Release 2021.03.03

Known issues (these issues existed in previous API as well)
* Mix playlists only give 1 page (25 vids)
* Trending only gives 1 video
* History gives 5 pages (200 vids)

Co-authored-by: colethedj, pukkandan

.gitattributes

@@ -0,0 +1 @@
+Makefile* text whitespace=-tab-in-indent

.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 2021.02.19. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. 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 Issues with outdated version will be REJECTED.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 **2021.02.19**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
 - [ ] I've checked that all provided URLs are alive and playable in a browser
 - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped
 - [ ] I've searched the bugtracker for similar issues including closed ones
@@ -44,7 +44,7 @@ Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your com
  [debug] User config: []
  [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
  [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
- [debug] yt-dlp version 2021.02.19
+ [debug] yt-dlp version 2021.03.01
  [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 2021.02.19. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. 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 **2021.02.19**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
 - [ ] 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 2021.02.19. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. 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 **2021.02.19**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
 - [ ] 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 2021.02.19. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. 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 **2021.02.19**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
 - [ ] I've checked that all provided URLs are alive and playable in a browser
 - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped
 - [ ] I've searched the bugtracker for similar bug reports including closed ones
@@ -46,7 +46,7 @@ Add the `-v` flag to your command line you run yt-dlp with (`yt-dlp -v <your com
  [debug] User config: []
  [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
  [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
- [debug] yt-dlp version 2021.02.19
+ [debug] yt-dlp version 2021.03.01
  [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 2021.02.19. If it's not, see https://github.com/yt-dlp/yt-dlp on how to update. Issues with outdated version will be REJECTED.
+- First of, make sure you are using the latest version of yt-dlp. Run `yt-dlp --version` and ensure your version is 2021.03.01. 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 **2021.02.19**
+- [ ] I've verified that I'm running yt-dlp version **2021.03.01**
 - [ ] I've searched the bugtracker for similar feature requests including closed ones
 
 

.github/workflows/build.yml

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

.github/workflows/core.yml

@@ -3,7 +3,7 @@ on: [push, pull_request]
 jobs:
   tests:
     name: Core Tests
-    if: "!contains(github.event.head_commit.message, 'ci skip all')"
+    if: "!contains(github.event.head_commit.message, 'ci skip')"
     runs-on: ${{ matrix.os }}
     strategy:
       fail-fast: true

.github/workflows/download.yml

@@ -3,7 +3,7 @@ on: [push, pull_request]
 jobs:
   tests:
     name: Download Tests
-    if: "!contains(github.event.head_commit.message, 'ci skip dl') && !contains(github.event.head_commit.message, 'ci skip all')"
+    if: "contains(github.event.head_commit.message, 'ci run dl')"
     runs-on: ${{ matrix.os }}
     strategy:
       fail-fast: true

.gitignore

@@ -8,6 +8,7 @@ dist/
 zip/
 tmp/
 venv/
+completions/
 
 # Misc
 *~
@@ -24,7 +25,9 @@ updates_key.pem
 *.class
 
 # Generated
+AUTHORS
 README.txt
+.mailmap
 *.1
 *.bash-completion
 *.fish
@@ -34,8 +37,9 @@ README.txt
 *.spec
 
 # Binary
-youtube-dl
+/youtube-dl
-youtube-dlc
+/youtube-dlc
+/yt-dlp
 yt-dlp.zip
 *.exe
 
@@ -50,12 +54,14 @@ yt-dlp.zip
 *.m4v
 *.mp3
 *.3gp
+*.webm
 *.wav
 *.ape
 *.mkv
 *.swf
 *.part
 *.ytdl
+*.dump
 *.frag
 *.frag.urls
 *.aria2

.readthedocs.yml

@@ -20,4 +20,3 @@ python:
     version: 3
     install:
     - requirements: docs/requirements.txt
-    - requirements: requirements.txt

