
                     'speed': speed,
                     'elapsed': now - ctx.start_time,
-                })
+                }, info_dict)
 
                 if data_len is not None and byte_counter == data_len:
                     break
 
+                if speed and speed < (self.params.get('throttledratelimit') or 0):
+                    # The speed must stay below the limit for 3 seconds
+                    # This prevents raising error when the speed temporarily goes down
+                    if throttle_start is None:
+                        throttle_start = now
+                    elif now - throttle_start > 3:
+                        if ctx.stream is not None and ctx.tmpfilename != '-':
+                            ctx.stream.close()
+                        raise ThrottledDownload()
+                else:
+                    throttle_start = None
+
             if not is_test and ctx.chunk_size and ctx.data_len is not None and byte_counter < ctx.data_len:
                 ctx.resume_len = byte_counter
                 # ctx.block_size = block_size
@@ -343,7 +357,7 @@ class HttpFD(FileDownloader):
                 'filename': ctx.filename,
                 'status': 'finished',
                 'elapsed': time.time() - ctx.start_time,
-            })
+            }, info_dict)
 
             return True
 

yt_dlp/downloader/ism.py

@@ -246,7 +246,7 @@ class IsmFD(FragmentFD):
             'total_frags': len(segments),
         }
 
-        self._prepare_and_start_frag_download(ctx)
+        self._prepare_and_start_frag_download(ctx, info_dict)
 
         extra_state = ctx.setdefault('extra_state', {
             'ism_track_written': False,
@@ -284,6 +284,6 @@ class IsmFD(FragmentFD):
                 self.report_error('giving up after %s fragment retries' % fragment_retries)
                 return False
 
-        self._finish_frag_download(ctx)
+        self._finish_frag_download(ctx, info_dict)
 
         return True

yt_dlp/downloader/mhtml.py

@@ -0,0 +1,202 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import io
+import quopri
+import re
+import uuid
+
+from .fragment import FragmentFD
+from ..utils import (
+    escapeHTML,
+    formatSeconds,
+    srt_subtitles_timecode,
+    urljoin,
+)
+from ..version import __version__ as YT_DLP_VERSION
+
+
+class MhtmlFD(FragmentFD):
+    FD_NAME = 'mhtml'
+
+    _STYLESHEET = """\
+html, body {
+    margin: 0;
+    padding: 0;
+    height: 100vh;
+}
+
+html {
+    overflow-y: scroll;
+    scroll-snap-type: y mandatory;
+}
+
+body {
+    scroll-snap-type: y mandatory;
+    display: flex;
+    flex-flow: column;
+}
+
+body > figure {
+    max-width: 100vw;
+    max-height: 100vh;
+    scroll-snap-align: center;
+}
+
+body > figure > figcaption {
+    text-align: center;
+    height: 2.5em;
+}
+
+body > figure > img {
+    display: block;
+    margin: auto;
+    max-width: 100%;
+    max-height: calc(100vh - 5em);
+}
+"""
+    _STYLESHEET = re.sub(r'\s+', ' ', _STYLESHEET)
+    _STYLESHEET = re.sub(r'\B \B|(?<=[\w\-]) (?=[^\w\-])|(?<=[^\w\-]) (?=[\w\-])', '', _STYLESHEET)
+
+    @staticmethod
+    def _escape_mime(s):
+        return '=?utf-8?Q?' + (b''.join(
+            bytes((b,)) if b >= 0x20 else b'=%02X' % b
+            for b in quopri.encodestring(s.encode('utf-8'), header=True)
+        )).decode('us-ascii') + '?='
+
+    def _gen_cid(self, i, fragment, frag_boundary):
+        return '%u.%s@yt-dlp.github.io.invalid' % (i, frag_boundary)
+
+    def _gen_stub(self, *, fragments, frag_boundary, title):
+        output = io.StringIO()
+
+        output.write((
+            '<!DOCTYPE html>'
+            '<html>'
+            '<head>'
+            ''  '<meta name="generator" content="yt-dlp {version}">'
+            ''  '<title>{title}</title>'
+            ''  '<style>{styles}</style>'
+            '<body>'
+        ).format(
+            version=escapeHTML(YT_DLP_VERSION),
+            styles=self._STYLESHEET,
+            title=escapeHTML(title)
+        ))
+
+        t0 = 0
+        for i, frag in enumerate(fragments):
+            output.write('<figure>')
+            try:
+                t1 = t0 + frag['duration']
+                output.write((
+                    '<figcaption>Slide #{num}: {t0} – {t1} (duration: {duration})</figcaption>'
+                ).format(
+                    num=i + 1,
+                    t0=srt_subtitles_timecode(t0),
+                    t1=srt_subtitles_timecode(t1),
+                    duration=formatSeconds(frag['duration'], msec=True)
+                ))
+            except (KeyError, ValueError, TypeError):
+                t1 = None
+                output.write((
+                    '<figcaption>Slide #{num}</figcaption>'
+                ).format(num=i + 1))
+            output.write('<img src="cid:{cid}">'.format(
+                cid=self._gen_cid(i, frag, frag_boundary)))
+            output.write('</figure>')
+            t0 = t1
+
+        return output.getvalue()
+
+    def real_download(self, filename, info_dict):
+        fragment_base_url = info_dict.get('fragment_base_url')
+        fragments = info_dict['fragments'][:1] if self.params.get(
+            'test', False) else info_dict['fragments']
+        title = info_dict['title']
+        origin = info_dict['webpage_url']
+
+        ctx = {
+            'filename': filename,
+            'total_frags': len(fragments),
+        }
+
+        self._prepare_and_start_frag_download(ctx, info_dict)
+
+        extra_state = ctx.setdefault('extra_state', {
+            'header_written': False,
+            'mime_boundary': str(uuid.uuid4()).replace('-', ''),
+        })
+
+        frag_boundary = extra_state['mime_boundary']
+
+        if not extra_state['header_written']:
+            stub = self._gen_stub(
+                fragments=fragments,
+                frag_boundary=frag_boundary,
+                title=title
+            )
+
+            ctx['dest_stream'].write((
+                'MIME-Version: 1.0\r\n'
+                'From: <nowhere@yt-dlp.github.io.invalid>\r\n'
+                'To: <nowhere@yt-dlp.github.io.invalid>\r\n'
+                'Subject: {title}\r\n'
+                'Content-type: multipart/related; '
+                ''  'boundary="{boundary}"; '
+                ''  'type="text/html"\r\n'
+                'X.yt-dlp.Origin: {origin}\r\n'
+                '\r\n'
+                '--{boundary}\r\n'
+                'Content-Type: text/html; charset=utf-8\r\n'
+                'Content-Length: {length}\r\n'
+                '\r\n'
+                '{stub}\r\n'
+            ).format(
+                origin=origin,
+                boundary=frag_boundary,
+                length=len(stub),
+                title=self._escape_mime(title),
+                stub=stub
+            ).encode('utf-8'))
+            extra_state['header_written'] = True
+
+        for i, fragment in enumerate(fragments):
+            if (i + 1) <= ctx['fragment_index']:
+                continue
+
+            fragment_url = urljoin(fragment_base_url, fragment['path'])
+            success, frag_content = self._download_fragment(ctx, fragment_url, info_dict)
+            if not success:
+                continue
+
+            mime_type = b'image/jpeg'
+            if frag_content.startswith(b'\x89PNG\r\n\x1a\n'):
+                mime_type = b'image/png'
+            if frag_content.startswith((b'GIF87a', b'GIF89a')):
+                mime_type = b'image/gif'
+            if frag_content.startswith(b'RIFF') and frag_content[8:12] == 'WEBP':
+                mime_type = b'image/webp'
+
+            frag_header = io.BytesIO()
+            frag_header.write(
+                b'--%b\r\n' % frag_boundary.encode('us-ascii'))
+            frag_header.write(
+                b'Content-ID: <%b>\r\n' % self._gen_cid(i, fragment, frag_boundary).encode('us-ascii'))
+            frag_header.write(
+                b'Content-type: %b\r\n' % mime_type)
+            frag_header.write(
+                b'Content-length: %u\r\n' % len(frag_content))
+            frag_header.write(
+                b'Content-location: %b\r\n' % fragment_url.encode('us-ascii'))
+            frag_header.write(
+                b'X.yt-dlp.Duration: %f\r\n' % fragment['duration'])
+            frag_header.write(b'\r\n')
+            self._append_fragment(
+                ctx, frag_header.getvalue() + frag_content + b'\r\n')
+
+        ctx['dest_stream'].write(
+            b'--%b--\r\n\r\n' % frag_boundary.encode('us-ascii'))
+        self._finish_frag_download(ctx, info_dict)
+        return True

yt_dlp/downloader/rtmp.py

@@ -66,7 +66,7 @@ class RtmpFD(FileDownloader):
                             'eta': eta,
                             'elapsed': time_now - start,
                             'speed': speed,
-                        })
+                        }, info_dict)
                         cursor_in_new_line = False
                     else:
                         # no percent for live streams
@@ -82,7 +82,7 @@ class RtmpFD(FileDownloader):
                                 'status': 'downloading',
                                 'elapsed': time_now - start,
                                 'speed': speed,
-                            })
+                            }, info_dict)
                             cursor_in_new_line = False
                         elif self.params.get('verbose', False):
                             if not cursor_in_new_line:
@@ -208,7 +208,7 @@ class RtmpFD(FileDownloader):
                 'filename': filename,
                 'status': 'finished',
                 'elapsed': time.time() - started,
-            })
+            }, info_dict)
             return True
         else:
             self.to_stderr('\n')

yt_dlp/downloader/rtsp.py

@@ -39,7 +39,7 @@ class RtspFD(FileDownloader):
                 'total_bytes': fsize,
                 'filename': filename,
                 'status': 'finished',
-            })
+            }, info_dict)
             return True
         else:
             self.to_stderr('\n')

yt_dlp/downloader/websocket.py

@@ -0,0 +1,59 @@
+import os
+import signal
+import asyncio
+import threading
+
+try:
+    import websockets
+    has_websockets = True
+except ImportError:
+    has_websockets = False
+
+from .common import FileDownloader
+from .external import FFmpegFD
+
+
+class FFmpegSinkFD(FileDownloader):
+    """ A sink to ffmpeg for downloading fragments in any form """
+
+    def real_download(self, filename, info_dict):
+        info_copy = info_dict.copy()
+        info_copy['url'] = '-'
+
+        async def call_conn(proc, stdin):
+            try:
+                await self.real_connection(stdin, info_dict)
+            except (BrokenPipeError, OSError):
+                pass
+            finally:
+                try:
+                    stdin.flush()
+                    stdin.close()
+                except OSError:
+                    pass
+                os.kill(os.getpid(), signal.SIGINT)
+
+        class FFmpegStdinFD(FFmpegFD):
+            @classmethod
+            def get_basename(cls):
+                return FFmpegFD.get_basename()
+
+            def on_process_started(self, proc, stdin):
+                thread = threading.Thread(target=asyncio.run, daemon=True, args=(call_conn(proc, stdin), ))
+                thread.start()
+
+        return FFmpegStdinFD(self.ydl, self.params or {}).download(filename, info_copy)
+
+    async def real_connection(self, sink, info_dict):
+        """ Override this in subclasses """
+        raise NotImplementedError('This method must be implemented by subclasses')
+
+
+class WebSocketFragmentFD(FFmpegSinkFD):
+    async def real_connection(self, sink, info_dict):
+        async with websockets.connect(info_dict['url'], extra_headers=info_dict.get('http_headers', {})) as ws:
+            while True:
+                recv = await ws.recv()
+                if isinstance(recv, str):
+                    recv = recv.encode('utf8')
+                sink.write(recv)

yt_dlp/downloader/youtube_live_chat.py

@@ -1,20 +1,23 @@
 from __future__ import division, unicode_literals
 
 import json
+import time
 
 from .fragment import FragmentFD
 from ..compat import compat_urllib_error
 from ..utils import (
     try_get,
+    dict_get,
+    int_or_none,
     RegexNotFoundError,
 )
 from ..extractor.youtube import YoutubeBaseInfoExtractor as YT_BaseIE
 
 
-class YoutubeLiveChatReplayFD(FragmentFD):
-    """ Downloads YouTube live chat replays fragment by fragment """
+class YoutubeLiveChatFD(FragmentFD):
+    """ Downloads YouTube live chats fragment by fragment """
 
-    FD_NAME = 'youtube_live_chat_replay'
+    FD_NAME = 'youtube_live_chat'
 
     def real_download(self, filename, info_dict):
         video_id = info_dict['video_id']
@@ -31,6 +34,8 @@ class YoutubeLiveChatReplayFD(FragmentFD):
 
         ie = YT_BaseIE(self.ydl)
 
+        start_time = int(time.time() * 1000)
+
         def dl_fragment(url, data=None, headers=None):
             http_headers = info_dict.get('http_headers', {})
             if headers:
@@ -38,15 +43,80 @@ class YoutubeLiveChatReplayFD(FragmentFD):
                 http_headers.update(headers)
             return self._download_fragment(ctx, url, info_dict, http_headers, data)
 
-        def download_and_parse_fragment(url, frag_index, request_data):
+        def parse_actions_replay(live_chat_continuation):
+            offset = continuation_id = click_tracking_params = None
+            processed_fragment = bytearray()
+            for action in live_chat_continuation.get('actions', []):
+                if 'replayChatItemAction' in action:
+                    replay_chat_item_action = action['replayChatItemAction']
+                    offset = int(replay_chat_item_action['videoOffsetTimeMsec'])
+                processed_fragment.extend(
+                    json.dumps(action, ensure_ascii=False).encode('utf-8') + b'\n')
+            if offset is not None:
+                continuation = try_get(
+                    live_chat_continuation,
+                    lambda x: x['continuations'][0]['liveChatReplayContinuationData'], dict)
+                if continuation:
+                    continuation_id = continuation.get('continuation')
+                    click_tracking_params = continuation.get('clickTrackingParams')
+            self._append_fragment(ctx, processed_fragment)
+            return continuation_id, offset, click_tracking_params
+
+        def try_refresh_replay_beginning(live_chat_continuation):
+            # choose the second option that contains the unfiltered live chat replay
+            refresh_continuation = try_get(
+                live_chat_continuation,
+                lambda x: x['header']['liveChatHeaderRenderer']['viewSelector']['sortFilterSubMenuRenderer']['subMenuItems'][1]['continuation']['reloadContinuationData'], dict)
+            if refresh_continuation:
+                # no data yet but required to call _append_fragment
+                self._append_fragment(ctx, b'')
+                refresh_continuation_id = refresh_continuation.get('continuation')
+                offset = 0
+                click_tracking_params = refresh_continuation.get('trackingParams')
+                return refresh_continuation_id, offset, click_tracking_params
+            return parse_actions_replay(live_chat_continuation)
+
+        live_offset = 0
+
+        def parse_actions_live(live_chat_continuation):
+            nonlocal live_offset
+            continuation_id = click_tracking_params = None
+            processed_fragment = bytearray()
+            for action in live_chat_continuation.get('actions', []):
+                timestamp = self.parse_live_timestamp(action)
+                if timestamp is not None:
+                    live_offset = timestamp - start_time
+                # compatibility with replay format
+                pseudo_action = {
+                    'replayChatItemAction': {'actions': [action]},
+                    'videoOffsetTimeMsec': str(live_offset),
+                    'isLive': True,
+                }
+                processed_fragment.extend(
+                    json.dumps(pseudo_action, ensure_ascii=False).encode('utf-8') + b'\n')
+            continuation_data_getters = [
+                lambda x: x['continuations'][0]['invalidationContinuationData'],
+                lambda x: x['continuations'][0]['timedContinuationData'],
+            ]
+            continuation_data = try_get(live_chat_continuation, continuation_data_getters, dict)
+            if continuation_data:
+                continuation_id = continuation_data.get('continuation')
+                click_tracking_params = continuation_data.get('clickTrackingParams')
+                timeout_ms = int_or_none(continuation_data.get('timeoutMs'))
+                if timeout_ms is not None:
+                    time.sleep(timeout_ms / 1000)
+            self._append_fragment(ctx, processed_fragment)
+            return continuation_id, live_offset, click_tracking_params
+
+        def download_and_parse_fragment(url, frag_index, request_data=None, headers=None):
             count = 0
             while count <= fragment_retries:
                 try:
-                    success, raw_fragment = dl_fragment(url, request_data, {'content-type': 'application/json'})
+                    success, raw_fragment = dl_fragment(url, request_data, headers)
                     if not success:
-                        return False, None, None
+                        return False, None, None, None
                     try:
-                        data = ie._extract_yt_initial_data(video_id, raw_fragment.decode('utf-8', 'replace'))
+                        data = ie.extract_yt_initial_data(video_id, raw_fragment.decode('utf-8', 'replace'))
                     except RegexNotFoundError:
                         data = None
                     if not data:
@@ -54,36 +124,29 @@ class YoutubeLiveChatReplayFD(FragmentFD):
                     live_chat_continuation = try_get(
                         data,
                         lambda x: x['continuationContents']['liveChatContinuation'], dict) or {}
-                    offset = continuation_id = None
-                    processed_fragment = bytearray()
-                    for action in live_chat_continuation.get('actions', []):
-                        if 'replayChatItemAction' in action:
-                            replay_chat_item_action = action['replayChatItemAction']
-                            offset = int(replay_chat_item_action['videoOffsetTimeMsec'])
-                        processed_fragment.extend(
-                            json.dumps(action, ensure_ascii=False).encode('utf-8') + b'\n')
-                    if offset is not None:
-                        continuation_id = try_get(
-                            live_chat_continuation,
-                            lambda x: x['continuations'][0]['liveChatReplayContinuationData']['continuation'])
-                    self._append_fragment(ctx, processed_fragment)
-
-                    return True, continuation_id, offset
+                    if info_dict['protocol'] == 'youtube_live_chat_replay':
+                        if frag_index == 1:
+                            continuation_id, offset, click_tracking_params = try_refresh_replay_beginning(live_chat_continuation)
+                        else:
+                            continuation_id, offset, click_tracking_params = parse_actions_replay(live_chat_continuation)
+                    elif info_dict['protocol'] == 'youtube_live_chat':
+                        continuation_id, offset, click_tracking_params = parse_actions_live(live_chat_continuation)
+                    return True, continuation_id, offset, click_tracking_params
                 except compat_urllib_error.HTTPError as err:
                     count += 1
                     if count <= fragment_retries:
                         self.report_retry_fragment(err, frag_index, count, fragment_retries)
             if count > fragment_retries:
                 self.report_error('giving up after %s fragment retries' % fragment_retries)
-                return False, None, None
+                return False, None, None, None
 
-        self._prepare_and_start_frag_download(ctx)
+        self._prepare_and_start_frag_download(ctx, info_dict)
 
         success, raw_fragment = dl_fragment(info_dict['url'])
         if not success:
             return False
         try:
-            data = ie._extract_yt_initial_data(video_id, raw_fragment.decode('utf-8', 'replace'))
+            data = ie.extract_yt_initial_data(video_id, raw_fragment.decode('utf-8', 'replace'))
         except RegexNotFoundError:
             return False
         continuation_id = try_get(
@@ -92,7 +155,7 @@ class YoutubeLiveChatReplayFD(FragmentFD):
         # no data yet but required to call _append_fragment
         self._append_fragment(ctx, b'')
 
-        ytcfg = ie._extract_ytcfg(video_id, raw_fragment.decode('utf-8', 'replace'))
+        ytcfg = ie.extract_ytcfg(video_id, raw_fragment.decode('utf-8', 'replace'))
 
         if not ytcfg:
             return False
@@ -100,9 +163,16 @@ class YoutubeLiveChatReplayFD(FragmentFD):
         innertube_context = try_get(ytcfg, lambda x: x['INNERTUBE_CONTEXT'])
         if not api_key or not innertube_context:
             return False
-        url = 'https://www.youtube.com/youtubei/v1/live_chat/get_live_chat_replay?key=' + api_key
+        visitor_data = try_get(innertube_context, lambda x: x['client']['visitorData'], str)
+        if info_dict['protocol'] == 'youtube_live_chat_replay':
+            url = 'https://www.youtube.com/youtubei/v1/live_chat/get_live_chat_replay?key=' + api_key
+            chat_page_url = 'https://www.youtube.com/live_chat_replay?continuation=' + continuation_id
+        elif info_dict['protocol'] == 'youtube_live_chat':
+            url = 'https://www.youtube.com/youtubei/v1/live_chat/get_live_chat?key=' + api_key
+            chat_page_url = 'https://www.youtube.com/live_chat?continuation=' + continuation_id
 
         frag_index = offset = 0
+        click_tracking_params = None
         while continuation_id is not None:
             frag_index += 1
             request_data = {
@@ -111,12 +181,56 @@ class YoutubeLiveChatReplayFD(FragmentFD):
             }
             if frag_index > 1:
                 request_data['currentPlayerState'] = {'playerOffsetMs': str(max(offset - 5000, 0))}
-            success, continuation_id, offset = download_and_parse_fragment(
-                url, frag_index, json.dumps(request_data, ensure_ascii=False).encode('utf-8') + b'\n')
+                if click_tracking_params:
+                    request_data['context']['clickTracking'] = {'clickTrackingParams': click_tracking_params}
+                headers = ie.generate_api_headers(ytcfg, visitor_data=visitor_data)
+                headers.update({'content-type': 'application/json'})
+                fragment_request_data = json.dumps(request_data, ensure_ascii=False).encode('utf-8') + b'\n'
+                success, continuation_id, offset, click_tracking_params = download_and_parse_fragment(
+                    url, frag_index, fragment_request_data, headers)
+            else:
+                success, continuation_id, offset, click_tracking_params = download_and_parse_fragment(
+                    chat_page_url, frag_index)
             if not success:
                 return False
             if test:
                 break
 
-        self._finish_frag_download(ctx)
+        self._finish_frag_download(ctx, info_dict)
         return True
+
+    @staticmethod
+    def parse_live_timestamp(action):
+        action_content = dict_get(
+            action,
+            ['addChatItemAction', 'addLiveChatTickerItemAction', 'addBannerToLiveChatCommand'])
+        if not isinstance(action_content, dict):
+            return None
+        item = dict_get(action_content, ['item', 'bannerRenderer'])
+        if not isinstance(item, dict):
+            return None
+        renderer = dict_get(item, [
+            # text
+            'liveChatTextMessageRenderer', 'liveChatPaidMessageRenderer',
+            'liveChatMembershipItemRenderer', 'liveChatPaidStickerRenderer',
+            # ticker
+            'liveChatTickerPaidMessageItemRenderer',
+            'liveChatTickerSponsorItemRenderer',
+            # banner
+            'liveChatBannerRenderer',
+        ])
+        if not isinstance(renderer, dict):
+            return None
+        parent_item_getters = [
+            lambda x: x['showItemEndpoint']['showLiveChatItemEndpoint']['renderer'],
+            lambda x: x['contents'],
+        ]
+        parent_item = try_get(renderer, parent_item_getters, dict)
+        if parent_item:
+            renderer = dict_get(parent_item, [
+                'liveChatTextMessageRenderer', 'liveChatPaidMessageRenderer',
+                'liveChatMembershipItemRenderer', 'liveChatPaidStickerRenderer',
+            ])
+            if not isinstance(renderer, dict):
+                return None
+        return int_or_none(renderer.get('timestampUsec'), 1000)

yt_dlp/extractor/adobepass.py

@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import json
 import re
 import time
 import xml.etree.ElementTree as etree
@@ -61,6 +62,11 @@ MSO_INFO = {
         'username_field': 'IDToken1',
         'password_field': 'IDToken2',
     },
+    'Spectrum': {
+        'name': 'Spectrum',
+        'username_field': 'IDToken1',
+        'password_field': 'IDToken2',
+    },
     'Philo': {
         'name': 'Philo',
         'username_field': 'ident'
@@ -1524,6 +1530,41 @@ class AdobePassIE(InfoExtractor):
                         }), headers={
                             'Content-Type': 'application/x-www-form-urlencoded'
                         })
+                elif mso_id == 'Spectrum':
+                    # Spectrum's login for is dynamically loaded via JS so we need to hardcode the flow
+                    # as a one-off implementation.
+                    provider_redirect_page, urlh = provider_redirect_page_res
+                    provider_login_page_res = post_form(
+                        provider_redirect_page_res, self._DOWNLOADING_LOGIN_PAGE)
+                    saml_login_page, urlh = provider_login_page_res
+                    relay_state = self._search_regex(
+                        r'RelayState\s*=\s*"(?P<relay>.+?)";',
+                        saml_login_page, 'RelayState', group='relay')
+                    saml_request = self._search_regex(
+                        r'SAMLRequest\s*=\s*"(?P<saml_request>.+?)";',
+                        saml_login_page, 'SAMLRequest', group='saml_request')
+                    login_json = {
+                        mso_info['username_field']: username,
+                        mso_info['password_field']: password,
+                        'RelayState': relay_state,
+                        'SAMLRequest': saml_request,
+                    }
+                    saml_response_json = self._download_json(
+                        'https://tveauthn.spectrum.net/tveauthentication/api/v1/manualAuth', video_id,
+                        'Downloading SAML Response',
+                        data=json.dumps(login_json).encode(),
+                        headers={
+                            'Content-Type': 'application/json',
+                            'Accept': 'application/json',
+                        })
+                    self._download_webpage(
+                        saml_response_json['SAMLRedirectUri'], video_id,
+                        'Confirming Login', data=urlencode_postdata({
+                            'SAMLResponse': saml_response_json['SAMLResponse'],
+                            'RelayState': relay_state,
+                        }), headers={
+                            'Content-Type': 'application/x-www-form-urlencoded'
+                        })
                 else:
                     # Some providers (e.g. DIRECTV NOW) have another meta refresh
                     # based redirect that should be followed.

yt_dlp/extractor/appleconnect.py

@@ -9,10 +9,10 @@ from ..utils import (
 
 
 class AppleConnectIE(InfoExtractor):
-    _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)'
-    _TEST = {
+    _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/(?:id)?sa\.(?P<id>[\w-]+)'
+    _TESTS = [{
         'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
-        'md5': 'e7c38568a01ea45402570e6029206723',
+        'md5': 'c1d41f72c8bcaf222e089434619316e4',
         'info_dict': {
             'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
             'ext': 'm4v',
@@ -22,7 +22,10 @@ class AppleConnectIE(InfoExtractor):
             'upload_date': '20150710',
             'timestamp': 1436545535,
         },
-    }
+    }, {
+        'url': 'https://itunes.apple.com/us/post/sa.0fe0229f-2457-11e5-9f40-1bb645f2d5d9',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -36,7 +39,7 @@ class AppleConnectIE(InfoExtractor):
 
         video_data = self._parse_json(video_json, video_id)
         timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp'))
-        like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count'))
+        like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count', default=None))
 
         return {
             'id': video_id,

yt_dlp/extractor/bilibili.py

@@ -281,7 +281,7 @@ class BiliBiliIE(InfoExtractor):
             webpage)
         if uploader_mobj:
             info.update({
-                'uploader': uploader_mobj.group('name'),
+                'uploader': uploader_mobj.group('name').strip(),
                 'uploader_id': uploader_mobj.group('id'),
             })
 

yt_dlp/extractor/bravotv.py

@@ -8,6 +8,9 @@ from ..utils import (
     smuggle_url,
     update_url_query,
     int_or_none,
+    float_or_none,
+    try_get,
+    dict_get,
 )
 
 
@@ -24,6 +27,11 @@ class BravoTVIE(AdobePassIE):
             'uploader': 'NBCU-BRAV',
             'upload_date': '20190314',
             'timestamp': 1552591860,
+            'season_number': 16,
+            'episode_number': 15,
+            'series': 'Top Chef',
+            'episode': 'The Top Chef Season 16 Winner Is...',
+            'duration': 190.0,
         }
     }, {
         'url': 'http://www.bravotv.com/below-deck/season-3/ep-14-reunion-part-1',
@@ -79,12 +87,34 @@ class BravoTVIE(AdobePassIE):
                 'episode_number': int_or_none(metadata.get('episode_num')),
             })
             query['switch'] = 'progressive'
+
+        tp_url = 'http://link.theplatform.com/s/%s/%s' % (account_pid, tp_path)
+
+        tp_metadata = self._download_json(
+            update_url_query(tp_url, {'format': 'preview'}),
+            display_id, fatal=False)
+        if tp_metadata:
+            info.update({
+                'title': tp_metadata.get('title'),
+                'description': tp_metadata.get('description'),
+                'duration': float_or_none(tp_metadata.get('duration'), 1000),
+                'season_number': int_or_none(
+                    dict_get(tp_metadata, ('pl1$seasonNumber', 'nbcu$seasonNumber'))),
+                'episode_number': int_or_none(
+                    dict_get(tp_metadata, ('pl1$episodeNumber', 'nbcu$episodeNumber'))),
+                # For some reason the series is sometimes wrapped into a single element array.
+                'series': try_get(
+                    dict_get(tp_metadata, ('pl1$show', 'nbcu$show')),
+                    lambda x: x[0] if isinstance(x, list) else x,
+                    expected_type=str),
+                'episode': dict_get(
+                    tp_metadata, ('pl1$episodeName', 'nbcu$episodeName', 'title')),
+            })
+
         info.update({
             '_type': 'url_transparent',
             'id': release_pid,
-            'url': smuggle_url(update_url_query(
-                'http://link.theplatform.com/s/%s/%s' % (account_pid, tp_path),
-                query), {'force_smil_url': True}),
+            'url': smuggle_url(update_url_query(tp_url, query), {'force_smil_url': True}),
             'ie_key': 'ThePlatform',
         })
         return info

yt_dlp/extractor/canvas.py

@@ -24,7 +24,7 @@ class CanvasIE(InfoExtractor):
     _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet|vrt(?:video|nieuws)|sporza|dako)/assets/(?P<id>[^/?#&]+)'
     _TESTS = [{
         'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
-        'md5': '68993eda72ef62386a15ea2cf3c93107',
+        'md5': '37b2b7bb9b3dcaa05b67058dc3a714a9',
         'info_dict': {
             'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
             'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
@@ -32,9 +32,9 @@ class CanvasIE(InfoExtractor):
             'title': 'Nachtwacht: De Greystook',
             'description': 'Nachtwacht: De Greystook',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 1468.04,
+            'duration': 1468.02,
         },
-        'expected_warnings': ['is not a supported codec', 'Unknown MIME type'],
+        'expected_warnings': ['is not a supported codec'],
     }, {
         'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
         'only_matching': True,

yt_dlp/extractor/common.py

@@ -19,7 +19,6 @@ from ..compat import (
     compat_etree_Element,
     compat_etree_fromstring,
     compat_getpass,
-    compat_integer_types,
     compat_http_client,
     compat_os_name,
     compat_str,
@@ -70,6 +69,7 @@ from ..utils import (
     str_or_none,
     str_to_int,
     strip_or_none,
+    traverse_obj,
     unescapeHTML,
     unified_strdate,
     unified_timestamp,
@@ -78,6 +78,7 @@ from ..utils import (
     urljoin,
     url_basename,
     url_or_none,
+    variadic,
     xpath_element,
     xpath_text,
     xpath_with_ns,
@@ -228,6 +229,7 @@ class InfoExtractor(object):
                         * "resolution" (optional, string "{width}x{height}",
                                         deprecated)
                         * "filesize" (optional, int)
+                        * "_test_url" (optional, bool) - If true, test the URL
     thumbnail:      Full URL to a video thumbnail image.
     description:    Full video description.
     uploader:       Full name of the video uploader.
@@ -290,10 +292,13 @@ class InfoExtractor(object):
     categories:     A list of categories that the video falls in, for example
                     ["Sports", "Berlin"]
     tags:           A list of tags assigned to the video, e.g. ["sweden", "pop music"]
+    cast:           A list of the video cast
     is_live:        True, False, or None (=unknown). Whether this video is a
                     live stream that goes on instead of a fixed-length video.
     was_live:       True, False, or None (=unknown). Whether this video was
                     originally a live stream.
+    live_status:    'is_live', 'upcoming', 'was_live', 'not_live' or None (=unknown)
+                    If absent, automatically set from is_live, was_live
     start_time:     Time in seconds where the reproduction should start, as
                     specified in the URL.
     end_time:       Time in seconds where the reproduction should end, as
@@ -626,14 +631,10 @@ class InfoExtractor(object):
         assert isinstance(err, compat_urllib_error.HTTPError)
         if expected_status is None:
             return False
-        if isinstance(expected_status, compat_integer_types):
-            return err.code == expected_status
-        elif isinstance(expected_status, (list, tuple)):
-            return err.code in expected_status
         elif callable(expected_status):
             return expected_status(err.code) is True
         else:
-            assert False
+            return err.code in variadic(expected_status)
 
     def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, data=None, headers={}, query={}, expected_status=None):
         """
@@ -1036,7 +1037,9 @@ class InfoExtractor(object):
             metadata_available=False, method='any'):
         if metadata_available and self.get_param('ignore_no_formats_error'):
             self.report_warning(msg)
-        raise ExtractorError('%s. %s' % (msg, self._LOGIN_HINTS[method]), expected=True)
+        if method is not None:
+            msg = '%s. %s' % (msg, self._LOGIN_HINTS[method])
+        raise ExtractorError(msg, expected=True)
 
     def raise_geo_restricted(
             self, msg='This video is not available from your location due to geo restriction',
@@ -1111,6 +1114,8 @@ class InfoExtractor(object):
             if group is None:
                 # return the first matching group
                 return next(g for g in mobj.groups() if g is not None)
+            elif isinstance(group, (list, tuple)):
+                return tuple(mobj.group(g) for g in group)
             else:
                 return mobj.group(group)
         elif default is not NO_DEFAULT:
@@ -1203,8 +1208,7 @@ class InfoExtractor(object):
                     [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop)
 
     def _og_search_property(self, prop, html, name=None, **kargs):
-        if not isinstance(prop, (list, tuple)):
-            prop = [prop]
+        prop = variadic(prop)
         if name is None:
             name = 'OpenGraph %s' % prop[0]
         og_regexes = []
@@ -1234,8 +1238,7 @@ class InfoExtractor(object):
         return self._og_search_property('url', html, **kargs)
 
     def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
-        if not isinstance(name, (list, tuple)):
-            name = [name]
+        name = variadic(name)
         if display_name is None:
             display_name = name[0]
         return self._html_search_regex(
@@ -1473,7 +1476,7 @@ class InfoExtractor(object):
     class FormatSort:
         regex = r' *((?P<reverse>\+)?(?P<field>[a-zA-Z0-9_]+)((?P<separator>[~:])(?P<limit>.*?))?)? *$'
 
-        default = ('hidden', 'hasvid', 'ie_pref', 'lang', 'quality',
+        default = ('hidden', 'aud_or_vid', 'hasvid', 'ie_pref', 'lang', 'quality',
                    'res', 'fps', 'codec:vp9.2', 'size', 'br', 'asr',
                    'proto', 'ext', 'hasaud', 'source', 'format_id')  # These must not be aliases
         ytdl_default = ('hasaud', 'quality', 'tbr', 'filesize', 'vbr',
@@ -1486,7 +1489,7 @@ class InfoExtractor(object):
             'acodec': {'type': 'ordered', 'regex': True,
                        'order': ['opus', 'vorbis', 'aac', 'mp?4a?', 'mp3', 'e?a?c-?3', 'dts', '', None, 'none']},
             'proto': {'type': 'ordered', 'regex': True, 'field': 'protocol',
-                      'order': ['(ht|f)tps', '(ht|f)tp$', 'm3u8.+', 'm3u8', '.*dash', '', 'mms|rtsp', 'none', 'f4']},
+                      'order': ['(ht|f)tps', '(ht|f)tp$', 'm3u8.+', '.*dash', 'ws|websocket', '', 'mms|rtsp', 'none', 'f4']},
             'vext': {'type': 'ordered', 'field': 'video_ext',
                      'order': ('mp4', 'webm', 'flv', '', 'none'),
                      'order_free': ('webm', 'mp4', 'flv', '', 'none')},
@@ -1494,6 +1497,9 @@ class InfoExtractor(object):
                      'order': ('m4a', 'aac', 'mp3', 'ogg', 'opus', 'webm', '', 'none'),
                      'order_free': ('opus', 'ogg', 'webm', 'm4a', 'mp3', 'aac', '', 'none')},
             'hidden': {'visible': False, 'forced': True, 'type': 'extractor', 'max': -1000},
+            'aud_or_vid': {'visible': False, 'forced': True, 'type': 'multiple', 'default': 1,
+                           'field': ('vcodec', 'acodec'),
+                           'function': lambda it: int(any(v != 'none' for v in it))},
             'ie_pref': {'priority': True, 'type': 'extractor'},
             'hasvid': {'priority': True, 'field': 'vcodec', 'type': 'boolean', 'not_in_list': ('none',)},
             'hasaud': {'field': 'acodec', 'type': 'boolean', 'not_in_list': ('none',)},
@@ -1701,9 +1707,7 @@ class InfoExtractor(object):
 
                 def wrapped_function(values):
                     values = tuple(filter(lambda x: x is not None, values))
-                    return (self._get_field_setting(field, 'function')(*values) if len(values) > 1
-                            else values[0] if values
-                            else None)
+                    return self._get_field_setting(field, 'function')(values) if values else None
 
                 value = wrapped_function((get_value(f) for f in actual_fields))
             else:
@@ -1719,7 +1723,7 @@ class InfoExtractor(object):
             if not format.get('ext') and 'url' in format:
                 format['ext'] = determine_ext(format['url'])
             if format.get('vcodec') == 'none':
-                format['audio_ext'] = format['ext']
+                format['audio_ext'] = format['ext'] if format.get('acodec') != 'none' else 'none'
                 format['video_ext'] = 'none'
             else:
                 format['video_ext'] = format['ext']
@@ -1976,24 +1980,33 @@ class InfoExtractor(object):
             preference=None, quality=None, m3u8_id=None, live=False, note=None,
             errnote=None, fatal=True, data=None, headers={}, query={},
             video_id=None):
+        formats, subtitles = [], {}
 
         if '#EXT-X-FAXS-CM:' in m3u8_doc:  # Adobe Flash Access
-            return [], {}
+            return formats, subtitles
 
         if (not self.get_param('allow_unplayable_formats')
                 and re.search(r'#EXT-X-SESSION-KEY:.*?URI="skd://', m3u8_doc)):  # Apple FairPlay
-            return [], {}
+            return formats, subtitles
 
-        formats = []
+        def format_url(url):
+            return url if re.match(r'^https?://', url) else compat_urlparse.urljoin(m3u8_url, url)
 
-        subtitles = {}
+        if self.get_param('hls_split_discontinuity', False):
+            def _extract_m3u8_playlist_indices(manifest_url=None, m3u8_doc=None):
+                if not m3u8_doc:
+                    if not manifest_url:
+                        return []
+                    m3u8_doc = self._download_webpage(
+                        manifest_url, video_id, fatal=fatal, data=data, headers=headers,
+                        note=False, errnote='Failed to download m3u8 playlist information')
+                    if m3u8_doc is False:
+                        return []
+                return range(1 + sum(line.startswith('#EXT-X-DISCONTINUITY') for line in m3u8_doc.splitlines()))
 
-        format_url = lambda u: (
-            u
-            if re.match(r'^https?://', u)
-            else compat_urlparse.urljoin(m3u8_url, u))
-
-        split_discontinuity = self.get_param('hls_split_discontinuity', False)
+        else:
+            def _extract_m3u8_playlist_indices(*args, **kwargs):
+                return [None]
 
         # References:
         # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-21
@@ -2011,68 +2024,16 @@ class InfoExtractor(object):
         # media playlist and MUST NOT appear in master playlist thus we can
         # clearly detect media playlist with this criterion.
 
-        def _extract_m3u8_playlist_formats(format_url=None, m3u8_doc=None, video_id=None,
-                                           fatal=True, data=None, headers={}):
-            if not m3u8_doc:
-                if not format_url:
-                    return []
-                res = self._download_webpage_handle(
-                    format_url, video_id,
-                    note=False,
-                    errnote='Failed to download m3u8 playlist information',
-                    fatal=fatal, data=data, headers=headers)
-
-                if res is False:
-                    return []
-
-                m3u8_doc, urlh = res
-                format_url = urlh.geturl()
-
-            playlist_formats = []
-            i = (
-                0
-                if split_discontinuity
-                else None)
-            format_info = {
-                'index': i,
-                'key_data': None,
-                'files': [],
-            }
-            for line in m3u8_doc.splitlines():
-                if not line.startswith('#'):
-                    format_info['files'].append(line)
-                elif split_discontinuity and line.startswith('#EXT-X-DISCONTINUITY'):
-                    i += 1
-                    playlist_formats.append(format_info)
-                    format_info = {
-                        'index': i,
-                        'url': format_url,
-                        'files': [],
-                    }
-            playlist_formats.append(format_info)
-            return playlist_formats
-
         if '#EXT-X-TARGETDURATION' in m3u8_doc:  # media playlist, return as is
-
-            playlist_formats = _extract_m3u8_playlist_formats(m3u8_doc=m3u8_doc)
-
-            for format in playlist_formats:
-                format_id = []
-                if m3u8_id:
-                    format_id.append(m3u8_id)
-                format_index = format.get('index')
-                if format_index:
-                    format_id.append(str(format_index))
-                f = {
-                    'format_id': '-'.join(format_id),
-                    'format_index': format_index,
-                    'url': m3u8_url,
-                    'ext': ext,
-                    'protocol': entry_protocol,
-                    'preference': preference,
-                    'quality': quality,
-                }
-                formats.append(f)
+            formats = [{
+                'format_id': '-'.join(map(str, filter(None, [m3u8_id, idx]))),
+                'format_index': idx,
+                'url': m3u8_url,
+                'ext': ext,
+                'protocol': entry_protocol,
+                'preference': preference,
+                'quality': quality,
+            } for idx in _extract_m3u8_playlist_indices(m3u8_doc=m3u8_doc)]
 
             return formats, subtitles
 
@@ -2112,31 +2073,19 @@ class InfoExtractor(object):
             media_url = media.get('URI')
             if media_url:
                 manifest_url = format_url(media_url)
-                format_id = []
-                playlist_formats = _extract_m3u8_playlist_formats(manifest_url, video_id=video_id,
-                                                                  fatal=fatal, data=data, headers=headers)
-
-                for format in playlist_formats:
-                    format_index = format.get('index')
-                    for v in (m3u8_id, group_id, name):
-                        if v:
-                            format_id.append(v)
-                    if format_index:
-                        format_id.append(str(format_index))
-                    f = {
-                        'format_id': '-'.join(format_id),
-                        'format_index': format_index,
-                        'url': manifest_url,
-                        'manifest_url': m3u8_url,
-                        'language': media.get('LANGUAGE'),
-                        'ext': ext,
-                        'protocol': entry_protocol,
-                        'preference': preference,
-                        'quality': quality,
-                    }
-                    if media_type == 'AUDIO':
-                        f['vcodec'] = 'none'
-                    formats.append(f)
+                formats.extend({
+                    'format_id': '-'.join(map(str, filter(None, (m3u8_id, group_id, name, idx)))),
+                    'format_note': name,
+                    'format_index': idx,
+                    'url': manifest_url,
+                    'manifest_url': m3u8_url,
+                    'language': media.get('LANGUAGE'),
+                    'ext': ext,
+                    'protocol': entry_protocol,
+                    'preference': preference,
+                    'quality': quality,
+                    'vcodec': 'none' if media_type == 'AUDIO' else None,
+                } for idx in _extract_m3u8_playlist_indices(manifest_url))
 
         def build_stream_name():
             # Despite specification does not mention NAME attribute for
@@ -2175,25 +2124,17 @@ class InfoExtractor(object):
                     or last_stream_inf.get('BANDWIDTH'), scale=1000)
                 manifest_url = format_url(line.strip())
 
-                playlist_formats = _extract_m3u8_playlist_formats(manifest_url, video_id=video_id,
-                                                                  fatal=fatal, data=data, headers=headers)
-
-                for frmt in playlist_formats:
-                    format_id = []
-                    if m3u8_id:
-                        format_id.append(m3u8_id)
-                    format_index = frmt.get('index')
-                    stream_name = build_stream_name()
+                for idx in _extract_m3u8_playlist_indices(manifest_url):
+                    format_id = [m3u8_id, None, idx]
                     # Bandwidth of live streams may differ over time thus making
                     # format_id unpredictable. So it's better to keep provided
                     # format_id intact.
                     if not live:
-                        format_id.append(stream_name if stream_name else '%d' % (tbr if tbr else len(formats)))
-                    if format_index:
-                        format_id.append(str(format_index))
+                        stream_name = build_stream_name()
+                        format_id[1] = stream_name if stream_name else '%d' % (tbr if tbr else len(formats))
                     f = {
-                        'format_id': '-'.join(format_id),
-                        'format_index': format_index,
+                        'format_id': '-'.join(map(str, filter(None, format_id))),
+                        'format_index': idx,
                         'url': manifest_url,
                         'manifest_url': m3u8_url,
                         'tbr': tbr,
@@ -2268,7 +2209,7 @@ class InfoExtractor(object):
                 out.append('{%s}%s' % (namespace, c))
         return '/'.join(out)
 