.travis.yml.disabled

@@ -1,38 +0,0 @@
-language: python
-python:
-  - "2.6"
-  - "2.7"
-  - "3.2"
-  - "3.3"
-  - "3.4"
-  - "3.5"
-  - "3.6"
-  - "pypy"
-  - "pypy3"
-dist: trusty
-env:
-  - YTDL_TEST_SET=core
-jobs:
-  include:
-    - python: 3.7
-      dist: xenial
-      env: YTDL_TEST_SET=core
-    - python: 3.8
-      dist: xenial
-      env: YTDL_TEST_SET=core
-    - python: 3.8-dev
-      dist: xenial
-      env: YTDL_TEST_SET=core
-    - env: JYTHON=true; YTDL_TEST_SET=core
-    - name: flake8
-      python: 3.8
-      dist: xenial
-      install: pip install flake8
-      script: flake8 .
-  fast_finish: true
-  allow_failures:
-    - env: YTDL_TEST_SET=download
-    - env: JYTHON=true; YTDL_TEST_SET=core
-before_install:
-  - if [ "$JYTHON" == "true" ]; then ./devscripts/install_jython.sh; export PATH="$HOME/jython/bin:$PATH"; fi
-script: ./devscripts/run_tests.sh

CONTRIBUTORS

@@ -23,3 +23,8 @@ tsukumi
 bbepis
 Pccode66
 Ashish
+RobinD42
+hseg
+colethedj
+DennyDai
+codeasashu

Changelog.md

@@ -17,9 +17,44 @@
 -->
 
 
+### 2021.03.03
+* [youtube] Use new browse API for continuation page extraction by @colethedj and @pukkandan
+* Fix HLS playlist downloading by @shirt
+* **Merge youtube-dl:** Upto [2021.03.03](https://github.com/ytdl-org/youtube-dl/releases/tag/2021.03.03)
+* [mtv] Fix extractor
+* [nick] Fix extractor by @DennyDai
+* [mxplayer] Add new extractor by@codeasashu
+* [youtube] Throw error when `--extractor-retries` are exhausted
+* Reduce default of `--extractor-retries` to 3
+* Fix packaging bugs by @hseg
+
+
+### 2021.03.01
+* Allow specifying path in `--external-downloader`
+* Add option `--sleep-requests` to sleep b/w requests
+* Add option `--extractor-retries` to retry on known extractor errors
+* Extract comments only when needed
+* `--get-comments` doesn't imply `--write-info-json` if `-J`, `-j` or `--print-json` are used
+* Fix `get_executable_path` by [shirt](https://github.com/shirt-dev)
+* [youtube] Retry on more known errors than just HTTP-5xx
+* [youtube] Fix inconsistent `webpage_url`
+* [tennistv] Fix format sorting
+* [bilibiliaudio] Recognize the file as audio-only
+* [hrfensehen] Fix wrong import
+* [viki] Fix viki play pass authentication by [RobinD42](https://github.com/RobinD42)
+* [readthedocs] Improvements by [shirt](https://github.com/shirt-dev)
+* [hls] Fix bug with m3u8 format extraction
+* [hls] Enable `--hls-use-mpegts` by default when downloading live-streams
+* [embedthumbnail] Fix bug with deleting original thumbnail
+* [build] Fix completion paths, zsh pip completion install by [hseg](https://github.com/hseg)
+* [ci] Disable download tests unless specifically invoked
+* Cleanup some code and fix typos
+
+
 ### 2021.02.24
 * Moved project to an organization [yt-dlp](https://github.com/yt-dlp)
 * **Completely changed project name to yt-dlp** by [Pccode66](https://github.com/Pccode66) and [pukkandan](https://github.com/pukkandan)
+    * Also, `youtube-dlc` config files are no longer loaded
 * **Merge youtube-dl:** Upto [commit/4460329](https://github.com/ytdl-org/youtube-dl/commit/44603290e5002153f3ebad6230cc73aef42cc2cd) (except tmz, gedi)
 * [Readthedocs](https://yt-dlp.readthedocs.io) support by [shirt](https://github.com/shirt-dev)
 * [youtube] Show if video was a live stream in info (`was_live`)
@@ -28,12 +63,12 @@
 * [tennistv] Fix extractor
 * [hls] Support media initialization by [shirt](https://github.com/shirt-dev)
 * [hls] Added options `--hls-split-discontinuity` to better support media discontinuity by [shirt](https://github.com/shirt-dev)
-* [ffmpeg] Allow passing custom arguments before -i using `--ppa "ffmpeg_i1:ARGS"` synatax
+* [ffmpeg] Allow passing custom arguments before -i using `--ppa "ffmpeg_i1:ARGS"` syntax
 * Fix `--windows-filenames` removing `/` from UNIX paths
 * [hls] Show warning if pycryptodome is not found
 * [documentation] Improvements
     * Fix documentation of `Extractor Options`
-    * Document `all` in format selection (Closes #101)
+    * Document `all` in format selection
     * Document `playable_in_embed` in output templates
 