-    def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None, transform_source=None):
+    def _extract_smil_formats_and_subtitles(self, smil_url, video_id, fatal=True, f4m_params=None, transform_source=None):
         smil = self._download_smil(smil_url, video_id, fatal=fatal, transform_source=transform_source)
 
         if smil is False:
@@ -2277,8 +2218,21 @@ class InfoExtractor(object):
 
         namespace = self._parse_smil_namespace(smil)
 
-        return self._parse_smil_formats(
+        fmts = self._parse_smil_formats(
             smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params)
+        subs = self._parse_smil_subtitles(
+            smil, namespace=namespace)
+
+        return fmts, subs
+
+    def _extract_smil_formats(self, *args, **kwargs):
+        fmts, subs = self._extract_smil_formats_and_subtitles(*args, **kwargs)
+        if subs:
+            self.report_warning(bug_reports_message(
+                "Ignoring subtitle tracks found in the SMIL manifest; "
+                "if any subtitle tracks are missing,"
+            ))
+        return fmts
 
     def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None):
         smil = self._download_smil(smil_url, video_id, fatal=fatal)
@@ -2636,7 +2590,7 @@ class InfoExtractor(object):
                     mime_type = representation_attrib['mimeType']
                     content_type = representation_attrib.get('contentType', mime_type.split('/')[0])
 
-                    if content_type in ('video', 'audio', 'text'):
+                    if content_type in ('video', 'audio', 'text') or mime_type == 'image/jpeg':
                         base_url = ''
                         for element in (representation, adaptation_set, period, mpd_doc):
                             base_url_e = element.find(_add_ns('BaseURL'))
@@ -2653,9 +2607,15 @@ class InfoExtractor(object):
                         url_el = representation.find(_add_ns('BaseURL'))
                         filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
                         bandwidth = int_or_none(representation_attrib.get('bandwidth'))
+                        if representation_id is not None:
+                            format_id = representation_id
+                        else:
+                            format_id = content_type
+                        if mpd_id:
+                            format_id = mpd_id + '-' + format_id
                         if content_type in ('video', 'audio'):
                             f = {
-                                'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id,
+                                'format_id': format_id,
                                 'manifest_url': mpd_url,
                                 'ext': mimetype2ext(mime_type),
                                 'width': int_or_none(representation_attrib.get('width')),
@@ -2675,6 +2635,17 @@ class InfoExtractor(object):
                                 'manifest_url': mpd_url,
                                 'filesize': filesize,
                             }
+                        elif mime_type == 'image/jpeg':
+                            # See test case in VikiIE
+                            # https://www.viki.com/videos/1175236v-choosing-spouse-by-lottery-episode-1
+                            f = {
+                                'format_id': format_id,
+                                'ext': 'mhtml',
+                                'manifest_url': mpd_url,
+                                'format_note': 'DASH storyboards (jpeg)',
+                                'acodec': 'none',
+                                'vcodec': 'none',
+                            }
                         representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)
 
                         def prepare_template(template_name, identifiers):
@@ -2693,7 +2664,8 @@ class InfoExtractor(object):
                                     t += c
                             # Next, $...$ templates are translated to their
                             # %(...) counterparts to be used with % operator
-                            t = t.replace('$RepresentationID$', representation_id)
+                            if representation_id is not None:
+                                t = t.replace('$RepresentationID$', representation_id)
                             t = re.sub(r'\$(%s)\$' % '|'.join(identifiers), r'%(\1)d', t)
                             t = re.sub(r'\$(%s)%%([^$]+)\$' % '|'.join(identifiers), r'%(\1)\2', t)
                             t.replace('$$', '$')
@@ -2810,7 +2782,7 @@ class InfoExtractor(object):
                                 'url': mpd_url or base_url,
                                 'fragment_base_url': base_url,
                                 'fragments': [],
-                                'protocol': 'http_dash_segments',
+                                'protocol': 'http_dash_segments' if mime_type != 'image/jpeg' else 'mhtml',
                             })
                             if 'initialization_url' in representation_ms_info:
                                 initialization_url = representation_ms_info['initialization_url']
@@ -2821,7 +2793,7 @@ class InfoExtractor(object):
                         else:
                             # Assuming direct URL to unfragmented media.
                             f['url'] = base_url
-                        if content_type in ('video', 'audio'):
+                        if content_type in ('video', 'audio') or mime_type == 'image/jpeg':
                             formats.append(f)
                         elif content_type == 'text':
                             subtitles.setdefault(lang or 'und', []).append(f)
@@ -3484,16 +3456,8 @@ class InfoExtractor(object):
         return ret
 
     @classmethod
-    def _merge_subtitles(cls, *dicts, **kwargs):
+    def _merge_subtitles(cls, *dicts, target=None):
         """ Merge subtitle dictionaries, language by language. """
-
-        target = (lambda target=None: target)(**kwargs)
-        # The above lambda extracts the keyword argument 'target' from kwargs
-        # while ensuring there are no stray ones. When Python 2 support
-        # is dropped, remove it and change the function signature to:
-        #
-        #     def _merge_subtitles(cls, *dicts, target=None):
-
         if target is None:
             target = {}
         for d in dicts:
@@ -3546,6 +3510,19 @@ class InfoExtractor(object):
             else 'public' if all_known
             else None)
 
+    def _configuration_arg(self, key, default=NO_DEFAULT, casesense=False):
+        '''
+        @returns            A list of values for the extractor argument given by "key"
+                            or "default" if no such key is present
+        @param default      The default value to return when the key is not present (default: [])
+        @param casesense    When false, the values are converted to lower case
+        '''
+        val = traverse_obj(
+            self._downloader.params, ('extractor_args', self.ie_key().lower(), key))
+        if val is None:
+            return [] if default is NO_DEFAULT else default
+        return list(val) if casesense else [x.lower() for x in val]
+
 
 class SearchInfoExtractor(InfoExtractor):
     """

yt_dlp/extractor/crunchyroll.py

@@ -636,7 +636,7 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login|media-\d+))(?P<id>[\w\-]+))/?(?:\?|$)'
 
     _TESTS = [{
-        'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
+        'url': 'https://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
         'info_dict': {
             'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
             'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
@@ -661,7 +661,8 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
         show_id = self._match_id(url)
 
         webpage = self._download_webpage(
-            self._add_skip_wall(url), show_id,
+            # https:// gives a 403, but http:// does not
+            self._add_skip_wall(url).replace('https://', 'http://'), show_id,
             headers=self.geo_verification_headers())
         title = self._html_search_meta('name', webpage, default=None)
 

yt_dlp/extractor/curiositystream.py

@@ -143,9 +143,9 @@ class CuriosityStreamIE(CuriosityStreamBaseIE):
         }
 
 
-class CuriosityStreamCollectionsIE(CuriosityStreamBaseIE):
-    IE_NAME = 'curiositystream:collections'
-    _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/collections/(?P<id>\d+)'
+class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
+    IE_NAME = 'curiositystream:collection'
+    _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/(?:collections?|series)/(?P<id>\d+)'
     _API_BASE_URL = 'https://api.curiositystream.com/v2/collections/'
     _TESTS = [{
         'url': 'https://curiositystream.com/collections/86',
@@ -155,6 +155,20 @@ class CuriosityStreamCollectionsIE(CuriosityStreamBaseIE):
             'description': 'Wondering where to start? Here are a few of our favorite series and films... from our couch to yours.',
         },
         'playlist_mincount': 7,
+    }, {
+        'url': 'https://app.curiositystream.com/collection/2',
+        'info_dict': {
+            'id': '2',
+            'title': 'Curious Minds: The Internet',
+            'description': 'How is the internet shaping our lives in the 21st Century?',
+        },
+        'playlist_mincount': 16,
+    }, {
+        'url': 'https://curiositystream.com/series/2',
+        'only_matching': True,
+    }, {
+        'url': 'https://curiositystream.com/collections/36',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -163,25 +177,10 @@ class CuriosityStreamCollectionsIE(CuriosityStreamBaseIE):
         entries = []
         for media in collection.get('media', []):
             media_id = compat_str(media.get('id'))
-            media_type, ie = ('series', CuriosityStreamSeriesIE) if media.get('is_collection') else ('video', CuriosityStreamIE)
+            media_type, ie = ('series', CuriosityStreamCollectionIE) if media.get('is_collection') else ('video', CuriosityStreamIE)
             entries.append(self.url_result(
                 'https://curiositystream.com/%s/%s' % (media_type, media_id),
                 ie=ie.ie_key(), video_id=media_id))
         return self.playlist_result(
             entries, collection_id,
             collection.get('title'), collection.get('description'))
-
-
-class CuriosityStreamSeriesIE(CuriosityStreamCollectionsIE):
-    IE_NAME = 'curiositystream:series'
-    _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/series/(?P<id>\d+)'
-    _API_BASE_URL = 'https://api.curiositystream.com/v2/series/'
-    _TESTS = [{
-        'url': 'https://app.curiositystream.com/series/2',
-        'info_dict': {
-            'id': '2',
-            'title': 'Curious Minds: The Internet',
-            'description': 'How is the internet shaping our lives in the 21st Century?',
-        },
-        'playlist_mincount': 16,
-    }]

yt_dlp/extractor/douyin.py

@@ -0,0 +1,145 @@
+# coding: utf-8
+
+from ..utils import (
+    int_or_none,
+    traverse_obj,
+    url_or_none,
+)
+from .common import (
+    InfoExtractor,
+    compat_urllib_parse_unquote,
+)
+
+
+class DouyinIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'https://www.douyin.com/video/6961737553342991651',
+        'md5': '10523312c8b8100f353620ac9dc8f067',
+        'info_dict': {
+            'id': '6961737553342991651',
+            'ext': 'mp4',
+            'title': '#杨超越  小小水手带你去远航❤️',
+            'uploader': '杨超越',
+            'upload_date': '20210513',
+            'timestamp': 1620905839,
+            'uploader_id': '110403406559',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+        }
+    }, {
+        'url': 'https://www.douyin.com/video/6982497745948921092',
+        'md5': 'd78408c984b9b5102904cf6b6bc2d712',
+        'info_dict': {
+            'id': '6982497745948921092',
+            'ext': 'mp4',
+            'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
+            'uploader': '杨超越工作室',
+            'upload_date': '20210708',
+            'timestamp': 1625739481,
+            'uploader_id': '408654318141572',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+        }
+    }, {
+        'url': 'https://www.douyin.com/video/6953975910773099811',
+        'md5': '72e882e24f75064c218b76c8b713c185',
+        'info_dict': {
+            'id': '6953975910773099811',
+            'ext': 'mp4',
+            'title': '#一起看海  出现在你的夏日里',
+            'uploader': '杨超越',
+            'upload_date': '20210422',
+            'timestamp': 1619098692,
+            'uploader_id': '110403406559',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+        }
+    }, {
+        'url': 'https://www.douyin.com/video/6950251282489675042',
+        'md5': 'b4db86aec367ef810ddd38b1737d2fed',
+        'info_dict': {
+            'id': '6950251282489675042',
+            'ext': 'mp4',
+            'title': '哈哈哈，成功了哈哈哈哈哈哈',
+            'uploader': '杨超越',
+            'upload_date': '20210412',
+            'timestamp': 1618231483,
+            'uploader_id': '110403406559',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+        }
+    }, {
+        'url': 'https://www.douyin.com/video/6963263655114722595',
+        'md5': '1abe1c477d05ee62efb40bf2329957cf',
+        'info_dict': {
+            'id': '6963263655114722595',
+            'ext': 'mp4',
+            'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
+            'uploader': '杨超越',
+            'upload_date': '20210517',
+            'timestamp': 1621261163,
+            'uploader_id': '110403406559',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+        }
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        render_data = self._parse_json(
+            self._search_regex(
+                r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
+                webpage, 'render data'),
+            video_id, transform_source=compat_urllib_parse_unquote)
+        details = traverse_obj(render_data, (..., 'aweme', 'detail'), get_all=False)
+
+        thumbnails = [{'url': self._proto_relative_url(url)} for url in traverse_obj(
+            details, ('video', ('cover', 'dynamicCover', 'originCover')), expected_type=url_or_none, default=[])]
+
+        common = {
+            'width': traverse_obj(details, ('video', 'width'), expected_type=int),
+            'height': traverse_obj(details, ('video', 'height'), expected_type=int),
+            'ext': 'mp4',
+        }
+        formats = [{**common, 'url': self._proto_relative_url(url)} for url in traverse_obj(
+            details, ('video', 'playAddr', ..., 'src'), expected_type=url_or_none, default=[]) if url]
+        self._remove_duplicate_formats(formats)
+
+        download_url = traverse_obj(details, ('download', 'url'), expected_type=url_or_none)
+        if download_url:
+            formats.append({
+                **common,
+                'format_id': 'download',
+                'url': self._proto_relative_url(download_url),
+                'quality': 1,
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': details.get('desc') or self._html_search_meta('title', webpage),
+            'formats': formats,
+            'thumbnails': thumbnails,
+            'uploader': traverse_obj(details, ('authorInfo', 'nickname'), expected_type=str),
+            'uploader_id': traverse_obj(details, ('authorInfo', 'uid'), expected_type=str),
+            'uploader_url': 'https://www.douyin.com/user/%s' % traverse_obj(
+                details, ('authorInfo', 'secUid'), expected_type=str),
+            'timestamp': int_or_none(details.get('createTime')),
+            'duration': traverse_obj(details, ('video', 'duration'), expected_type=int),
+            'view_count': traverse_obj(details, ('stats', 'playCount'), expected_type=int),
+            'like_count': traverse_obj(details, ('stats', 'diggCount'), expected_type=int),
+            'repost_count': traverse_obj(details, ('stats', 'shareCount'), expected_type=int),
+            'comment_count': traverse_obj(details, ('stats', 'commentCount'), expected_type=int),
+        }

yt_dlp/extractor/egghead.py

@@ -22,16 +22,19 @@ class EggheadBaseIE(InfoExtractor):
 class EggheadCourseIE(EggheadBaseIE):
     IE_DESC = 'egghead.io course'
     IE_NAME = 'egghead:course'
-    _VALID_URL = r'https://egghead\.io/courses/(?P<id>[^/?#&]+)'
-    _TEST = {
+    _VALID_URL = r'https://(?:app\.)?egghead\.io/(?:course|playlist)s/(?P<id>[^/?#&]+)'
+    _TESTS = [{
         'url': 'https://egghead.io/courses/professor-frisby-introduces-composable-functional-javascript',
         'playlist_count': 29,
         'info_dict': {
-            'id': '72',
+            'id': '432655',
             'title': 'Professor Frisby Introduces Composable Functional JavaScript',
             'description': 're:(?s)^This course teaches the ubiquitous.*You\'ll start composing functionality before you know it.$',
         },
-    }
+    }, {
+        'url': 'https://app.egghead.io/playlists/professor-frisby-introduces-composable-functional-javascript',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
@@ -65,7 +68,7 @@ class EggheadCourseIE(EggheadBaseIE):
 class EggheadLessonIE(EggheadBaseIE):
     IE_DESC = 'egghead.io lesson'
     IE_NAME = 'egghead:lesson'
-    _VALID_URL = r'https://egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)'
+    _VALID_URL = r'https://(?:app\.)?egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)'
     _TESTS = [{
         'url': 'https://egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box',
         'info_dict': {
@@ -88,6 +91,9 @@ class EggheadLessonIE(EggheadBaseIE):
     }, {
         'url': 'https://egghead.io/api/v1/lessons/react-add-redux-to-a-react-application',
         'only_matching': True,
+    }, {
+        'url': 'https://app.egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

yt_dlp/extractor/extractors.py

@@ -291,8 +291,7 @@ from .ctvnews import CTVNewsIE
 from .cultureunplugged import CultureUnpluggedIE
 from .curiositystream import (
     CuriosityStreamIE,
-    CuriosityStreamCollectionsIE,
-    CuriosityStreamSeriesIE,
+    CuriosityStreamCollectionIE,
 )
 from .cwtv import CWTVIE
 from .dailymail import DailyMailIE
@@ -322,6 +321,7 @@ from .discoveryplusindia import (
     DiscoveryPlusIndiaShowIE,
 )
 from .dotsub import DotsubIE
+from .douyin import DouyinIE
 from .douyutv import (
     DouyuShowIE,
     DouyuTVIE,
@@ -399,7 +399,11 @@ from .facebook import (
     FacebookIE,
     FacebookPluginsVideoIE,
 )
-from .fancode import FancodeVodIE
+from .fancode import (
+    FancodeVodIE,
+    FancodeLiveIE
+)
+
 from .faz import FazIE
 from .fc2 import (
     FC2IE,
@@ -456,7 +460,11 @@ from .frontendmasters import (
     FrontendMastersCourseIE
 )
 from .fujitv import FujiTVFODPlus7IE
-from .funimation import FunimationIE
+from .funimation import (
+    FunimationIE,
+    FunimationPageIE,
+    FunimationShowIE,
+)
 from .funk import FunkIE
 from .fusion import FusionIE
 from .gaia import GaiaIE
@@ -655,10 +663,6 @@ from .linkedin import (
 from .linuxacademy import LinuxAcademyIE
 from .litv import LiTVIE
 from .livejournal import LiveJournalIE
-from .liveleak import (
-    LiveLeakIE,
-    LiveLeakEmbedIE,
-)
 from .livestream import (
     LivestreamIE,
     LivestreamOriginalIE,
@@ -1012,6 +1016,7 @@ from .popcorntimes import PopcorntimesIE
 from .popcorntv import PopcornTVIE
 from .porn91 import Porn91IE
 from .porncom import PornComIE
+from .pornflip import PornFlipIE
 from .pornhd import PornHdIE
 from .pornhub import (
     PornHubIE,
@@ -1065,6 +1070,10 @@ from .rcs import (
     RCSEmbedsIE,
     RCSVariousIE,
 )
+from .rcti import (
+    RCTIPlusIE,
+    RCTIPlusSeriesIE,
+)
 from .rds import RDSIE
 from .redbulltv import (
     RedBullTVIE,

yt_dlp/extractor/facebook.py

@@ -629,16 +629,11 @@ class FacebookIE(InfoExtractor):
 
         process_formats(formats)
 
+        description = self._html_search_meta('description', webpage, default=None)
         video_title = self._html_search_regex(
-            r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage,
-            'title', default=None)
-        if not video_title:
-            video_title = self._html_search_regex(
-                r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
-                webpage, 'alternative title', default=None)
-        if not video_title:
-            video_title = self._html_search_meta(
-                'description', webpage, 'title', default=None)
+            (r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>',
+             r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>'),
+            webpage, 'title', default=None) or self._og_search_title(webpage, default=None) or description
         if video_title:
             video_title = limit_length(video_title, 80)
         else:
@@ -662,6 +657,7 @@ class FacebookIE(InfoExtractor):
             'formats': formats,
             'uploader': uploader,
             'timestamp': timestamp,
+            'description': description,
             'thumbnail': thumbnail,
             'view_count': view_count,
             'subtitles': subtitles,

yt_dlp/extractor/fancode.py

@@ -7,7 +7,8 @@ from ..compat import compat_str
 from ..utils import (
     parse_iso8601,
     ExtractorError,
-    try_get
+    try_get,
+    mimetype2ext
 )
 
 
@@ -38,16 +39,63 @@ class FancodeVodIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    _ACCESS_TOKEN = None
+    _NETRC_MACHINE = 'fancode'
+
+    _LOGIN_HINT = 'Use "--user refresh --password <refresh_token>" to login using a refresh token'
+
+    headers = {
+        'content-type': 'application/json',
+        'origin': 'https://fancode.com',
+        'referer': 'https://fancode.com',
+    }
+
+    def _login(self):
+        # Access tokens are shortlived, so get them using the refresh token.
+        username, password = self._get_login_info()
+        if username == 'refresh' and password is not None:
+            self.report_login()
+            data = '''{
+                "query":"mutation RefreshToken($refreshToken: String\\u0021) { refreshToken(refreshToken: $refreshToken) { accessToken }}",
+                "variables":{
+                    "refreshToken":"%s"
+                },
+                "operationName":"RefreshToken"
+            }''' % password
+
+            token_json = self.download_gql('refresh token', data, "Getting the Access token")
+            self._ACCESS_TOKEN = try_get(token_json, lambda x: x['data']['refreshToken']['accessToken'])
+            if self._ACCESS_TOKEN is None:
+                self.report_warning('Failed to get Access token')
+            else:
+                self.headers.update({'Authorization': 'Bearer %s' % self._ACCESS_TOKEN})
+        elif username is not None:
+            self.report_warning(f'Login using username and password is not currently supported. {self._LOGIN_HINT}')
+
+    def _real_initialize(self):
+        self._login()
+
+    def _check_login_required(self, is_available, is_premium):
+        msg = None
+        if is_premium and self._ACCESS_TOKEN is None:
+            msg = f'This video is only available for registered users. {self._LOGIN_HINT}'
+        elif not is_available and self._ACCESS_TOKEN is not None:
+            msg = 'This video isn\'t available to the current logged in account'
+        if msg:
+            self.raise_login_required(msg, metadata_available=True, method=None)
+
+    def download_gql(self, variable, data, note, fatal=False, headers=headers):
+        return self._download_json(
+            'https://www.fancode.com/graphql', variable,
+            data=data.encode(), note=note,
+            headers=headers, fatal=fatal)
+
     def _real_extract(self, url):
 
         BRIGHTCOVE_URL_TEMPLATE = 'https://players.brightcove.net/%s/default_default/index.html?videoId=%s'
-
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        brightcove_user_id = self._html_search_regex(
-            r'(?:https?://)?players\.brightcove\.net/(\d+)/default_default/index(?:\.min)?\.js',
-            webpage, 'user id')
 
+        brightcove_user_id = '6008340455001'
         data = '''{
             "query":"query Video($id: Int\\u0021, $filter: SegmentFilter) { media(id: $id, filter: $filter) { id contentId title contentId publishedTime totalViews totalUpvotes provider thumbnail { src } mediaSource {brightcove } duration isPremium isUserEntitled tags duration }}",
             "variables":{
@@ -57,15 +105,9 @@ class FancodeVodIE(InfoExtractor):
                 }
             },
             "operationName":"Video"
-            }''' % video_id
+        }''' % video_id
 
-        metadata_json = self._download_json(
-            'https://www.fancode.com/graphql', video_id, data=data.encode(), note='Downloading metadata',
-            headers={
-                'content-type': 'application/json',
-                'origin': 'https://fancode.com',
-                'referer': url,
-            })
+        metadata_json = self.download_gql(video_id, data, note='Downloading metadata')
 
         media = try_get(metadata_json, lambda x: x['data']['media'], dict) or {}
         brightcove_video_id = try_get(media, lambda x: x['mediaSource']['brightcove'], compat_str)
@@ -74,8 +116,8 @@ class FancodeVodIE(InfoExtractor):
             raise ExtractorError('Unable to extract brightcove Video ID')
 
         is_premium = media.get('isPremium')
-        if is_premium:
-            self.report_warning('this video requires a premium account', video_id)
+
+        self._check_login_required(media.get('isUserEntitled'), is_premium)
 
         return {
             '_type': 'url_transparent',
@@ -89,3 +131,57 @@ class FancodeVodIE(InfoExtractor):
             'release_timestamp': parse_iso8601(media.get('publishedTime')),
             'availability': self._availability(needs_premium=is_premium),
         }
+
+
+class FancodeLiveIE(FancodeVodIE):
+    IE_NAME = 'fancode:live'
+
+    _VALID_URL = r'https?://(www\.)?fancode\.com/match/(?P<id>[0-9]+).+'
+
+    _TESTS = [{
+        'url': 'https://fancode.com/match/35328/cricket-fancode-ecs-hungary-2021-bub-vs-blb?slug=commentary',
+        'info_dict': {
+            'id': '35328',
+            'ext': 'mp4',
+            'title': 'BUB vs BLB',
+            "timestamp": 1624863600,
+            'is_live': True,
+            'upload_date': '20210628',
+        },
+        'skip': 'Ended'
+    }, {
+        'url': 'https://fancode.com/match/35328/',
+        'only_matching': True,
+    }, {
+        'url': 'https://fancode.com/match/35567?slug=scorecard',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+
+        id = self._match_id(url)
+        data = '''{
+            "query":"query MatchResponse($id: Int\\u0021, $isLoggedIn: Boolean\\u0021) { match: matchWithScores(id: $id) { id matchDesc mediaId videoStreamId videoStreamUrl { ...VideoSource } liveStreams { videoStreamId videoStreamUrl { ...VideoSource } contentId } name startTime streamingStatus isPremium isUserEntitled @include(if: $isLoggedIn) status metaTags bgImage { src } sport { name slug } tour { id name } squads { name shortName } liveStreams { contentId } mediaId }}fragment VideoSource on VideoSource { title description posterUrl url deliveryType playerType}",
+            "variables":{
+                "id":%s,
+                "isLoggedIn":true
+            },
+            "operationName":"MatchResponse"
+        }''' % id
+
+        info_json = self.download_gql(id, data, "Info json")
+
+        match_info = try_get(info_json, lambda x: x['data']['match'])
+
+        if match_info.get('status') != "LIVE":
+            raise ExtractorError('The stream can\'t be accessed', expected=True)
+        self._check_login_required(match_info.get('isUserEntitled'), True)  # all live streams are premium only
+
+        return {
+            'id': id,
+            'title': match_info.get('name'),
+            'formats': self._extract_akamai_formats(try_get(match_info, lambda x: x['videoStreamUrl']['url']), id),
+            'ext': mimetype2ext(try_get(match_info, lambda x: x['videoStreamUrl']['deliveryType'])),
+            'is_live': True,
+            'release_timestamp': parse_iso8601(match_info.get('startTime'))
+        }

yt_dlp/extractor/funimation.py

@@ -2,59 +2,124 @@
 from __future__ import unicode_literals
 
 import random
+import re
 import string
 
 from .common import InfoExtractor
 from ..compat import compat_HTTPError
 from ..utils import (
     determine_ext,
+    dict_get,
     int_or_none,
     js_to_json,
+    str_or_none,
+    try_get,
+    urlencode_postdata,
     ExtractorError,
-    urlencode_postdata
 )
 
 
-class FunimationIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/(?:[^/]+/)?shows/[^/]+/(?P<id>[^/?#&]+)'
-
-    _NETRC_MACHINE = 'funimation'
-    _TOKEN = None
+class FunimationPageIE(InfoExtractor):
+    IE_NAME = 'funimation:page'
+    _VALID_URL = r'(?P<origin>https?://(?:www\.)?funimation(?:\.com|now\.uk))/(?P<lang>[^/]+/)?(?P<path>shows/(?P<id>[^/]+/[^/?#&]+).*$)'
 
     _TESTS = [{
-        'url': 'https://www.funimation.com/shows/hacksign/role-play/',
-        'info_dict': {
-            'id': '91144',
-            'display_id': 'role-play',
-            'ext': 'mp4',
-            'title': '.hack//SIGN - Role Play',
-            'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
-            'thumbnail': r're:https?://.*\.jpg',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-    }, {
         'url': 'https://www.funimation.com/shows/attack-on-titan-junior-high/broadcast-dub-preview/',
         'info_dict': {
-            'id': '210051',
-            'display_id': 'broadcast-dub-preview',
+            'id': '210050',
             'ext': 'mp4',
-            'title': 'Attack on Titan: Junior High - Broadcast Dub Preview',
-            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+            'title': 'Broadcast Dub Preview',
+            # Other metadata is tested in FunimationIE
         },
         'params': {
-            # m3u8 download
-            'skip_download': True,
+            'skip_download': 'm3u8',
         },
+        'add_ie': ['Funimation'],
     }, {
-        'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
+        # Not available in US
+        'url': 'https://www.funimation.com/shows/hacksign/role-play/',
         'only_matching': True,
     }, {
         # with lang code
         'url': 'https://www.funimation.com/en/shows/hacksign/role-play/',
         'only_matching': True,
+    }, {
+        'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id').replace('/', '_')
+        if not mobj.group('lang'):
+            url = '%s/en/%s' % (mobj.group('origin'), mobj.group('path'))
+
+        webpage = self._download_webpage(url, display_id)
+        title_data = self._parse_json(self._search_regex(
+            r'TITLE_DATA\s*=\s*({[^}]+})',
+            webpage, 'title data', default=''),
+            display_id, js_to_json, fatal=False) or {}
+
+        video_id = (
+            title_data.get('id')
+            or self._search_regex(
+                (r"KANE_customdimensions.videoID\s*=\s*'(\d+)';", r'<iframe[^>]+src="/player/(\d+)'),
+                webpage, 'video_id', default=None)
+            or self._search_regex(
+                r'/player/(\d+)',
+                self._html_search_meta(['al:web:url', 'og:video:url', 'og:video:secure_url'], webpage, fatal=True),
+                'video id'))
+        return self.url_result(f'https://www.funimation.com/player/{video_id}', FunimationIE.ie_key(), video_id)
+
+
+class FunimationIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?funimation\.com/player/(?P<id>\d+)'
+
+    _NETRC_MACHINE = 'funimation'
+    _TOKEN = None
+
+    _TESTS = [{
+        'url': 'https://www.funimation.com/player/210051',
+        'info_dict': {
+            'id': '210050',
+            'display_id': 'broadcast-dub-preview',
+            'ext': 'mp4',
+            'title': 'Broadcast Dub Preview',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+            'episode': 'Broadcast Dub Preview',
+            'episode_id': '210050',
+            'season': 'Extras',
+            'season_id': '166038',
+            'season_number': 99,
+            'series': 'Attack on Titan: Junior High',
+            'description': '',
+            'duration': 154,
+        },
+        'params': {
+            'skip_download': 'm3u8',
+        },
+    }, {
+        'note': 'player_id should be extracted with the relevent compat-opt',
+        'url': 'https://www.funimation.com/player/210051',
+        'info_dict': {
+            'id': '210051',
+            'display_id': 'broadcast-dub-preview',
+            'ext': 'mp4',
+            'title': 'Broadcast Dub Preview',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+            'episode': 'Broadcast Dub Preview',
+            'episode_id': '210050',
+            'season': 'Extras',
+            'season_id': '166038',
+            'season_number': 99,
+            'series': 'Attack on Titan: Junior High',
+            'description': '',
+            'duration': 154,
+        },
+        'params': {
+            'skip_download': 'm3u8',
+            'compat_opts': ['seperate-video-versions'],
+        },
     }]
 
     def _login(self):
@@ -78,81 +143,184 @@ class FunimationIE(InfoExtractor):
     def _real_initialize(self):
         self._login()
 
+    @staticmethod
+    def _get_experiences(episode):
+        for lang, lang_data in episode.get('languages', {}).items():
+            for video_data in lang_data.values():
+                for version, f in video_data.items():
+                    yield lang, version.title(), f
+
+    def _get_episode(self, webpage, experience_id=None, episode_id=None, fatal=True):
+        ''' Extract the episode, season and show objects given either episode/experience id '''
+        show = self._parse_json(
+            self._search_regex(
+                r'show\s*=\s*({.+?})\s*;', webpage, 'show data', fatal=fatal),
+            experience_id, transform_source=js_to_json, fatal=fatal) or []
+        for season in show.get('seasons', []):
+            for episode in season.get('episodes', []):
+                if episode_id is not None:
+                    if str(episode.get('episodePk')) == episode_id:
+                        return episode, season, show
+                    continue
+                for _, _, f in self._get_experiences(episode):
+                    if f.get('experienceId') == experience_id:
+                        return episode, season, show
+        if fatal:
+            raise ExtractorError('Unable to find episode information')
+        else:
+            self.report_warning('Unable to find episode information')
+        return {}, {}, {}
+
     def _real_extract(self, url):
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
+        initial_experience_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, initial_experience_id, note=f'Downloading player webpage for {initial_experience_id}')
+        episode, season, show = self._get_episode(webpage, experience_id=int(initial_experience_id))
+        episode_id = str(episode['episodePk'])
+        display_id = episode.get('slug') or episode_id
 
-        def _search_kane(name):
-            return self._search_regex(
-                r"KANE_customdimensions\.%s\s*=\s*'([^']+)';" % name,
-                webpage, name, default=None)
+        formats, subtitles, thumbnails, duration = [], {}, [], 0
+        requested_languages, requested_versions = self._configuration_arg('language'), self._configuration_arg('version')
+        only_initial_experience = 'seperate-video-versions' in self.get_param('compat_opts', [])
 
-        title_data = self._parse_json(self._search_regex(
-            r'TITLE_DATA\s*=\s*({[^}]+})',
-            webpage, 'title data', default=''),
-            display_id, js_to_json, fatal=False) or {}
+        for lang, version, fmt in self._get_experiences(episode):
+            experience_id = str(fmt['experienceId'])
+            if (only_initial_experience and experience_id != initial_experience_id
+                    or requested_languages and lang.lower() not in requested_languages
+                    or requested_versions and version.lower() not in requested_versions):
+                continue
+            thumbnails.append({'url': fmt.get('poster')})
+            duration = max(duration, fmt.get('duration', 0))
+            format_name = '%s %s (%s)' % (version, lang, experience_id)
+            self.extract_subtitles(
+                subtitles, experience_id, display_id=display_id, format_name=format_name,
+                episode=episode if experience_id == initial_experience_id else episode_id)
 
-        video_id = title_data.get('id') or self._search_regex([
-            r"KANE_customdimensions.videoID\s*=\s*'(\d+)';",
-            r'<iframe[^>]+src="/player/(\d+)',
-        ], webpage, 'video_id', default=None)
-        if not video_id:
-            player_url = self._html_search_meta([
-                'al:web:url',
-                'og:video:url',
-                'og:video:secure_url',
-            ], webpage, fatal=True)
-            video_id = self._search_regex(r'/player/(\d+)', player_url, 'video id')
-
-        title = episode = title_data.get('title') or _search_kane('videoTitle') or self._og_search_title(webpage)
-        series = _search_kane('showName')
-        if series:
-            title = '%s - %s' % (series, title)
-        description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
-
-        try:
             headers = {}
             if self._TOKEN:
                 headers['Authorization'] = 'Token %s' % self._TOKEN
-            sources = self._download_json(
-                'https://www.funimation.com/api/showexperience/%s/' % video_id,
-                video_id, headers=headers, query={
+            page = self._download_json(
+                'https://www.funimation.com/api/showexperience/%s/' % experience_id,
+                display_id, headers=headers, expected_status=403, query={
                     'pinst_id': ''.join([random.choice(string.digits + string.ascii_letters) for _ in range(8)]),
-                })['items']
-        except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
-                error = self._parse_json(e.cause.read(), video_id)['errors'][0]
-                raise ExtractorError('%s said: %s' % (
-                    self.IE_NAME, error.get('detail') or error.get('title')), expected=True)
-            raise
+                }, note=f'Downloading {format_name} JSON')
+            sources = page.get('items') or []
+            if not sources:
+                error = try_get(page, lambda x: x['errors'][0], dict)
+                if error:
+                    self.report_warning('%s said: Error %s - %s' % (
+                        self.IE_NAME, error.get('code'), error.get('detail') or error.get('title')))
+                else:
+                    self.report_warning('No sources found for format')
 
-        formats = []
-        for source in sources:
-            source_url = source.get('src')
-            if not source_url:
-                continue
-            source_type = source.get('videoType') or determine_ext(source_url)
-            if source_type == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    source_url, video_id, 'mp4',
-                    m3u8_id='hls', fatal=False))
-            else:
-                formats.append({
-                    'format_id': source_type,
-                    'url': source_url,
-                })
+            current_formats = []
+            for source in sources:
+                source_url = source.get('src')
+                source_type = source.get('videoType') or determine_ext(source_url)
+                if source_type == 'm3u8':
+                    current_formats.extend(self._extract_m3u8_formats(
+                        source_url, display_id, 'mp4', m3u8_id='%s-%s' % (experience_id, 'hls'), fatal=False,
+                        note=f'Downloading {format_name} m3u8 information'))
+                else:
+                    current_formats.append({
+                        'format_id': '%s-%s' % (experience_id, source_type),
+                        'url': source_url,
+                    })
+                for f in current_formats:
+                    # TODO: Convert language to code
+                    f.update({'language': lang, 'format_note': version})
+                formats.extend(current_formats)
+        self._remove_duplicate_formats(formats)
         self._sort_formats(formats)
 
         return {
-            'id': video_id,
+            'id': initial_experience_id if only_initial_experience else episode_id,
             'display_id': display_id,
-            'title': title,
-            'description': description,
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'series': series,
-            'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
-            'episode_number': int_or_none(title_data.get('episodeNum')),
-            'episode': episode,
-            'season_id': title_data.get('seriesId'),
+            'duration': duration,
+            'title': episode['episodeTitle'],
+            'description': episode.get('episodeSummary'),
+            'episode': episode.get('episodeTitle'),
+            'episode_number': int_or_none(episode.get('episodeId')),
+            'episode_id': episode_id,
+            'season': season.get('seasonTitle'),
+            'season_number': int_or_none(season.get('seasonId')),
+            'season_id': str_or_none(season.get('seasonPk')),
+            'series': show.get('showTitle'),
             'formats': formats,
+            'thumbnails': thumbnails,
+            'subtitles': subtitles,
+        }
+
+    def _get_subtitles(self, subtitles, experience_id, episode, display_id, format_name):
+        if isinstance(episode, str):
+            webpage = self._download_webpage(
+                f'https://www.funimation.com/player/{experience_id}', display_id,
+                fatal=False, note=f'Downloading player webpage for {format_name}')
+            episode, _, _ = self._get_episode(webpage, episode_id=episode, fatal=False)
+
+        for _, version, f in self._get_experiences(episode):
+            for source in f.get('sources'):
+                for text_track in source.get('textTracks'):
+                    if not text_track.get('src'):
+                        continue
+                    sub_type = text_track.get('type').upper()
+                    sub_type = sub_type if sub_type != 'FULL' else None
+                    current_sub = {
+                        'url': text_track['src'],
+                        'name': ' '.join(filter(None, (version, text_track.get('label'), sub_type)))
+                    }
+                    lang = '_'.join(filter(None, (
+                        text_track.get('language', 'und'), version if version != 'Simulcast' else None, sub_type)))
+                    if current_sub not in subtitles.get(lang, []):
+                        subtitles.setdefault(lang, []).append(current_sub)
+        return subtitles
+
+
+class FunimationShowIE(FunimationIE):
+    IE_NAME = 'funimation:show'
+    _VALID_URL = r'(?P<url>https?://(?:www\.)?funimation(?:\.com|now\.uk)/(?P<locale>[^/]+)?/?shows/(?P<id>[^/?#&]+))/?(?:[?#]|$)'
+
+    _TESTS = [{
+        'url': 'https://www.funimation.com/en/shows/sk8-the-infinity',
+        'info_dict': {
+            'id': 1315000,
+            'title': 'SK8 the Infinity'
+        },
+        'playlist_count': 13,
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # without lang code
+        'url': 'https://www.funimation.com/shows/ouran-high-school-host-club/',
+        'info_dict': {
+            'id': 39643,
+            'title': 'Ouran High School Host Club'
+        },
+        'playlist_count': 26,
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        base_url, locale, display_id = re.match(self._VALID_URL, url).groups()
+
+        show_info = self._download_json(
+            'https://title-api.prd.funimationsvc.com/v2/shows/%s?region=US&deviceType=web&locale=%s'
+            % (display_id, locale or 'en'), display_id)
+        items = self._download_json(
+            'https://prod-api-funimationnow.dadcdigital.com/api/funimation/episodes/?limit=99999&title_id=%s'
+            % show_info.get('id'), display_id).get('items')
+        vod_items = map(lambda k: dict_get(k, ('mostRecentSvod', 'mostRecentAvod')).get('item'), items)
+
+        return {
+            '_type': 'playlist',
+            'id': show_info['id'],
+            'title': show_info['name'],
+            'entries': [
+                self.url_result(
+                    '%s/%s' % (base_url, vod_item.get('episodeSlug')), FunimationPageIE.ie_key(),
+                    vod_item.get('episodeId'), vod_item.get('episodeName'))
+                for vod_item in sorted(vod_items, key=lambda x: x.get('episodeOrder'))],
         }

yt_dlp/extractor/generic.py

@@ -84,7 +84,6 @@ from .jwplatform import JWPlatformIE
 from .digiteka import DigitekaIE
 from .arkena import ArkenaIE
 from .instagram import InstagramIE
-from .liveleak import LiveLeakIE
 from .threeqsdn import ThreeQSDNIE
 from .theplatform import ThePlatformIE
 from .kaltura import KalturaIE
@@ -1632,31 +1631,6 @@ class GenericIE(InfoExtractor):
                 'upload_date': '20160409',
             },
         },
-        # LiveLeak embed
-        {
-            'url': 'http://www.wykop.pl/link/3088787/',
-            'md5': '7619da8c820e835bef21a1efa2a0fc71',
-            'info_dict': {
-                'id': '874_1459135191',
-                'ext': 'mp4',
-                'title': 'Man shows poor quality of new apartment building',
-                'description': 'The wall is like a sand pile.',
-                'uploader': 'Lake8737',
-            },
-            'add_ie': [LiveLeakIE.ie_key()],
-        },
-        # Another LiveLeak embed pattern (#13336)
-        {
-            'url': 'https://milo.yiannopoulos.net/2017/06/concealed-carry-robbery/',
-            'info_dict': {
-                'id': '2eb_1496309988',
-                'ext': 'mp4',
-                'title': 'Thief robs place where everyone was armed',
-                'description': 'md5:694d73ee79e535953cf2488562288eee',
-                'uploader': 'brazilwtf',
-            },
-            'add_ie': [LiveLeakIE.ie_key()],
-        },
         # Duplicated embedded video URLs
         {
             'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443',
@@ -2488,7 +2462,7 @@ class GenericIE(InfoExtractor):
 
         # Is it an M3U playlist?
         if first_bytes.startswith(b'#EXTM3U'):
-            info_dict['formats'] = self._extract_m3u8_formats(url, video_id, 'mp4')
+            info_dict['formats'], info_dict['subtitles'] = self._extract_m3u8_formats_and_subtitles(url, video_id, 'mp4')
             self._sort_formats(info_dict['formats'])
             return info_dict
 
@@ -3204,11 +3178,6 @@ class GenericIE(InfoExtractor):
             return self.url_result(
                 self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key())
 
-        # Look for LiveLeak embeds
-        liveleak_urls = LiveLeakIE._extract_urls(webpage)
-        if liveleak_urls:
-            return self.playlist_from_matches(liveleak_urls, video_id, video_title)
-
         # Look for 3Q SDN embeds
         threeqsdn_url = ThreeQSDNIE._extract_url(webpage)
         if threeqsdn_url:
@@ -3441,6 +3410,7 @@ class GenericIE(InfoExtractor):
             if not isinstance(sources, list):
                 sources = [sources]
             formats = []
+            subtitles = {}
             for source in sources:
                 src = source.get('src')
                 if not src or not isinstance(src, compat_str):
@@ -3453,12 +3423,16 @@ class GenericIE(InfoExtractor):
                 if src_type == 'video/youtube':
                     return self.url_result(src, YoutubeIE.ie_key())
                 if src_type == 'application/dash+xml' or ext == 'mpd':
-                    formats.extend(self._extract_mpd_formats(
-                        src, video_id, mpd_id='dash', fatal=False))
+                    fmts, subs = self._extract_mpd_formats_and_subtitles(
+                        src, video_id, mpd_id='dash', fatal=False)
+                    formats.extend(fmts)
+                    self._merge_subtitles(subs, target=subtitles)
                 elif src_type == 'application/x-mpegurl' or ext == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
+                    fmts, subs = self._extract_m3u8_formats_and_subtitles(
                         src, video_id, 'mp4', entry_protocol='m3u8_native',
-                        m3u8_id='hls', fatal=False))
+                        m3u8_id='hls', fatal=False)
+                    formats.extend(fmts)
+                    self._merge_subtitles(subs, target=subtitles)
                 else:
                     formats.append({
                         'url': src,
@@ -3468,9 +3442,10 @@ class GenericIE(InfoExtractor):
                             'Referer': full_response.geturl(),
                         },
                     })
-            if formats:
+            if formats or subtitles:
                 self._sort_formats(formats)
                 info_dict['formats'] = formats
+                info_dict['subtitles'] = subtitles
                 return info_dict
 
         # Looking for http://schema.org/VideoObject
@@ -3605,13 +3580,13 @@ class GenericIE(InfoExtractor):
 
             ext = determine_ext(video_url)
             if ext == 'smil':
-                entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id)
+                entry_info_dict = {**self._extract_smil_info(video_url, video_id), **entry_info_dict}
             elif ext == 'xspf':
                 return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
             elif ext == 'm3u8':
-                entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4')
+                entry_info_dict['formats'], entry_info_dict['subtitles'] = self._extract_m3u8_formats_and_subtitles(video_url, video_id, ext='mp4')
             elif ext == 'mpd':
-                entry_info_dict['formats'] = self._extract_mpd_formats(video_url, video_id)
+                entry_info_dict['formats'], entry_info_dict['subtitles'] = self._extract_mpd_formats_and_subtitles(video_url, video_id)
             elif ext == 'f4m':
                 entry_info_dict['formats'] = self._extract_f4m_formats(video_url, video_id)
             elif re.search(r'(?i)\.(?:ism|smil)/manifest', video_url) and video_url != url:

yt_dlp/extractor/hotstar.py

@@ -27,8 +27,8 @@ from ..utils import (
 class HotStarBaseIE(InfoExtractor):
     _AKAMAI_ENCRYPTION_KEY = b'\x05\xfc\x1a\x01\xca\xc9\x4b\xc4\x12\xfc\x53\x12\x07\x75\xf9\xee'
 
-    def _call_api_impl(self, path, video_id, query):
-        st = int(time.time())
+    def _call_api_impl(self, path, video_id, query, st=None):
+        st = int_or_none(st) or int(time.time())
         exp = st + 6000
         auth = 'st=%d~exp=%d~acl=/*' % (st, exp)
         auth += '~hmac=' + hmac.new(self._AKAMAI_ENCRYPTION_KEY, auth.encode(), hashlib.sha256).hexdigest()
@@ -75,9 +75,9 @@ class HotStarBaseIE(InfoExtractor):
             'tas': 10000,
         })
 
-    def _call_api_v2(self, path, video_id):
+    def _call_api_v2(self, path, video_id, st=None):
         return self._call_api_impl(
-            '%s/content/%s' % (path, video_id), video_id, {
+            '%s/content/%s' % (path, video_id), video_id, st=st, query={
                 'desired-config': 'audio_channel:stereo|dynamic_range:sdr|encryption:plain|ladder:tv|package:dash|resolution:hd|subs-tag:HotstarVIP|video_codec:vp9',
                 'device-id': compat_str(uuid.uuid4()),
                 'os-name': 'Windows',
@@ -131,7 +131,8 @@ class HotStarIE(HotStarBaseIE):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage, urlh = self._download_webpage_handle(url, video_id)
+        st = urlh.headers.get('x-origin-date')
         app_state = self._parse_json(self._search_regex(
             r'<script>window\.APP_STATE\s*=\s*({.+?})</script>',
             webpage, 'app state'), video_id)
@@ -155,7 +156,7 @@ class HotStarIE(HotStarBaseIE):
         formats = []
         geo_restricted = False
         # change to v2 in the future
-        playback_sets = self._call_api_v2('play/v1/playback', video_id)['playBackSets']
+        playback_sets = self._call_api_v2('play/v1/playback', video_id, st=st)['playBackSets']
         for playback_set in playback_sets:
             if not isinstance(playback_set, dict):
                 continue

yt_dlp/extractor/instagram.py

@@ -19,6 +19,7 @@ from ..utils import (
     std_headers,
     try_get,
     url_or_none,
+    variadic,
 )
 
 
@@ -188,9 +189,7 @@ class InstagramIE(InfoExtractor):
             uploader_id = media.get('owner', {}).get('username')
 
             def get_count(keys, kind):
-                if not isinstance(keys, (list, tuple)):
-                    keys = [keys]
-                for key in keys:
+                for key in variadic(keys):
                     count = int_or_none(try_get(
                         media, (lambda x: x['edge_media_%s' % key]['count'],
                                 lambda x: x['%ss' % kind]['count'])))

yt_dlp/extractor/liveleak.py

@@ -1,191 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import int_or_none
-
-
-class LiveLeakIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?.*?\b[it]=(?P<id>[\w_]+)'
-    _TESTS = [{
-        'url': 'http://www.liveleak.com/view?i=757_1364311680',
-        'md5': '0813c2430bea7a46bf13acf3406992f4',
-        'info_dict': {
-            'id': '757_1364311680',
-            'ext': 'mp4',
-            'description': 'extremely bad day for this guy..!',
-            'uploader': 'ljfriel2',
-            'title': 'Most unlucky car accident',
-            'thumbnail': r're:^https?://.*\.jpg$'
-        }
-    }, {
-        'url': 'http://www.liveleak.com/view?i=f93_1390833151',
-        'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
-        'info_dict': {
-            'id': 'f93_1390833151',
-            'ext': 'mp4',
-            'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.',
-            'uploader': 'ARD_Stinkt',
-            'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
-            'thumbnail': r're:^https?://.*\.jpg$'
-        }
-    }, {
-        # Prochan embed
-        'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
-        'md5': '42c6d97d54f1db107958760788c5f48f',
-        'info_dict': {
-            'id': '4f7_1392687779',
-            'ext': 'mp4',
-            'description': "The guy with the cigarette seems amazingly nonchalant about the whole thing...  I really hope my friends' reactions would be a bit stronger.\r\n\r\nAction-go to 0:55.",
-            'uploader': 'CapObveus',
-            'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck',
-            'age_limit': 18,
-        },
-        'skip': 'Video is dead',
-    }, {
-        # Covers https://github.com/ytdl-org/youtube-dl/pull/5983
-        # Multiple resolutions
-        'url': 'http://www.liveleak.com/view?i=801_1409392012',
-        'md5': 'c3a449dbaca5c0d1825caecd52a57d7b',
-        'info_dict': {
-            'id': '801_1409392012',
-            'ext': 'mp4',
-            'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
-            'uploader': 'bony333',
-            'title': 'Crazy Hungarian tourist films close call waterspout in Croatia',
-            'thumbnail': r're:^https?://.*\.jpg$'
-        }
-    }, {
-        # Covers https://github.com/ytdl-org/youtube-dl/pull/10664#issuecomment-247439521
-        'url': 'http://m.liveleak.com/view?i=763_1473349649',
-        'add_ie': ['Youtube'],
-        'info_dict': {
-            'id': '763_1473349649',
-            'ext': 'mp4',
-            'title': 'Reporters and public officials ignore epidemic of black on asian violence in Sacramento | Colin Flaherty',
-            'description': 'Colin being the warrior he is and showing the injustice Asians in Sacramento are being subjected to.',
-            'uploader': 'Ziz',
-            'upload_date': '20160908',
-            'uploader_id': 'UCEbta5E_jqlZmEJsriTEtnw'
-        },
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'https://www.liveleak.com/view?i=677_1439397581',
-        'info_dict': {
-            'id': '677_1439397581',
-            'title': 'Fuel Depot in China Explosion caught on video',
-        },
-        'playlist_count': 3,
-    }, {
-        'url': 'https://www.liveleak.com/view?t=HvHi_1523016227',
-        'only_matching': True,
-    }, {
-        # No original video
-        'url': 'https://www.liveleak.com/view?t=C26ZZ_1558612804',
-        'only_matching': True,
-    }]
-
-    @staticmethod
-    def _extract_urls(webpage):
-        return re.findall(
-            r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[ift]=[\w_]+[^"]+)"',
-            webpage)
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
-        video_description = self._og_search_description(webpage)
-        video_uploader = self._html_search_regex(
-            r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
-        age_limit = int_or_none(self._search_regex(
-            r'you confirm that you are ([0-9]+) years and over.',
-            webpage, 'age limit', default=None))
-        video_thumbnail = self._og_search_thumbnail(webpage)
-
-        entries = self._parse_html5_media_entries(url, webpage, video_id)
-        if not entries:
-            # Maybe an embed?
-            embed_url = self._search_regex(
-                r'<iframe[^>]+src="((?:https?:)?//(?:www\.)?(?:prochan|youtube)\.com/embed[^"]+)"',
-                webpage, 'embed URL')
-            return {
-                '_type': 'url_transparent',
-                'url': embed_url,
-                'id': video_id,
-                'title': video_title,
-                'description': video_description,
-                'uploader': video_uploader,
-                'age_limit': age_limit,
-            }
-
-        for idx, info_dict in enumerate(entries):
-            formats = []
-            for a_format in info_dict['formats']:
-                if not a_format.get('height'):
-                    a_format['height'] = int_or_none(self._search_regex(
-                        r'([0-9]+)p\.mp4', a_format['url'], 'height label',
-                        default=None))
-                formats.append(a_format)
-
-                # Removing '.*.mp4' gives the raw video, which is essentially
-                # the same video without the LiveLeak logo at the top (see
-                # https://github.com/ytdl-org/youtube-dl/pull/4768)
-                orig_url = re.sub(r'\.mp4\.[^.]+', '', a_format['url'])
-                if a_format['url'] != orig_url:
-                    format_id = a_format.get('format_id')
-                    format_id = 'original' + ('-' + format_id if format_id else '')
-                    if self._is_valid_url(orig_url, video_id, format_id):
-                        formats.append({
-                            'format_id': format_id,
-                            'url': orig_url,
-                            'quality': 1,
-                        })
-            self._sort_formats(formats)
-            info_dict['formats'] = formats
-
-            # Don't append entry ID for one-video pages to keep backward compatibility
-            if len(entries) > 1:
-                info_dict['id'] = '%s_%s' % (video_id, idx + 1)
-            else:
-                info_dict['id'] = video_id
-
-            info_dict.update({
-                'title': video_title,
-                'description': video_description,
-                'uploader': video_uploader,
-                'age_limit': age_limit,
-                'thumbnail': video_thumbnail,
-            })
-
-        return self.playlist_result(entries, video_id, video_title)
-
-
-class LiveLeakEmbedIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[ift])=(?P<id>[\w_]+)'
-
-    # See generic.py for actual test cases
-    _TESTS = [{
-        'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        kind, video_id = re.match(self._VALID_URL, url).groups()
-
-        if kind == 'f':
-            webpage = self._download_webpage(url, video_id)
-            liveleak_url = self._search_regex(
-                r'(?:logourl\s*:\s*|window\.open\()(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL,
-                webpage, 'LiveLeak URL', group='url')
-        else:
-            liveleak_url = 'http://www.liveleak.com/view?%s=%s' % (kind, video_id)
-
-        return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())

yt_dlp/extractor/mediasite.py

@@ -122,6 +122,52 @@ class MediasiteIE(InfoExtractor):
                 r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE,
                 webpage)]
 
+    def __extract_slides(self, *, stream_id, snum, Stream, duration, images):
+        slide_base_url = Stream['SlideBaseUrl']
+
+        fname_template = Stream['SlideImageFileNameTemplate']
+        if fname_template != 'slide_{0:D4}.jpg':
+            self.report_warning('Unusual slide file name template; report a bug if slide downloading fails')
+        fname_template = re.sub(r'\{0:D([0-9]+)\}', r'{0:0\1}', fname_template)
+
+        fragments = []
+        for i, slide in enumerate(Stream['Slides']):
+            if i == 0:
+                if slide['Time'] > 0:
+                    default_slide = images.get('DefaultSlide')
+                    if default_slide is None:
+                        default_slide = images.get('DefaultStreamImage')
+                    if default_slide is not None:
+                        default_slide = default_slide['ImageFilename']
+                    if default_slide is not None:
+                        fragments.append({
+                            'path': default_slide,
+                            'duration': slide['Time'] / 1000,
+                        })
+
+            next_time = try_get(None, [
+                lambda _: Stream['Slides'][i + 1]['Time'],
+                lambda _: duration,
+                lambda _: slide['Time'],
+            ], expected_type=(int, float))
+
+            fragments.append({
+                'path': fname_template.format(slide.get('Number', i + 1)),
+                'duration': (next_time - slide['Time']) / 1000
+            })
+
+        return {
+            'format_id': '%s-%u.slides' % (stream_id, snum),
+            'ext': 'mhtml',
+            'url': slide_base_url,
+            'protocol': 'mhtml',
+            'acodec': 'none',
+            'vcodec': 'none',
+            'format_note': 'Slides',
+            'fragments': fragments,
+            'fragment_base_url': slide_base_url,
+        }
+
     def _real_extract(self, url):
         url, data = unsmuggle_url(url, {})
         mobj = re.match(self._VALID_URL, url)
@@ -198,10 +244,15 @@ class MediasiteIE(InfoExtractor):
                         'ext': mimetype2ext(VideoUrl.get('MimeType')),
                     })
 
-            # TODO: if Stream['HasSlideContent']:
-            # synthesise an MJPEG video stream '%s-%u.slides' % (stream_type, snum)
-            # from Stream['Slides']
-            # this will require writing a custom downloader...
+            if Stream.get('HasSlideContent', False):
+                images = player_options['PlayerLayoutOptions']['Images']
+                stream_formats.append(self.__extract_slides(
+                    stream_id=stream_id,
+                    snum=snum,
+                    Stream=Stream,
+                    duration=presentation.get('Duration'),
+                    images=images,
+                ))
 
             # disprefer 'secondary' streams
             if stream_type != 0:

yt_dlp/extractor/mtv.py

@@ -249,6 +249,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
             if info:
                 entries.append(info)
 
+        # TODO: should be multi-video
         return self.playlist_result(
             entries, playlist_title=title, playlist_description=description)
 

yt_dlp/extractor/nebula.py

@@ -2,9 +2,11 @@
 from __future__ import unicode_literals
 
 import json
+import time
 
+from urllib.error import HTTPError
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import compat_str, compat_urllib_parse_unquote, compat_urllib_parse_quote
 from ..utils import (
     ExtractorError,
     parse_iso8601,
@@ -78,7 +80,9 @@ class NebulaIE(InfoExtractor):
     ]
     _NETRC_MACHINE = 'watchnebula'
 
-    def _retrieve_nebula_auth(self, video_id):
+    _nebula_token = None
+
+    def _retrieve_nebula_auth(self):
         """
         Log in to Nebula, and returns a Nebula API token
         """
@@ -91,7 +95,7 @@ class NebulaIE(InfoExtractor):
         data = json.dumps({'email': username, 'password': password}).encode('utf8')
         response = self._download_json(
             'https://api.watchnebula.com/api/v1/auth/login/',
-            data=data, fatal=False, video_id=video_id,
+            data=data, fatal=False, video_id=None,
             headers={
                 'content-type': 'application/json',
                 # Submitting the 'sessionid' cookie always causes a 403 on auth endpoint
@@ -101,6 +105,19 @@ class NebulaIE(InfoExtractor):
             errnote='Authentication failed or rejected')
         if not response or not response.get('key'):
             self.raise_login_required()
+
+        # save nebula token as cookie
+        self._set_cookie(
+            'nebula.app', 'nebula-auth',
+            compat_urllib_parse_quote(
+                json.dumps({
+                    "apiToken": response["key"],
+                    "isLoggingIn": False,
+                    "isLoggingOut": False,
+                }, separators=(",", ":"))),
+            expire_time=int(time.time()) + 86400 * 365,
+        )
+
         return response['key']
 
     def _retrieve_zype_api_key(self, page_url, display_id):
@@ -139,8 +156,17 @@ class NebulaIE(InfoExtractor):
             'Authorization': 'Token {access_token}'.format(access_token=access_token)
         }, note=note)
 
-    def _fetch_zype_access_token(self, video_id, nebula_token):
-        user_object = self._call_nebula_api('/auth/user/', video_id, nebula_token, note='Retrieving Zype access token')
+    def _fetch_zype_access_token(self, video_id):
+        try:
+            user_object = self._call_nebula_api('/auth/user/', video_id, self._nebula_token, note='Retrieving Zype access token')
+        except ExtractorError as exc:
+            # if 401, attempt credential auth and retry
+            if exc.cause and isinstance(exc.cause, HTTPError) and exc.cause.code == 401:
+                self._nebula_token = self._retrieve_nebula_auth()
+                user_object = self._call_nebula_api('/auth/user/', video_id, self._nebula_token, note='Retrieving Zype access token')
+            else:
+                raise
+
         access_token = try_get(user_object, lambda x: x['zype_auth_info']['access_token'], compat_str)
         if not access_token:
             if try_get(user_object, lambda x: x['is_subscribed'], bool):
@@ -162,9 +188,21 @@ class NebulaIE(InfoExtractor):
             if category.get('value'):
                 return category['value'][0]
 
+    def _real_initialize(self):
+        # check cookie jar for valid token
+        nebula_cookies = self._get_cookies('https://nebula.app')
+        nebula_cookie = nebula_cookies.get('nebula-auth')
+        if nebula_cookie:
+            self.to_screen('Authenticating to Nebula with token from cookie jar')
+            nebula_cookie_value = compat_urllib_parse_unquote(nebula_cookie.value)
+            self._nebula_token = self._parse_json(nebula_cookie_value, None).get('apiToken')
+
+        # try to authenticate using credentials if no valid token has been found
+        if not self._nebula_token:
+            self._nebula_token = self._retrieve_nebula_auth()
+
     def _real_extract(self, url):
         display_id = self._match_id(url)
-        nebula_token = self._retrieve_nebula_auth(display_id)
         api_key = self._retrieve_zype_api_key(url, display_id)
 
         response = self._call_zype_api('/videos', {'friendly_title': display_id},
@@ -174,7 +212,7 @@ class NebulaIE(InfoExtractor):
         video_meta = response['response'][0]
 
         video_id = video_meta['_id']
-        zype_access_token = self._fetch_zype_access_token(display_id, nebula_token=nebula_token)
+        zype_access_token = self._fetch_zype_access_token(display_id)
 
         channel_title = self._extract_channel_title(video_meta)
 
@@ -187,13 +225,12 @@ class NebulaIE(InfoExtractor):
             'title': video_meta.get('title'),
             'description': video_meta.get('description'),
             'timestamp': parse_iso8601(video_meta.get('published_at')),
-            'thumbnails': [
-                {
-                    'id': tn.get('name'),  # this appears to be null
-                    'url': tn['url'],
-                    'width': tn.get('width'),
-                    'height': tn.get('height'),
-                } for tn in video_meta.get('thumbnails', [])],
+            'thumbnails': [{
+                'id': tn.get('name'),  # this appears to be null
+                'url': tn['url'],
+                'width': tn.get('width'),
+                'height': tn.get('height'),
+            } for tn in video_meta.get('thumbnails', [])],
             'duration': video_meta.get('duration'),
             'channel': channel_title,
             'uploader': channel_title,  # we chose uploader = channel name

yt_dlp/extractor/nrk.py

@@ -58,7 +58,7 @@ class NRKBaseIE(InfoExtractor):
 
     def _call_api(self, path, video_id, item=None, note=None, fatal=True, query=None):
         return self._download_json(
-            urljoin('http://psapi.nrk.no/', path),
+            urljoin('https://psapi.nrk.no/', path),
             video_id, note or 'Downloading %s JSON' % item,
             fatal=fatal, query=query,
             headers={'Accept-Encoding': 'gzip, deflate, br'})

yt_dlp/extractor/orf.py

@@ -98,6 +98,9 @@ class ORFTVthekIE(InfoExtractor):
                 elif ext == 'f4m':
                     formats.extend(self._extract_f4m_formats(
                         src, video_id, f4m_id=format_id, fatal=False))
+                elif ext == 'mpd':
+                    formats.extend(self._extract_mpd_formats(
+                        src, video_id, mpd_id=format_id, fatal=False))
                 else:
                     formats.append({
                         'format_id': format_id,

yt_dlp/extractor/peertube.py

@@ -569,15 +569,15 @@ class PeerTubeIE(InfoExtractor):
             formats.append(f)
         self._sort_formats(formats)
 
-        full_description = self._call_api(
-            host, video_id, 'description', note='Downloading description JSON',
-            fatal=False)
+        description = video.get('description')
+        if len(description) >= 250:
+            # description is shortened
+            full_description = self._call_api(
+                host, video_id, 'description', note='Downloading description JSON',
+                fatal=False)
 
-        description = None
-        if isinstance(full_description, dict):
-            description = str_or_none(full_description.get('description'))
-        if not description:
-            description = video.get('description')
+            if isinstance(full_description, dict):
+                description = str_or_none(full_description.get('description')) or description
 
         subtitles = self.extract_subtitles(host, video_id)
 

yt_dlp/extractor/periscope.py

@@ -12,6 +12,10 @@ from ..utils import (
 
 
 class PeriscopeBaseIE(InfoExtractor):
+    _M3U8_HEADERS = {
+        'Referer': 'https://www.periscope.tv/'
+    }
+
     def _call_api(self, method, query, item_id):
         return self._download_json(
             'https://api.periscope.tv/api/v2/%s' % method,
@@ -54,9 +58,11 @@ class PeriscopeBaseIE(InfoExtractor):
             m3u8_url, video_id, 'mp4',
             entry_protocol='m3u8_native'
             if state in ('ended', 'timed_out') else 'm3u8',
-            m3u8_id=format_id, fatal=fatal)
+            m3u8_id=format_id, fatal=fatal, headers=self._M3U8_HEADERS)
         if len(m3u8_formats) == 1:
             self._add_width_and_height(m3u8_formats[0], width, height)
+        for f in m3u8_formats:
+            f.setdefault('http_headers', {}).update(self._M3U8_HEADERS)
         return m3u8_formats
 
 

yt_dlp/extractor/plutotv.py

@@ -19,7 +19,7 @@ from ..utils import (
 
 
 class PlutoTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?pluto\.tv/on-demand/(?P<video_type>movies|series)/(?P<slug>.*)/?$'
+    _VALID_URL = r'https?://(?:www\.)?pluto\.tv(?:/en)?/on-demand/(?P<video_type>movies|series)/(?P<slug>.*)/?$'
     _INFO_URL = 'https://service-vod.clusters.pluto.tv/v3/vod/slugs/'
     _INFO_QUERY_PARAMS = {
         'appName': 'web',
@@ -48,24 +48,21 @@ class PlutoTVIE(InfoExtractor):
                 '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': {
@@ -75,7 +72,10 @@ class PlutoTVIE(InfoExtractor):
                 '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,
             }
-        },
+        }, {
+            'url': 'https://pluto.tv/en/on-demand/series/manhunters-fugitive-task-force/seasons/1/episode/third-times-the-charm-1-1',
+            'only_matching': True,
+        }
     ]
 
     def _to_ad_free_formats(self, video_id, formats, subtitles):

yt_dlp/extractor/pornflip.py

@@ -0,0 +1,82 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    parse_iso8601
+)
+
+
+class PornFlipIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?pornflip\.com/(?:(embed|sv|v)/)?(?P<id>[^/]+)'
+    _TESTS = [
+        {
+            'url': 'https://www.pornflip.com/dzv9Mtw1qj2/sv/brazzers-double-dare-two-couples-fucked-jenna-reid-maya-bijou',
+            'info_dict': {
+                'id': 'dzv9Mtw1qj2',
+                'ext': 'mp4',
+                'title': 'Brazzers - Double Dare Two couples fucked Jenna Reid Maya Bijou',
+                'description': 'md5:d2b69e6cc743c5fd158e162aa7f05821',
+                'duration': 476,
+                'like_count': int,
+                'dislike_count': int,
+                'view_count': int,
+                'timestamp': 1617846819,
+                'upload_date': '20210408',
+                'uploader': 'Brazzers',
+                'age_limit': 18,
+            },
+            'params': {
+                'format': 'bestvideo',
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'https://www.pornflip.com/v/IrJEC40i21L',
+            'only_matching': True,
+        },
+        {
+            'url': 'https://www.pornflip.com/Z3jzbChC5-P/sexintaxi-e-sereyna-gomez-czech-naked-couple',
+            'only_matching': True,
+        },
+        {
+            'url': 'https://www.pornflip.com/embed/bLcDFxnrZnU',
+            'only_matching': True,
+        },
+    ]
+    _HOST = 'www.pornflip.com'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            'https://{}/sv/{}'.format(self._HOST, video_id), video_id, headers={'host': self._HOST})
+        description = self._html_search_regex(r'&p\[summary\]=(.*?)\s*&p', webpage, 'description', fatal=False)
+        duration = self._search_regex(r'"duration":\s+"([^"]+)",', webpage, 'duration', fatal=False)
+        view_count = self._search_regex(r'"interactionCount":\s+"([^"]+)"', webpage, 'view_count', fatal=False)
+        title = self._html_search_regex(r'id="mediaPlayerTitleLink"[^>]*>(.+)</a>', webpage, 'title', fatal=False)
+        uploader = self._html_search_regex(r'class="title-chanel"[^>]*>[^<]*<a[^>]*>([^<]+)<', webpage, 'uploader', fatal=False)
+        upload_date = self._search_regex(r'"uploadDate":\s+"([^"]+)",', webpage, 'upload_date', fatal=False)
+        likes = self._html_search_regex(
+            r'class="btn btn-up-rating[^>]*>[^<]*<i[^>]*>[^<]*</i>[^>]*<span[^>]*>[^0-9]*([0-9]+)[^<0-9]*<', webpage, 'like_count', fatal=False)
+        dislikes = self._html_search_regex(
+            r'class="btn btn-down-rating[^>]*>[^<]*<i[^>]*>[^<]*</i>[^>]*<span[^>]*>[^0-9]*([0-9]+)[^<0-9]*<', webpage, 'dislike_count', fatal=False)
+        mpd_url = self._search_regex(r'"([^"]+userscontent.net/dash/[0-9]+/manifest.mpd[^"]*)"', webpage, 'mpd_url').replace('&amp;', '&')
+        formats = self._extract_mpd_formats(mpd_url, video_id, mpd_id='dash')
+        self._sort_formats(formats)
+
+        return {
+            'age_limit': 18,
+            'description': description,
+            'dislike_count': int_or_none(dislikes),
+            'duration': parse_duration(duration),
+            'formats': formats,
+            'id': video_id,
+            'like_count': int_or_none(likes),
+            'timestamp': parse_iso8601(upload_date),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'title': title,
+            'uploader': uploader,
+            'view_count': int_or_none(view_count),
+        }

yt_dlp/extractor/pornhub.py

@@ -14,6 +14,7 @@ from ..compat import (
 )
 from .openload import PhantomJSwrapper
 from ..utils import (
+    clean_html,
     determine_ext,
     ExtractorError,
     int_or_none,
@@ -30,6 +31,7 @@ from ..utils import (
 
 class PornHubBaseIE(InfoExtractor):
     _NETRC_MACHINE = 'pornhub'
+    _PORNHUB_HOST_RE = r'(?:(?P<host>pornhub(?:premium)?\.(?:com|net|org))|pornhubthbh7ap3u\.onion)'
 
     def _download_webpage_handle(self, *args, **kwargs):
         def dl(*args, **kwargs):
@@ -122,11 +124,13 @@ class PornHubIE(PornHubBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
-                            (?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
+                            (?:[^/]+\.)?
+                            %s
+                            /(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
                             (?:www\.)?thumbzilla\.com/video/
                         )
                         (?P<id>[\da-z]+)
-                    '''
+                    ''' % PornHubBaseIE._PORNHUB_HOST_RE
     _TESTS = [{
         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
         'md5': 'a6391306d050e4547f62b3f485dd9ba9',
@@ -145,6 +149,7 @@ class PornHubIE(PornHubBaseIE):
             'age_limit': 18,
             'tags': list,
             'categories': list,
+            'cast': list,
         },
     }, {
         # non-ASCII title
@@ -236,6 +241,13 @@ class PornHubIE(PornHubBaseIE):
     }, {
         'url': 'https://www.pornhubpremium.com/view_video.php?viewkey=ph5f75b0f4b18e3',
         'only_matching': True,
+    }, {
+        # geo restricted
+        'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5a9813bfa7156',
+        'only_matching': True,
+    }, {
+        'url': 'http://pornhubthbh7ap3u.onion/view_video.php?viewkey=ph5a9813bfa7156',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -275,6 +287,11 @@ class PornHubIE(PornHubBaseIE):
                 'PornHub said: %s' % error_msg,
                 expected=True, video_id=video_id)
 
+        if any(re.search(p, webpage) for p in (
+                r'class=["\']geoBlocked["\']',
+                r'>\s*This content is unavailable in your country')):
+            self.raise_geo_restricted()
+
         # video_title from flashvars contains whitespace instead of non-ASCII (see
         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
         # on that anymore.
@@ -408,17 +425,14 @@ class PornHubIE(PornHubBaseIE):
                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
                     m3u8_id='hls', fatal=False))
                 return
-            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'))
+            if not height:
+                height = int_or_none(self._search_regex(
+                    r'(?P<height>\d+)[pP]?_\d+[kK]', format_url, 'height',
+                    default=None))
             formats.append({
                 'url': format_url,
                 'format_id': '%dp' % height if height else None,
                 'height': height,
-                'tbr': tbr,
             })
 
         for video_url, height in video_urls:
@@ -440,7 +454,10 @@ class PornHubIE(PornHubBaseIE):
                         add_format(video_url, height)
                 continue
             add_format(video_url)
-        self._sort_formats(formats)
+
+        # field_preference is unnecessary here, but kept for code-similarity with youtube-dl
+        self._sort_formats(
+            formats, field_preference=('height', 'width', 'fps', 'format_id'))
 
         video_uploader = self._html_search_regex(
             r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
@@ -464,7 +481,7 @@ class PornHubIE(PornHubBaseIE):
                 r'(?s)<div[^>]+\bclass=["\'].*?\b%sWrapper[^>]*>(.+?)</div>'
                 % meta_key, webpage, meta_key, default=None)
             if div:
-                return re.findall(r'<a[^>]+\bhref=[^>]+>([^<]+)', div)
+                return [clean_html(x).strip() for x in re.findall(r'(?s)<a[^>]+\bhref=[^>]+>.+?</a>', div)]
 
         info = self._search_json_ld(webpage, video_id, default={})
         # description provided in JSON-LD is irrelevant
@@ -485,6 +502,7 @@ class PornHubIE(PornHubBaseIE):
             'age_limit': 18,
             'tags': extract_list('tags'),
             'categories': extract_list('categories'),
+            'cast': extract_list('pornstars'),
             'subtitles': subtitles,
         }, info)
 
@@ -513,7 +531,7 @@ class PornHubPlaylistBaseIE(PornHubBaseIE):
 
 
 class PornHubUserIE(PornHubPlaylistBaseIE):
-    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)'
+    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?%s/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)' % PornHubBaseIE._PORNHUB_HOST_RE
     _TESTS = [{
         'url': 'https://www.pornhub.com/model/zoe_ph',
         'playlist_mincount': 118,
@@ -542,6 +560,9 @@ class PornHubUserIE(PornHubPlaylistBaseIE):
         # Same as before, multi page
         'url': 'https://www.pornhubpremium.com/pornstar/lily-labeau',
         'only_matching': True,
+    }, {
+        'url': 'https://pornhubthbh7ap3u.onion/model/zoe_ph',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -617,7 +638,7 @@ class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE):
 
 
 class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
-    _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?P<id>(?:[^/]+/)*[^/?#&]+)'
+    _VALID_URL = r'https?://(?:[^/]+\.)?%s/(?P<id>(?:[^/]+/)*[^/?#&]+)' % PornHubBaseIE._PORNHUB_HOST_RE
     _TESTS = [{
         'url': 'https://www.pornhub.com/model/zoe_ph/videos',
         'only_matching': True,
@@ -722,6 +743,9 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
     }, {
         'url': 'https://de.pornhub.com/playlist/4667351',
         'only_matching': True,
+    }, {
+        'url': 'https://pornhubthbh7ap3u.onion/model/zoe_ph/videos',
+        'only_matching': True,
     }]
 
     @classmethod
@@ -732,7 +756,7 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
 
 
 class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE):
-    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)'
+    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?%s/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)' % PornHubBaseIE._PORNHUB_HOST_RE
     _TESTS = [{
         'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
         'info_dict': {
@@ -742,4 +766,7 @@ class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE):
     }, {
         'url': 'https://www.pornhub.com/model/zoe_ph/videos/upload',
         'only_matching': True,
+    }, {
+        'url': 'http://pornhubthbh7ap3u.onion/pornstar/jenny-blighe/videos/upload',
+        'only_matching': True,
     }]

yt_dlp/extractor/rcti.py

@@ -0,0 +1,242 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import itertools
+import re
+
+from .openload import PhantomJSwrapper
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    RegexNotFoundError,
+    strip_or_none,
+    try_get
+)
+
+
+class RCTIPlusBaseIE(InfoExtractor):
+    def _real_initialize(self):
+        self._AUTH_KEY = self._download_json(
+            'https://api.rctiplus.com/api/v1/visitor?platform=web',  # platform can be web, mweb, android, ios
+            None, 'Fetching authorization key')['data']['access_token']
+
+    def _call_api(self, url, video_id, note=None):
+        json = self._download_json(
+            url, video_id, note=note, headers={'Authorization': self._AUTH_KEY})
+        if json.get('status', {}).get('code', 0) != 0:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, json["status"]["message_client"]), cause=json)
+        return json.get('data'), json.get('meta')
+
+
+class RCTIPlusIE(RCTIPlusBaseIE):
+    _VALID_URL = r'https://www\.rctiplus\.com/programs/\d+?/.*?/(?P<type>episode|clip|extra)/(?P<id>\d+)/(?P<display_id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://www.rctiplus.com/programs/1259/kiko-untuk-lola/episode/22124/untuk-lola',
+        'md5': '56ed45affad45fa18d5592a1bc199997',
+        'info_dict': {
+            'id': 'v_e22124',
+            'title': 'Untuk Lola',
+            'display_id': 'untuk-lola',
+            'description': 'md5:2b809075c0b1e071e228ad6d13e41deb',
+            'ext': 'mp4',
+            'duration': 1400,
+            'timestamp': 1615978800,
+            'upload_date': '20210317',
+            'series': 'Kiko : Untuk Lola',
+            'season_number': 1,
+            'episode_number': 1,
+            'channel': 'RCTI',
+        },
+        'params': {
+            'fixup': 'never',
+        },
+    }, {  # Clip; Series title doesn't appear on metadata JSON
+        'url': 'https://www.rctiplus.com/programs/316/cahaya-terindah/clip/3921/make-a-wish',
+        'md5': 'd179b2ff356f0e91a53bcc6a4d8504f0',
+        'info_dict': {
+            'id': 'v_c3921',
+            'title': 'Make A Wish',
+            'display_id': 'make-a-wish',
+            'description': 'Make A Wish',
+            'ext': 'mp4',
+            'duration': 288,
+            'timestamp': 1571652600,
+            'upload_date': '20191021',
+            'series': 'Cahaya Terindah',
+            'channel': 'RCTI',
+        },
+        'params': {
+            'fixup': 'never',
+        },
+    }, {  # Extra
+        'url': 'https://www.rctiplus.com/programs/616/inews-malam/extra/9438/diungkapkan-melalui-surat-terbuka-ceo-ruangguru-belva-devara-mundur-dari-staf-khusus-presiden',
+        'md5': 'c48106afdbce609749f5e0c007d9278a',
+        'info_dict': {
+            'id': 'v_ex9438',
+            'title': 'md5:2ede828c0f8bde249e0912be150314ca',
+            'display_id': 'md5:62b8d4e9ff096db527a1ad797e8a9933',
+            'description': 'md5:2ede828c0f8bde249e0912be150314ca',
+            'ext': 'mp4',
+            'duration': 93,
+            'timestamp': 1587561540,
+            'upload_date': '20200422',
+            'series': 'iNews Malam',
+            'channel': 'INews',
+        },
+        'params': {
+            'format': 'bestvideo',
+        },
+    }]
+
+    def _search_auth_key(self, webpage):
+        try:
+            self._AUTH_KEY = self._search_regex(
+                r'\'Authorization\':"(?P<auth>[^"]+)"', webpage, 'auth-key')
+        except RegexNotFoundError:
+            pass
+
+    def _real_extract(self, url):
+        video_type, video_id, display_id = re.match(self._VALID_URL, url).groups()
+        webpage = self._download_webpage(url, display_id)
+        self._search_auth_key(webpage)
+
+        video_json = self._call_api(
+            'https://api.rctiplus.com/api/v1/%s/%s/url?appierid=.1' % (video_type, video_id), display_id, 'Downloading video URL JSON')[0]
+        video_url = video_json['url']
+        if 'akamaized' in video_url:
+            # Akamai's CDN requires a session to at least be made via Conviva's API
+            # TODO: Reverse-engineer Conviva's heartbeat code to avoid phantomJS
+            phantom = None
+            try:
+                phantom = PhantomJSwrapper(self)
+                phantom.get(url, webpage, display_id, note2='Initiating video session')
+            except ExtractorError:
+                self.report_warning('PhantomJS is highly recommended for this video, as it might load incredibly slowly otherwise.'
+                                    'You can also try opening the page in this device\'s browser first')
+
+        video_meta, meta_paths = self._call_api(
+            'https://api.rctiplus.com/api/v1/%s/%s' % (video_type, video_id), display_id, 'Downloading video metadata')
+
+        thumbnails, image_path = [], meta_paths.get('image_path', 'https://rstatic.akamaized.net/media/')
+        if video_meta.get('portrait_image'):
+            thumbnails.append({
+                'id': 'portrait_image',
+                'url': '%s%d%s' % (image_path, 2000, video_meta['portrait_image'])  # 2000px seems to be the highest resolution that can be given
+            })
+        if video_meta.get('landscape_image'):
+            thumbnails.append({
+                'id': 'landscape_image',
+                'url': '%s%d%s' % (image_path, 2000, video_meta['landscape_image'])
+            })
+
+        formats = self._extract_m3u8_formats(video_url, display_id, 'mp4', headers={'Referer': 'https://www.rctiplus.com/'})
+        for f in formats:
+            if 'akamaized' in f['url']:
+                f.setdefault('http_headers', {})['Referer'] = 'https://www.rctiplus.com/'  # Referer header is required for akamai CDNs
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_meta.get('product_id') or video_json.get('product_id'),
+            'title': video_meta.get('title') or video_json.get('content_name'),
+            'display_id': display_id,
+            'description': video_meta.get('summary'),
+            'timestamp': video_meta.get('release_date'),
+            'duration': video_meta.get('duration'),
+            'categories': [video_meta.get('genre')],
+            'average_rating': video_meta.get('star_rating'),
+            'series': video_meta.get('program_title') or video_json.get('program_title'),
+            'season_number': video_meta.get('season'),
+            'episode_number': video_meta.get('episode'),
+            'channel': video_json.get('tv_name'),
+            'channel_id': video_json.get('tv_id'),
+            'formats': formats,
+            'thumbnails': thumbnails
+        }
+
+
+class RCTIPlusSeriesIE(RCTIPlusBaseIE):
+    _VALID_URL = r'https://www\.rctiplus\.com/programs/(?P<id>\d+)/(?P<display_id>[^/?#&]+)(?:\W)*$'
+    _TESTS = [{
+        'url': 'https://www.rctiplus.com/programs/540/upin-ipin',
+        'playlist_mincount': 417,
+        'info_dict': {
+            'id': '540',
+            'title': 'Upin & Ipin',
+            'description': 'md5:22cc912381f389664416844e1ec4f86b',
+        },
+    }, {
+        'url': 'https://www.rctiplus.com/programs/540/upin-ipin/#',
+        'only_matching': True,
+    }]
+    _AGE_RATINGS = {  # Based off https://id.wikipedia.org/wiki/Sistem_rating_konten_televisi with additional ratings
+        'S-SU': 2,
+        'SU': 2,
+        'P': 2,
+        'A': 7,
+        'R': 13,
+        'R-R/1': 17,  # Labelled as 17+ despite being R
+        'D': 18,
+    }
+
+    def _entries(self, url, display_id=None, note='Downloading entries JSON', metadata={}):
+        total_pages = 0
+        try:
+            total_pages = self._call_api(
+                '%s&length=20&page=0' % url,
+                display_id, note)[1]['pagination']['total_page']
+        except ExtractorError as e:
+            if 'not found' in str(e):
+                return []
+            raise e
+        if total_pages <= 0:
+            return []
+
+        for page_num in range(1, total_pages + 1):
+            episode_list = self._call_api(
+                '%s&length=20&page=%s' % (url, page_num),
+                display_id, '%s page %s' % (note, page_num))[0] or []
+
+            for video_json in episode_list:
+                link = video_json['share_link']
+                url_res = self.url_result(link, 'RCTIPlus', video_json.get('product_id'), video_json.get('title'))
+                url_res.update(metadata)
+                yield url_res
+
+    def _real_extract(self, url):
+        series_id, display_id = re.match(self._VALID_URL, url).groups()
+
+        series_meta, meta_paths = self._call_api(
+            'https://api.rctiplus.com/api/v1/program/%s/detail' % series_id, display_id, 'Downloading series metadata')
+        metadata = {
+            'age_limit': try_get(series_meta, lambda x: self._AGE_RATINGS[x['age_restriction'][0]['code']])
+        }
+
+        cast = []
+        for star in series_meta.get('starring', []):
+            cast.append(strip_or_none(star.get('name')))
+        for star in series_meta.get('creator', []):
+            cast.append(strip_or_none(star.get('name')))
+        for star in series_meta.get('writer', []):
+            cast.append(strip_or_none(star.get('name')))
+        metadata['cast'] = cast
+
+        tags = []
+        for tag in series_meta.get('tag', []):
+            tags.append(strip_or_none(tag.get('name')))
+        metadata['tag'] = tags
+
+        entries = []
+        seasons_list = self._call_api(
+            'https://api.rctiplus.com/api/v1/program/%s/season' % series_id, display_id, 'Downloading seasons list JSON')[0]
+        for season in seasons_list:
+            entries.append(self._entries('https://api.rctiplus.com/api/v2/program/%s/episode?season=%s' % (series_id, season['season']),
+                                         display_id, 'Downloading season %s episode entries' % season['season'], metadata))
+
+        entries.append(self._entries('https://api.rctiplus.com/api/v2/program/%s/clip?content_id=0' % series_id,
+                                     display_id, 'Downloading clip entries', metadata))
+        entries.append(self._entries('https://api.rctiplus.com/api/v2/program/%s/extra?content_id=0' % series_id,
+                                     display_id, 'Downloading extra entries', metadata))
+
+        return self.playlist_result(itertools.chain(*entries), series_id, series_meta.get('title'), series_meta.get('summary'), **metadata)

yt_dlp/extractor/rtp.py

@@ -2,10 +2,11 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import (
-    determine_ext,
-    js_to_json,
-)
+from ..utils import js_to_json
+import re
+import json
+import urllib.parse
+import base64
 
 
 class RTPIE(InfoExtractor):
@@ -25,6 +26,22 @@ class RTPIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    _RX_OBFUSCATION = re.compile(r'''(?xs)
+        atob\s*\(\s*decodeURIComponent\s*\(\s*
+            (\[[0-9A-Za-z%,'"]*\])
+        \s*\.\s*join\(\s*(?:""|'')\s*\)\s*\)\s*\)
+    ''')
+
+    def __unobfuscate(self, data, *, video_id):
+        if data.startswith('{'):
+            data = self._RX_OBFUSCATION.sub(
+                lambda m: json.dumps(
+                    base64.b64decode(urllib.parse.unquote(
+                        ''.join(self._parse_json(m.group(1), video_id))
+                    )).decode('iso-8859-1')),
+                data)
+        return js_to_json(data)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -32,30 +49,46 @@ class RTPIE(InfoExtractor):
         title = self._html_search_meta(
             'twitter:title', webpage, display_name='title', fatal=True)
 
-        config = self._parse_json(self._search_regex(
-            r'(?s)RTPPlayer\(({.+?})\);', webpage,
-            'player config'), video_id, js_to_json)
-        file_url = config['file']
-        ext = determine_ext(file_url)
-        if ext == 'm3u8':
-            file_key = config.get('fileKey')
-            formats = self._extract_m3u8_formats(
-                file_url, video_id, 'mp4', 'm3u8_native',
-                m3u8_id='hls', fatal=file_key)
-            if file_key:
-                formats.append({
-                    'url': 'https://cdn-ondemand.rtp.pt' + file_key,
-                    'quality': 1,
-                })
-            self._sort_formats(formats)
+        f, config = self._search_regex(
+            r'''(?sx)
+                var\s+f\s*=\s*(?P<f>".*?"|{[^;]+?});\s*
+                var\s+player1\s+=\s+new\s+RTPPlayer\s*\((?P<config>{(?:(?!\*/).)+?})\);(?!\s*\*/)
+            ''', webpage,
+            'player config', group=('f', 'config'))
+
+        f = self._parse_json(
+            f, video_id,
+            lambda data: self.__unobfuscate(data, video_id=video_id))
+        config = self._parse_json(
+            config, video_id,
+            lambda data: self.__unobfuscate(data, video_id=video_id))
+
+        formats = []
+        if isinstance(f, dict):
+            f_hls = f.get('hls')
+            if f_hls is not None:
+                formats.extend(self._extract_m3u8_formats(
+                    f_hls, video_id, 'mp4', 'm3u8_native', m3u8_id='hls'))
+
+            f_dash = f.get('dash')
+            if f_dash is not None:
+                formats.extend(self._extract_mpd_formats(f_dash, video_id, mpd_id='dash'))
         else:
-            formats = [{
-                'url': file_url,
-                'ext': ext,
-            }]
-        if config.get('mediaType') == 'audio':
-            for f in formats:
-                f['vcodec'] = 'none'
+            formats.append({
+                'format_id': 'f',
+                'url': f,
+                'vcodec': 'none' if config.get('mediaType') == 'audio' else None,
+            })
+
+        subtitles = {}
+
+        vtt = config.get('vtt')
+        if vtt is not None:
+            for lcode, lname, url in vtt:
+                subtitles.setdefault(lcode, []).append({
+                    'name': lname,
+                    'url': url,
+                })
 
         return {
             'id': video_id,
@@ -63,4 +96,5 @@ class RTPIE(InfoExtractor):
             'formats': formats,
             'description': self._html_search_meta(['description', 'twitter:description'], webpage),
             'thumbnail': config.get('poster') or self._og_search_thumbnail(webpage),
+            'subtitles': subtitles,
         }

yt_dlp/extractor/soundcloud.py

@@ -4,7 +4,7 @@ from __future__ import unicode_literals
 import itertools
 import re
 import json
-import random
+# import random
 
 from .common import (
     InfoExtractor,
@@ -164,23 +164,11 @@ class SoundcloudIE(InfoExtractor):
         },
         # downloadable song
         {
-            'url': 'https://soundcloud.com/oddsamples/bus-brakes',
-            'md5': '7624f2351f8a3b2e7cd51522496e7631',
+            'url': 'https://soundcloud.com/the80m/the-following',
+            'md5': '9ffcddb08c87d74fb5808a3c183a1d04',
             'info_dict': {
-                'id': '128590877',
-                'ext': 'mp3',
-                'title': 'Bus Brakes',
-                'description': 'md5:0053ca6396e8d2fd7b7e1595ef12ab66',
-                'uploader': 'oddsamples',
-                'uploader_id': '73680509',
-                'timestamp': 1389232924,
-                'upload_date': '20140109',
-                'duration': 17.346,
-                'license': 'cc-by-sa',
-                'view_count': int,
-                'like_count': int,
-                'comment_count': int,
-                'repost_count': int,
+                'id': '343609555',
+                'ext': 'wav',
             },
         },
         # private link, downloadable format
@@ -317,12 +305,13 @@ class SoundcloudIE(InfoExtractor):
                 raise
 
     def _real_initialize(self):
-        self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or "T5R4kgWS2PRf6lzLyIravUMnKlbIxQag"  # 'EXLwg5lHTO2dslU5EePe3xkw0m1h86Cd' # 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
+        self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or 'fXuVKzsVXlc6tzniWWS31etd7VHWFUuN'  # persistent `client_id`
         self._login()
 
-    _USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36"
+    _USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36'
     _API_AUTH_QUERY_TEMPLATE = '?client_id=%s'
     _API_AUTH_URL_PW = 'https://api-auth.soundcloud.com/web-auth/sign-in/password%s'
+    _API_VERIFY_AUTH_TOKEN = 'https://api-auth.soundcloud.com/connect/session%s'
     _access_token = None
     _HEADERS = {}
     _NETRC_MACHINE = 'soundcloud'
@@ -332,6 +321,23 @@ class SoundcloudIE(InfoExtractor):
         if username is None:
             return
 
+        if username == 'oauth' and password is not None:
+            self._access_token = password
+            query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
+            payload = {'session': {'access_token': self._access_token}}
+            token_verification = sanitized_Request(self._API_VERIFY_AUTH_TOKEN % query, json.dumps(payload).encode('utf-8'))
+            response = self._download_json(token_verification, None, note='Verifying login token...', fatal=False)
+            if response is not False:
+                self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
+                self.report_login()
+            else:
+                self.report_warning('Provided authorization token seems to be invalid. Continue as guest')
+        elif username is not None:
+            self.report_warning(
+                'Login using username and password is not currently supported. '
+                'Use "--user oauth --password <oauth_token>" to login using an oauth token')
+
+        r'''
         def genDevId():
             def genNumBlock():
                 return ''.join([str(random.randrange(10)) for i in range(6)])
@@ -358,6 +364,7 @@ class SoundcloudIE(InfoExtractor):
             self.report_warning('Unable to get access token, login may has failed')
         else:
             self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
+        '''
 
     # signature generation
     def sign(self, user, pw, clid):
@@ -370,9 +377,9 @@ class SoundcloudIE(InfoExtractor):
         b = 37
         k = 37
         c = 5
-        n = "0763ed7314c69015fd4a0dc16bbf4b90"  # _KEY
-        y = "8"  # _REV
-        r = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36"  # _USER_AGENT
+        n = '0763ed7314c69015fd4a0dc16bbf4b90'  # _KEY
+        y = '8'  # _REV
+        r = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36'  # _USER_AGENT
         e = user  # _USERNAME
         t = clid  # _CLIENT_ID
 

yt_dlp/extractor/tbs.py

@@ -16,7 +16,7 @@ from ..utils import (
 
 
 class TBSIE(TurnerBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?(?P<site>tbs|tntdrama)\.com(?P<path>/(?:movies|shows/[^/]+/(?:clips|season-\d+/episode-\d+))/(?P<id>[^/?#]+))'
+    _VALID_URL = r'https?://(?:www\.)?(?P<site>tbs|tntdrama)\.com(?P<path>/(?:movies|watchtnt|shows/[^/]+/(?:clips|season-\d+/episode-\d+))/(?P<id>[^/?#]+))'
     _TESTS = [{
         'url': 'http://www.tntdrama.com/shows/the-alienist/clips/monster',
         'info_dict': {
@@ -45,7 +45,8 @@ class TBSIE(TurnerBaseIE):
         drupal_settings = self._parse_json(self._search_regex(
             r'<script[^>]+?data-drupal-selector="drupal-settings-json"[^>]*?>({.+?})</script>',
             webpage, 'drupal setting'), display_id)
-        video_data = next(v for v in drupal_settings['turner_playlist'] if v.get('url') == path)
+        isLive = 'watchtnt' in path
+        video_data = next(v for v in drupal_settings['turner_playlist'] if isLive or v.get('url') == path)
 
         media_id = video_data['mediaID']
         title = video_data['title']
@@ -56,7 +57,8 @@ class TBSIE(TurnerBaseIE):
             media_id, tokenizer_query, {
                 'url': url,
                 'site_name': site[:3].upper(),
-                'auth_required': video_data.get('authRequired') == '1',
+                'auth_required': video_data.get('authRequired') == '1' or isLive,
+                'is_live': isLive
             })
 
         thumbnails = []
@@ -85,5 +87,6 @@ class TBSIE(TurnerBaseIE):
             'season_number': int_or_none(video_data.get('season')),
             'episode_number': int_or_none(video_data.get('episode')),
             'thumbnails': thumbnails,
+            'is_live': isLive
         })
         return info

yt_dlp/extractor/turner.py

@@ -221,6 +221,7 @@ class TurnerBaseIE(AdobePassIE):
         }
 
     def _extract_ngtv_info(self, media_id, tokenizer_query, ap_data=None):
+        is_live = ap_data.get('is_live')
         streams_data = self._download_json(
             'http://medium.ngtv.io/media/%s/tv' % media_id,
             media_id)['media']['tv']
@@ -237,11 +238,11 @@ class TurnerBaseIE(AdobePassIE):
                     'http://token.ngtv.io/token/token_spe',
                     m3u8_url, media_id, ap_data or {}, tokenizer_query)
             formats.extend(self._extract_m3u8_formats(
-                m3u8_url, media_id, 'mp4', m3u8_id='hls', fatal=False))
+                m3u8_url, media_id, 'mp4', m3u8_id='hls', live=is_live, fatal=False))
 
             duration = float_or_none(stream_data.get('totalRuntime'))
 
-            if not chapters:
+            if not chapters and not is_live:
                 for chapter in stream_data.get('contentSegments', []):
                     start_time = float_or_none(chapter.get('start'))
                     chapter_duration = float_or_none(chapter.get('duration'))

yt_dlp/extractor/twitcasting.py

@@ -5,12 +5,14 @@ import itertools
 import re
 
 from .common import InfoExtractor
+from ..downloader.websocket import has_websockets
 from ..utils import (
     clean_html,
     float_or_none,
     get_element_by_class,
     get_element_by_id,
     parse_duration,
+    qualities,
     str_to_int,
     try_get,
     unified_timestamp,
@@ -89,9 +91,24 @@ class TwitCastingIE(InfoExtractor):
             video_js_data = video_js_data[0]
             m3u8_url = try_get(video_js_data, lambda x: x['source']['url'])
 
+        stream_server_data = self._download_json(
+            'https://twitcasting.tv/streamserver.php?target=%s&mode=client' % uploader_id, video_id,
+            'Downloading live info', fatal=False)
+
         is_live = 'data-status="online"' in webpage
+        formats = []
         if is_live and not m3u8_url:
             m3u8_url = 'https://twitcasting.tv/%s/metastream.m3u8' % uploader_id
+        if is_live and has_websockets and stream_server_data:
+            qq = qualities(['base', 'mobilesource', 'main'])
+            for mode, ws_url in stream_server_data['llfmp4']['streams'].items():
+                formats.append({
+                    'url': ws_url,
+                    'format_id': 'ws-%s' % mode,
+                    'ext': 'mp4',
+                    'quality': qq(mode),
+                    'protocol': 'websocket_frag',  # TwitCasting simply sends moof atom directly over WS
+                })
 
         thumbnail = video_js_data.get('thumbnailUrl') or self._og_search_thumbnail(webpage)
         description = clean_html(get_element_by_id(
@@ -106,10 +123,9 @@ class TwitCastingIE(InfoExtractor):
             r'data-toggle="true"[^>]+datetime="([^"]+)"',
             webpage, 'datetime', None))
 
-        formats = None
         if m3u8_url:
-            formats = self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', live=is_live)
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', live=is_live))
         self._sort_formats(formats)
 
         return {

yt_dlp/extractor/umg.py

@@ -28,7 +28,7 @@ class UMGDeIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
         video_data = self._download_json(
-            'https://api.universal-music.de/graphql',
+            'https://graphql.universal-music.de/',
             video_id, query={
                 'query': '''{
   universalMusic(channel:16) {
@@ -56,11 +56,9 @@ class UMGDeIE(InfoExtractor):
         formats = []
 
         def add_m3u8_format(format_id):
-            m3u8_formats = self._extract_m3u8_formats(
+            formats.extend(self._extract_m3u8_formats(
                 hls_url_template % format_id, video_id, 'mp4',
-                'm3u8_native', m3u8_id='hls', fatal='False')
-            if m3u8_formats and m3u8_formats[0].get('height'):
-                formats.extend(m3u8_formats)
+                'm3u8_native', m3u8_id='hls', fatal=False))
 
         for f in video_data.get('formats', []):
             f_url = f.get('url')

yt_dlp/extractor/videa.py

@@ -12,6 +12,7 @@ from ..utils import (
     mimetype2ext,
     parse_codecs,
     update_url_query,
+    urljoin,
     xpath_element,
     xpath_text,
 )
@@ -19,6 +20,7 @@ from ..compat import (
     compat_b64decode,
     compat_ord,
     compat_struct_pack,
+    compat_urlparse,
 )
 
 
@@ -95,9 +97,13 @@ class VideaIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        query = {'v': video_id}
-        player_page = self._download_webpage(
-            'https://videa.hu/player', video_id, query=query)
+
+        video_page = self._download_webpage(url, video_id)
+
+        player_url = self._search_regex(
+            r'<iframe.*?src="(/player\?[^"]+)"', video_page, 'player url')
+        player_url = urljoin(url, player_url)
+        player_page = self._download_webpage(player_url, video_id)
 
         nonce = self._search_regex(
             r'_xt\s*=\s*"([^"]+)"', player_page, 'nonce')
@@ -107,6 +113,7 @@ class VideaIE(InfoExtractor):
         for i in range(0, 32):
             result += s[i - (self._STATIC_SECRET.index(l[i]) - 31)]
 
+        query = compat_urlparse.parse_qs(compat_urlparse.urlparse(player_url).query)
         random_seed = ''.join(random.choice(string.ascii_letters + string.digits) for _ in range(8))
         query['_s'] = random_seed
         query['_t'] = result[:16]

yt_dlp/extractor/viki.py

@@ -1,39 +1,28 @@
 # coding: utf-8
 from __future__ import unicode_literals
-
-import base64
 import hashlib
 import hmac
-import itertools
 import json
-import re
 import time
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_parse_qs,
-    compat_urllib_parse_urlparse,
-)
 from ..utils import (
     ExtractorError,
     int_or_none,
-    HEADRequest,
     parse_age_limit,
     parse_iso8601,
-    sanitized_Request,
-    std_headers,
     try_get,
 )
 
 
 class VikiBaseIE(InfoExtractor):
     _VALID_URL_BASE = r'https?://(?:www\.)?viki\.(?:com|net|mx|jp|fr)/'
-    _API_QUERY_TEMPLATE = '/v4/%sapp=%s&t=%s&site=www.viki.com'
-    _API_URL_TEMPLATE = 'https://api.viki.io%s&sig=%s'
+    _API_URL_TEMPLATE = 'https://api.viki.io%s'
 
+    _DEVICE_ID = '86085977d'  # used for android api
     _APP = '100005a'
-    _APP_VERSION = '6.0.0'
-    _APP_SECRET = 'MM_d*yP@`&1@]@!AVrXf_o-HVEnoTnm$O-ti4[G~$JDI/Dc-&piU&z&5.;:}95=Iad'
+    _APP_VERSION = '6.11.3'
+    _APP_SECRET = 'd96704b180208dbb2efa30fe44c48bd8690441af9f567ba8fd710a72badc85198f7472'
 
     _GEO_BYPASS = False
     _NETRC_MACHINE = 'viki'
@@ -46,53 +35,57 @@ class VikiBaseIE(InfoExtractor):
         'paywall': 'Sorry, this content is only available to Viki Pass Plus subscribers',
     }
 
-    def _prepare_call(self, path, timestamp=None, post_data=None):
+    def _stream_headers(self, timestamp, sig):
+        return {
+            'X-Viki-manufacturer': 'vivo',
+            'X-Viki-device-model': 'vivo 1606',
+            'X-Viki-device-os-ver': '6.0.1',
+            'X-Viki-connection-type': 'WIFI',
+            'X-Viki-carrier': '',
+            'X-Viki-as-id': '100005a-1625321982-3932',
+            'timestamp': str(timestamp),
+            'signature': str(sig),
+            'x-viki-app-ver': self._APP_VERSION
+        }
+
+    def _api_query(self, path, version=4, **kwargs):
         path += '?' if '?' not in path else '&'
-        if not timestamp:
-            timestamp = int(time.time())
-        query = self._API_QUERY_TEMPLATE % (path, self._APP, timestamp)
+        query = f'/v{version}/{path}app={self._APP}'
         if self._token:
             query += '&token=%s' % self._token
+        return query + ''.join(f'&{name}={val}' for name, val in kwargs.items())
+
+    def _sign_query(self, path):
+        timestamp = int(time.time())
+        query = self._api_query(path, version=5)
         sig = hmac.new(
-            self._APP_SECRET.encode('ascii'),
-            query.encode('ascii'),
-            hashlib.sha1
-        ).hexdigest()
-        url = self._API_URL_TEMPLATE % (query, sig)
-        return sanitized_Request(
-            url, json.dumps(post_data).encode('utf-8')) if post_data else url
+            self._APP_SECRET.encode('ascii'), f'{query}&t={timestamp}'.encode('ascii'), hashlib.sha1).hexdigest()
+        return timestamp, sig, self._API_URL_TEMPLATE % query
 
-    def _call_api(self, path, video_id, note, timestamp=None, post_data=None):
+    def _call_api(
+            self, path, video_id, note='Downloading JSON metadata', data=None, query=None, fatal=True):
+        if query is None:
+            timestamp, sig, url = self._sign_query(path)
+        else:
+            url = self._API_URL_TEMPLATE % self._api_query(path, version=4)
         resp = self._download_json(
-            self._prepare_call(path, timestamp, post_data),
-            video_id, note,
-            headers={
-                'x-client-user-agent': std_headers['User-Agent'],
-                'x-viki-as-id': self._APP,
-                'x-viki-app-ver': self._APP_VERSION,
-            })
-
-        error = resp.get('error')
-        if error:
-            if error == 'invalid timestamp':
-                resp = self._download_json(
-                    self._prepare_call(path, int(resp['current_timestamp']), post_data),
-                    video_id, '%s (retry)' % note,
-                    headers={
-                        'x-client-user-agent': std_headers['User-Agent'],
-                        'x-viki-as-id': self._APP,
-                        'x-viki-app-ver': self._APP_VERSION,
-                    })
-                error = resp.get('error')
-            if error:
-                self._raise_error(resp['error'])
+            url, video_id, note, fatal=fatal, query=query,
+            data=json.dumps(data).encode('utf-8') if data else None,
+            headers=({'x-viki-app-ver': self._APP_VERSION} if data
+                     else self._stream_headers(timestamp, sig) if query is None
+                     else None)) or {}
 
+        self._raise_error(resp.get('error'), fatal)
         return resp
 
-    def _raise_error(self, error):
-        raise ExtractorError(
-            '%s returned error: %s' % (self.IE_NAME, error),
-            expected=True)
+    def _raise_error(self, error, fatal=True):
+        if error is None:
+            return
+        msg = '%s said: %s' % (self.IE_NAME, error)
+        if fatal:
+            raise ExtractorError(msg, expected=True)
+        else:
+            self.report_warning(msg)
 
     def _check_errors(self, data):
         for reason, status in (data.get('blocking') or {}).items():
@@ -101,9 +94,10 @@ class VikiBaseIE(InfoExtractor):
                 if reason == 'geo':
                     self.raise_geo_restricted(msg=message)
                 elif reason == 'paywall':
+                    if try_get(data, lambda x: x['paywallable']['tvod']):
+                        self._raise_error('This video is for rent only or TVOD (Transactional Video On demand)')
                     self.raise_login_required(message)
-                raise ExtractorError('%s said: %s' % (
-                    self.IE_NAME, message), expected=True)
+                self._raise_error(message)
 
     def _real_initialize(self):
         self._login()
@@ -113,35 +107,24 @@ class VikiBaseIE(InfoExtractor):
         if username is None:
             return
 
-        login_form = {
-            'login_id': username,
-            'password': password,
-        }
-
-        login = self._call_api(
-            'sessions.json', None,
-            'Logging in', post_data=login_form)
-
-        self._token = login.get('token')
+        self._token = self._call_api(
+            'sessions.json', None, 'Logging in', fatal=False,
+            data={'username': username, 'password': password}).get('token')
         if not self._token:
-            self.report_warning('Unable to get session token, login has probably failed')
+            self.report_warning('Login Failed: Unable to get session token')
 
     @staticmethod
-    def dict_selection(dict_obj, preferred_key, allow_fallback=True):
+    def dict_selection(dict_obj, preferred_key):
         if preferred_key in dict_obj:
-            return dict_obj.get(preferred_key)
-
-        if not allow_fallback:
-            return
-
-        filtered_dict = list(filter(None, [dict_obj.get(k) for k in dict_obj.keys()]))
-        return filtered_dict[0] if filtered_dict else None
+            return dict_obj[preferred_key]
+        return (list(filter(None, dict_obj.values())) or [None])[0]
 
 
 class VikiIE(VikiBaseIE):
     IE_NAME = 'viki'
     _VALID_URL = r'%s(?:videos|player)/(?P<id>[0-9]+v)' % VikiBaseIE._VALID_URL_BASE
     _TESTS = [{
+        'note': 'Free non-DRM video with storyboards in MPD',
         'url': 'https://www.viki.com/videos/1175236v-choosing-spouse-by-lottery-episode-1',
         'info_dict': {
             'id': '1175236v',
@@ -155,7 +138,6 @@ class VikiIE(VikiBaseIE):
         'params': {
             'format': 'bestvideo',
         },
-        'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
     }, {
         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
         'info_dict': {
@@ -173,7 +155,6 @@ class VikiIE(VikiBaseIE):
             'format': 'bestvideo',
         },
         'skip': 'Blocked in the US',
-        'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
     }, {
         # clip
         'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
@@ -225,7 +206,6 @@ class VikiIE(VikiBaseIE):
         'params': {
             'format': 'bestvideo',
         },
-        'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
     }, {
         # youtube external
         'url': 'http://www.viki.com/videos/50562v-poor-nastya-complete-episode-1',
@@ -264,23 +244,14 @@ class VikiIE(VikiBaseIE):
         'params': {
             'format': 'bestvideo',
         },
-        'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'],
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-
-        resp = self._download_json(
-            'https://www.viki.com/api/videos/' + video_id,
-            video_id, 'Downloading video JSON', headers={
-                'x-client-user-agent': std_headers['User-Agent'],
-                'x-viki-app-ver': '3.0.0',
-            })
-        video = resp['video']
-
+        video = self._call_api(f'videos/{video_id}.json', video_id, 'Downloading video JSON', query={})
         self._check_errors(video)
 
-        title = self.dict_selection(video.get('titles', {}), 'en', allow_fallback=False)
+        title = try_get(video, lambda x: x['titles']['en'], str)
         episode_number = int_or_none(video.get('number'))
         if not title:
             title = 'Episode %d' % episode_number if video.get('type') == 'episode' else video.get('id') or video_id
@@ -288,116 +259,46 @@ class VikiIE(VikiBaseIE):
             container_title = self.dict_selection(container_titles, 'en')
             title = '%s - %s' % (container_title, title)
 
-        description = self.dict_selection(video.get('descriptions', {}), 'en')
+        thumbnails = [{
+            'id': thumbnail_id,
+            'url': thumbnail['url'],
+        } for thumbnail_id, thumbnail in (video.get('images') or {}).items() if thumbnail.get('url')]
 
-        like_count = int_or_none(try_get(video, lambda x: x['likes']['count']))
+        resp = self._call_api(
+            'playback_streams/%s.json?drms=dt1,dt2&device_id=%s' % (video_id, self._DEVICE_ID),
+            video_id, 'Downloading video streams JSON')['main'][0]
 
-        thumbnails = []
-        for thumbnail_id, thumbnail in (video.get('images') or {}).items():
-            thumbnails.append({
-                'id': thumbnail_id,
-                'url': thumbnail.get('url'),
-            })
+        stream_id = try_get(resp, lambda x: x['properties']['track']['stream_id'])
+        subtitles = dict((lang, [{
+            'ext': ext,
+            'url': self._API_URL_TEMPLATE % self._api_query(
+                f'videos/{video_id}/auth_subtitles/{lang}.{ext}', stream_id=stream_id)
+        } for ext in ('srt', 'vtt')]) for lang in (video.get('subtitle_completions') or {}).keys())
 
-        subtitles = {}
-        for subtitle_lang, _ in (video.get('subtitle_completions') or {}).items():
-            subtitles[subtitle_lang] = [{
-                'ext': subtitles_format,
-                'url': self._prepare_call(
-                    'videos/%s/subtitles/%s.%s' % (video_id, subtitle_lang, subtitles_format)),
-            } for subtitles_format in ('srt', 'vtt')]
+        mpd_url = resp['url']
+        # 1080p is hidden in another mpd which can be found in the current manifest content
+        mpd_content = self._download_webpage(mpd_url, video_id, note='Downloading initial MPD manifest')
+        mpd_url = self._search_regex(
+            r'(?mi)<BaseURL>(http.+.mpd)', mpd_content, 'new manifest', default=mpd_url)
+        formats = self._extract_mpd_formats(mpd_url, video_id)
+        self._sort_formats(formats)
 
-        result = {
+        return {
             'id': video_id,
+            'formats': formats,
             'title': title,
-            'description': description,
+            'description': self.dict_selection(video.get('descriptions', {}), 'en'),
             'duration': int_or_none(video.get('duration')),
             'timestamp': parse_iso8601(video.get('created_at')),
             'uploader': video.get('author'),
             'uploader_url': video.get('author_url'),
-            'like_count': like_count,
+            'like_count': int_or_none(try_get(video, lambda x: x['likes']['count'])),
             'age_limit': parse_age_limit(video.get('rating')),
             'thumbnails': thumbnails,
             'subtitles': subtitles,
             'episode_number': episode_number,
         }
 
-        formats = []
-
-        def add_format(format_id, format_dict, protocol='http'):
-            # rtmps URLs does not seem to work
-            if protocol == 'rtmps':
-                return
-            format_url = format_dict.get('url')
-            if not format_url:
-                return
-            qs = compat_parse_qs(compat_urllib_parse_urlparse(format_url).query)
-            stream = qs.get('stream', [None])[0]
-            if stream:
-                format_url = base64.b64decode(stream).decode()
-            if format_id in ('m3u8', 'hls'):
-                m3u8_formats = self._extract_m3u8_formats(
-                    format_url, video_id, 'mp4',
-                    entry_protocol='m3u8_native',
-                    m3u8_id='m3u8-%s' % protocol, fatal=False)
-                # Despite CODECS metadata in m3u8 all video-only formats
-                # are actually video+audio
-                for f in m3u8_formats:
-                    if not self.get_param('allow_unplayable_formats') and '_drm/index_' in f['url']:
-                        continue
-                    if f.get('acodec') == 'none' and f.get('vcodec') != 'none':
-                        f['acodec'] = None
-                    formats.append(f)
-            elif format_id in ('mpd', 'dash'):
-                formats.extend(self._extract_mpd_formats(
-                    format_url, video_id, 'mpd-%s' % protocol, fatal=False))
-            elif format_url.startswith('rtmp'):
-                mobj = re.search(
-                    r'^(?P<url>rtmp://[^/]+/(?P<app>.+?))/(?P<playpath>mp4:.+)$',
-                    format_url)
-                if not mobj:
-                    return
-                formats.append({
-                    'format_id': 'rtmp-%s' % format_id,
-                    'ext': 'flv',
-                    'url': mobj.group('url'),
-                    'play_path': mobj.group('playpath'),
-                    'app': mobj.group('app'),
-                    'page_url': url,
-                })
-            else:
-                urlh = self._request_webpage(
-                    HEADRequest(format_url), video_id, 'Checking file size', fatal=False)
-                formats.append({
-                    'url': format_url,
-                    'format_id': '%s-%s' % (format_id, protocol),
-                    'height': int_or_none(self._search_regex(
-                        r'^(\d+)[pP]$', format_id, 'height', default=None)),
-                    'filesize': int_or_none(urlh.headers.get('Content-Length')),
-                })
-
-        for format_id, format_dict in (resp.get('streams') or {}).items():
-            add_format(format_id, format_dict)
-        if not formats:
-            streams = self._call_api(
-                'videos/%s/streams.json' % video_id, video_id,
-                'Downloading video streams JSON')
-
-            if 'external' in streams:
-                result.update({
-                    '_type': 'url_transparent',
-                    'url': streams['external']['url'],
-                })
-                return result
-
-            for format_id, stream_dict in streams.items():
-                for protocol, format_dict in stream_dict.items():
-                    add_format(format_id, format_dict, protocol)
-        self._sort_formats(formats)
-
-        result['formats'] = formats
-        return result
-
 
 class VikiChannelIE(VikiBaseIE):
     IE_NAME = 'viki:channel'
@@ -409,7 +310,7 @@ class VikiChannelIE(VikiBaseIE):
             'title': 'Boys Over Flowers',
             'description': 'md5:804ce6e7837e1fd527ad2f25420f4d59',
         },
-        'playlist_mincount': 71,
+        'playlist_mincount': 51,
     }, {
         'url': 'http://www.viki.com/tv/1354c-poor-nastya-complete',
         'info_dict': {
@@ -430,33 +331,35 @@ class VikiChannelIE(VikiBaseIE):
         'only_matching': True,
     }]
 
-    _PER_PAGE = 25
+    _video_types = ('episodes', 'movies', 'clips', 'trailers')
+
+    def _entries(self, channel_id):
+        params = {
+            'app': self._APP, 'token': self._token, 'only_ids': 'true',
+            'direction': 'asc', 'sort': 'number', 'per_page': 30
+        }
+        video_types = self._configuration_arg('video_types') or self._video_types
+        for video_type in video_types:
+            if video_type not in self._video_types:
+                self.report_warning(f'Unknown video_type: {video_type}')
+            page_num = 0
+            while True:
+                page_num += 1
+                params['page'] = page_num
+                res = self._call_api(
+                    f'containers/{channel_id}/{video_type}.json', channel_id, query=params, fatal=False,
+                    note='Downloading %s JSON page %d' % (video_type.title(), page_num))
+
+                for video_id in res.get('response') or []:
+                    yield self.url_result(f'https://www.viki.com/videos/{video_id}', VikiIE.ie_key(), video_id)
+                if not res.get('more'):
+                    break
 
     def _real_extract(self, url):
         channel_id = self._match_id(url)
-
-        channel = self._call_api(
-            'containers/%s.json' % channel_id, channel_id,
-            'Downloading channel JSON')
-
+        channel = self._call_api('containers/%s.json' % channel_id, channel_id, 'Downloading channel JSON')
         self._check_errors(channel)
-
-        title = self.dict_selection(channel['titles'], 'en')
-
-        description = self.dict_selection(channel['descriptions'], 'en')
-
-        entries = []
-        for video_type in ('episodes', 'clips', 'movies'):
-            for page_num in itertools.count(1):
-                page = self._call_api(
-                    'containers/%s/%s.json?per_page=%d&sort=number&direction=asc&with_paging=true&page=%d'
-                    % (channel_id, video_type, self._PER_PAGE, page_num), channel_id,
-                    'Downloading %s JSON page #%d' % (video_type, page_num))
-                for video in page['response']:
-                    video_id = video['id']
-                    entries.append(self.url_result(
-                        'https://www.viki.com/videos/%s' % video_id, 'Viki'))
-                if not page['pagination']['next']:
-                    break
-
-        return self.playlist_result(entries, channel_id, title, description)
+        return self.playlist_result(
+            self._entries(channel_id), channel_id,
+            self.dict_selection(channel['titles'], 'en'),
+            self.dict_selection(channel['descriptions'], 'en'))

yt_dlp/extractor/vlive.py

@@ -178,9 +178,15 @@ class VLiveIE(VLiveBaseIE):
         if video_type == 'VOD':
             inkey = self._call_api('video/v1.0/vod/%s/inkey', video_id)['inkey']
             vod_id = video['vodId']
-            return merge_dicts(
+            info_dict = merge_dicts(
                 get_common_fields(),
                 self._extract_video_info(video_id, vod_id, inkey))
+            thumbnail = video.get('thumb')
+            if thumbnail:
+                if not info_dict.get('thumbnails') and info_dict.get('thumbnail'):
+                    info_dict['thumbnails'] = [{'url': info_dict.pop('thumbnail')}]
+                info_dict.setdefault('thumbnails', []).append({'url': thumbnail, 'preference': 1})
+            return info_dict
         elif video_type == 'LIVE':
             status = video.get('status')
             if status == 'ON_AIR':

yt_dlp/extractor/yahoo.py

@@ -22,6 +22,7 @@ from ..utils import (
 )
 
 from .brightcove import BrightcoveNewIE
+from .youtube import YoutubeIE
 
 
 class YahooIE(InfoExtractor):
@@ -38,6 +39,7 @@ class YahooIE(InfoExtractor):
             'timestamp': 1369812016,
             'upload_date': '20130529',
         },
+        'skip': 'No longer exists',
     }, {
         'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
         'md5': '7993e572fac98e044588d0b5260f4352',
@@ -50,6 +52,7 @@ class YahooIE(InfoExtractor):
             'timestamp': 1406838636,
             'upload_date': '20140731',
         },
+        'skip': 'Unfortunately, this video is not available in your region',
     }, {
         'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
         'md5': '71298482f7c64cbb7fa064e4553ff1c1',
@@ -61,7 +64,8 @@ class YahooIE(InfoExtractor):
             'duration': 97,
             'timestamp': 1414489862,
             'upload_date': '20141028',
-        }
+        },
+        'skip': 'No longer exists',
     }, {
         'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
         'md5': '88e209b417f173d86186bef6e4d1f160',
@@ -120,6 +124,7 @@ class YahooIE(InfoExtractor):
             'season_number': 6,
             'episode_number': 1,
         },
+        'skip': 'No longer exists',
     }, {
         # ytwnews://cavideo/
         'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html',
@@ -156,7 +161,7 @@ class YahooIE(InfoExtractor):
                 'id': '352CFDOQrKg',
                 'ext': 'mp4',
                 'title': 'Kyndal Inskeep "Performs the Hell Out of" Sia\'s "Elastic Heart" - The Voice Knockouts 2019',
-                'description': 'md5:35b61e94c2ae214bc965ff4245f80d11',
+                'description': 'md5:7fe8e3d5806f96002e55f190d1d94479',
                 'uploader': 'The Voice',
                 'uploader_id': 'NBCTheVoice',
                 'upload_date': '20191029',
@@ -165,7 +170,7 @@ class YahooIE(InfoExtractor):
         'params': {
             'playlistend': 2,
         },
-        'expected_warnings': ['HTTP Error 404'],
+        'expected_warnings': ['HTTP Error 404', 'Ignoring subtitle tracks'],
     }, {
         'url': 'https://malaysia.news.yahoo.com/video/bystanders-help-ontario-policeman-bust-190932818.html',
         'only_matching': True,
@@ -280,12 +285,13 @@ class YahooIE(InfoExtractor):
         else:
             country = country.split('-')[0]
 
-        item = self._download_json(
+        items = self._download_json(
             'https://%s.yahoo.com/caas/content/article' % country, display_id,
             'Downloading content JSON metadata', query={
                 'url': url
-            })['items'][0]['data']['partnerData']
+            })['items'][0]
 
+        item = items['data']['partnerData']
         if item.get('type') != 'video':
             entries = []
 
@@ -299,9 +305,19 @@ class YahooIE(InfoExtractor):
             for e in (item.get('body') or []):
                 if e.get('type') == 'videoIframe':
                     iframe_url = e.get('url')
-                    if not iframe_url:
-                        continue
+                    if iframe_url:
+                        entries.append(self.url_result(iframe_url))
+
+            if item.get('type') == 'storywithleadvideo':
+                iframe_url = try_get(item, lambda x: x['meta']['player']['url'])
+                if iframe_url:
                     entries.append(self.url_result(iframe_url))
+                else:
+                    self.report_warning("Yahoo didn't provide an iframe url for this storywithleadvideo")
+
+            if items.get('markup'):
+                entries.extend(
+                    self.url_result(yt_url) for yt_url in YoutubeIE._extract_urls(items['markup']))
 
             return self.playlist_result(
                 entries, item.get('uuid'),
@@ -346,7 +362,7 @@ class YahooSearchIE(SearchInfoExtractor):
 
 class YahooGyaOPlayerIE(InfoExtractor):
     IE_NAME = 'yahoo:gyao:player'
-    _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:player|episode/[^/]+)|streaming\.yahoo\.co\.jp/c/y)/(?P<id>\d+/v\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
+    _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:player|episode(?:/[^/]+)?)|streaming\.yahoo\.co\.jp/c/y)/(?P<id>\d+/v\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
     _TESTS = [{
         'url': 'https://gyao.yahoo.co.jp/player/00998/v00818/v0000000000000008564/',
         'info_dict': {
@@ -368,6 +384,9 @@ class YahooGyaOPlayerIE(InfoExtractor):
     }, {
         'url': 'https://gyao.yahoo.co.jp/episode/%E3%81%8D%E3%81%AE%E3%81%86%E4%BD%95%E9%A3%9F%E3%81%B9%E3%81%9F%EF%BC%9F%20%E7%AC%AC2%E8%A9%B1%202019%2F4%2F12%E6%94%BE%E9%80%81%E5%88%86/5cb02352-b725-409e-9f8d-88f947a9f682',
         'only_matching': True,
+    }, {
+        'url': 'https://gyao.yahoo.co.jp/episode/5fa1226c-ef8d-4e93-af7a-fd92f4e30597',
+        'only_matching': True,
     }]
     _GEO_BYPASS = False
 

yt_dlp/options.py

@@ -19,6 +19,7 @@ from .utils import (
     preferredencoding,
     write_string,
 )
+from .cookies import SUPPORTED_BROWSERS
 from .version import __version__
 
 from .downloader.external import list_external_downloaders
@@ -137,14 +138,21 @@ def parseOpts(overrideArguments=None):
         else:
             raise optparse.OptionValueError(
                 'wrong %s formatting; it should be %s, not "%s"' % (opt_str, option.metavar, value))
-        val = process(val) if callable(process) else val
+        try:
+            val = process(val) if process else val
+        except Exception as err:
+            raise optparse.OptionValueError(
+                'wrong %s formatting; %s' % (opt_str, err))
         for key in keys:
             out_dict[key] = val
 
     # No need to wrap help messages if we're on a wide console
     columns = compat_get_terminal_size().columns
     max_width = columns if columns else 80
-    max_help_position = 80
+    # 47% is chosen because that is how README.md is currently formatted
+    # and moving help text even further to the right is undesirable.
+    # This can be reduced in the future to get a prettier output
+    max_help_position = int(0.47 * max_width)
 
     fmt = optparse.IndentedHelpFormatter(width=max_width, max_help_position=max_help_position)
     fmt.format_option_strings = _format_option_string
@@ -520,8 +528,12 @@ def parseOpts(overrideArguments=None):
         help="Don't give any special preference to free containers (default)")
     video_format.add_option(
         '--check-formats',
-        action='store_true', dest='check_formats', default=False,
-        help="Check that the formats selected are actually downloadable (Experimental)")
+        action='store_true', dest='check_formats', default=None,
+        help='Check that the formats selected are actually downloadable')
+    video_format.add_option(
+        '--no-check-formats',
+        action='store_false', dest='check_formats',
+        help='Do not check that the formats selected are actually downloadable')
     video_format.add_option(
         '-F', '--list-formats',
         action='store_true', dest='listformats',
@@ -599,6 +611,10 @@ def parseOpts(overrideArguments=None):
         '-r', '--limit-rate', '--rate-limit',
         dest='ratelimit', metavar='RATE',
         help='Maximum download rate in bytes per second (e.g. 50K or 4.2M)')
+    downloader.add_option(
+        '--throttled-rate',
+        dest='throttledratelimit', metavar='RATE',
+        help='Minimum download rate in bytes per second below which throttling is assumed and the video data is re-extracted (e.g. 100K)')
     downloader.add_option(
         '-R', '--retries',
         dest='retries', metavar='RETRIES', default=10,
@@ -712,7 +728,8 @@ def parseOpts(overrideArguments=None):
         help=(
             'Give these arguments to the external downloader. '
             'Specify the downloader name and the arguments separated by a colon ":". '
-            'You can use this option multiple times (Alias: --external-downloader-args)'))
+            'You can use this option multiple times to give different arguments to different downloaders '
+            '(Alias: --external-downloader-args)'))
 
     workarounds = optparse.OptionGroup(parser, 'Workarounds')
     workarounds.add_option(
@@ -1074,7 +1091,21 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option(
         '--no-cookies',
         action='store_const', const=None, dest='cookiefile', metavar='FILE',
-        help='Do not read/dump cookies (default)')
+        help='Do not read/dump cookies from/to file (default)')
+    filesystem.add_option(
+        '--cookies-from-browser',
+        dest='cookiesfrombrowser', metavar='BROWSER[:PROFILE]',
+        help=(
+            'Load cookies from a user profile of the given web browser. '
+            'Currently supported browsers are: {}. '
+            'You can specify the user profile name or directory using '
+            '"BROWSER:PROFILE_NAME" or "BROWSER:PROFILE_PATH". '
+            'If no profile is given, the most recently accessed one is used'.format(
+                '|'.join(sorted(SUPPORTED_BROWSERS)))))
+    filesystem.add_option(
+        '--no-cookies-from-browser',
+        action='store_const', const=None, dest='cookiesfrombrowser',
+        help='Do not load cookies from browser (default)')
     filesystem.add_option(
         '--cache-dir', dest='cachedir', default=None, metavar='DIR',
         help='Location in the filesystem where youtube-dl can store some downloaded information (such as client ids and signatures) permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl')
@@ -1165,7 +1196,7 @@ def parseOpts(overrideArguments=None):
             'to give the argument to the specified postprocessor/executable. Supported PP are: '
             'Merger, ExtractAudio, SplitChapters, Metadata, EmbedSubtitle, EmbedThumbnail, '
             'SubtitlesConvertor, ThumbnailsConvertor, VideoRemuxer, VideoConvertor, '
-            'SponSkrub, FixupStretched, FixupM4a and FixupM3u8. '
+            'SponSkrub, FixupStretched, FixupM4a, FixupM3u8, FixupTimestamp and FixupDuration. '
             'The supported executables are: AtomicParsley, FFmpeg, FFprobe, and SponSkrub. '
             '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, '
@@ -1200,19 +1231,19 @@ def parseOpts(overrideArguments=None):
     postproc.add_option(
         '--embed-thumbnail',
         action='store_true', dest='embedthumbnail', default=False,
-        help='Embed thumbnail in the audio as cover art')
+        help='Embed thumbnail in the video as cover art')
     postproc.add_option(
         '--no-embed-thumbnail',
         action='store_false', dest='embedthumbnail',
         help='Do not embed thumbnail (default)')
     postproc.add_option(
-        '--add-metadata',
+        '--embed-metadata', '--add-metadata',
         action='store_true', dest='addmetadata', default=False,
-        help='Write metadata to the video file')
+        help='Embed metadata including chapter markers (if supported by the format) to the video file (Alias: --add-metadata)')
     postproc.add_option(
-        '--no-add-metadata',
+        '--no-embed-metadata', '--no-add-metadata',
         action='store_false', dest='addmetadata',
-        help='Do not write metadata (default)')
+        help='Do not write metadata (default)  (Alias: --no-add-metadata)')
     postproc.add_option(
         '--metadata-from-title',
         metavar='FORMAT', dest='metafromtitle',
@@ -1230,10 +1261,12 @@ def parseOpts(overrideArguments=None):
     postproc.add_option(
         '--fixup',
         metavar='POLICY', dest='fixup', default=None,
+        choices=('never', 'ignore', 'warn', 'detect_or_warn', 'force'),
         help=(
             'Automatically correct known faults of the file. '
             'One of never (do nothing), warn (only emit a warning), '
-            'detect_or_warn (the default; fix file if we can, warn otherwise)'))
+            'detect_or_warn (the default; fix file if we can, warn otherwise), '
+            'force (try fixing even if file already exists'))
     postproc.add_option(
         '--prefer-avconv', '--no-prefer-ffmpeg',
         action='store_false', dest='prefer_ffmpeg',
@@ -1254,6 +1287,10 @@ def parseOpts(overrideArguments=None):
             'Similar syntax to the output template can be used to pass any field as arguments to the command. '
             'An additional field "filepath" that contains the final path of the downloaded file is also available. '
             'If no fields are passed, "%(filepath)s" is appended to the end of the command'))
+    postproc.add_option(
+        '--exec-before-download',
+        metavar='CMD', dest='exec_before_dl_cmd',
+        help='Execute a command before the actual download. The syntax is the same as --exec')
     postproc.add_option(
         '--convert-subs', '--convert-sub', '--convert-subtitles',
         metavar='FORMAT', dest='convertsubtitles', default=None,
@@ -1337,22 +1374,35 @@ def parseOpts(overrideArguments=None):
         '--no-hls-split-discontinuity',
         dest='hls_split_discontinuity', action='store_false',
         help='Do not split HLS playlists to different formats at discontinuities such as ad breaks (default)')
+    _extractor_arg_parser = lambda key, vals='': (key.strip().lower(), [val.strip() for val in vals.split(',')])
+    extractor.add_option(
+        '--extractor-args',
+        metavar='KEY:ARGS', dest='extractor_args', default={}, type='str',
+        action='callback', callback=_dict_from_options_callback,
+        callback_kwargs={
+            'multiple_keys': False,
+            'process': lambda val: dict(
+                _extractor_arg_parser(*arg.split('=', 1)) for arg in val.split(';'))
+        },
+        help=(
+            'Pass these arguments to the extractor. See "EXTRACTOR ARGUMENTS" for details. '
+            'You can use this option multiple times to give arguments for different extractors'))
     extractor.add_option(
         '--youtube-include-dash-manifest', '--no-youtube-skip-dash-manifest',
         action='store_true', dest='youtube_include_dash_manifest', default=True,
-        help='Download the DASH manifests and related data on YouTube videos (default) (Alias: --no-youtube-skip-dash-manifest)')
+        help=optparse.SUPPRESS_HELP)
     extractor.add_option(
         '--youtube-skip-dash-manifest', '--no-youtube-include-dash-manifest',
         action='store_false', dest='youtube_include_dash_manifest',
-        help='Do not download the DASH manifests and related data on YouTube videos (Alias: --no-youtube-include-dash-manifest)')
+        help=optparse.SUPPRESS_HELP)
     extractor.add_option(
         '--youtube-include-hls-manifest', '--no-youtube-skip-hls-manifest',
         action='store_true', dest='youtube_include_hls_manifest', default=True,
-        help='Download the HLS manifests and related data on YouTube videos (default) (Alias: --no-youtube-skip-hls-manifest)')
+        help=optparse.SUPPRESS_HELP)
     extractor.add_option(
         '--youtube-skip-hls-manifest', '--no-youtube-include-hls-manifest',
         action='store_false', dest='youtube_include_hls_manifest',
-        help='Do not download the HLS manifests and related data on YouTube videos (Alias: --no-youtube-include-hls-manifest)')
+        help=optparse.SUPPRESS_HELP)
 
     parser.add_option_group(general)
     parser.add_option_group(network)