 
@@ -276,7 +311,7 @@
     * Added `--video-multistreams`, `--no-video-multistreams`, `--audio-multistreams`, `--no-audio-multistreams`
     * Added `b`,`w`,`v`,`a` as alias for `best`, `worst`, `video` and `audio` respectively
 * **Shortcut Options:** Added `--write-link`, `--write-url-link`, `--write-webloc-link`, `--write-desktop-link` by [h-h-h-h](https://github.com/h-h-h-h) - See [Internet Shortcut Options](README.md#internet-shortcut-options) for details
-* **Sponskrub integration:** Added `--sponskrub`, `--sponskrub-cut`, `--sponskrub-force`, `--sponskrub-location`, `--sponskrub-args` - See [SponSkrub Options](README.md#sponskrub-options-sponsorblock) for details
+* **Sponskrub integration:** Added `--sponskrub`, `--sponskrub-cut`, `--sponskrub-force`, `--sponskrub-location`, `--sponskrub-args` - See [SponSkrub Options](README.md#sponskrub-sponsorblock-options) for details
 * Added `--force-download-archive` (`--force-write-archive`) by [h-h-h-h](https://github.com/h-h-h-h)
 * Added `--list-formats-as-table`,  `--list-formats-old`
 * **Negative Options:** Makes it possible to negate most boolean options by adding a `no-` to the switch. Usefull when you want to reverse an option that is defined in a config file

MANIFEST.in

@@ -1,9 +1,9 @@
-include README.md
-include LICENSE
 include AUTHORS
-include ChangeLog
+include Changelog.md
-include yt-dlp.bash-completion
+include LICENSE
-include yt-dlp.fish
+include README.md
+include completions/*/*
+include supportedsites.md
 include yt-dlp.1
-recursive-include docs Makefile conf.py *.rst
+recursive-include devscripts *
 recursive-include test *

Makefile

@@ -1,12 +1,28 @@
-all: yt-dlp doc man
+all: yt-dlp doc
+clean: clean-test clean-dist clean-cache
+completions: completion-bash completion-fish completion-zsh
 doc: README.md CONTRIBUTING.md issuetemplates supportedsites
-man: README.txt yt-dlp.1 yt-dlp.bash-completion yt-dlp.zsh yt-dlp.fish
+ot: offlinetest
+tar: yt-dlp.tar.gz
 
+# Keep this list in sync with MANIFEST.in
+# intended use: when building a source distribution,
+# make pypi-files && python setup.py sdist
+pypi-files: AUTHORS Changelog.md LICENSE README.md README.txt supportedsites completions yt-dlp.1 devscripts/* test/*
 
-clean:
+.PHONY: all clean install test tar pypi-files completions ot offlinetest codetest supportedsites
-	rm -rf yt-dlp.1.temp.md yt-dlp.1 yt-dlp.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ yt-dlp.tar.gz yt-dlp.zsh yt-dlp.fish yt_dlp/extractor/lazy_extractors.py *.dump *.part* *.ytdl *.info.json *.mp4 *.m4a *.flv *.mp3 *.avi *.mkv *.webm *.3gp *.wav *.ape *.swf *.jpg *.png *.spec *.frag *.frag.urls *.frag.aria2 CONTRIBUTING.md.tmp yt-dlp yt-dlp.exe
+
-	find . -name "*.pyc" -delete
+clean-test:
-	find . -name "*.class" -delete
+	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
+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:
+	find . -name "*.pyc" -o -name "*.class" -delete
+
+completion-bash: completions/bash/yt-dlp
+completion-fish: completions/fish/yt-dlp.fish
+completion-zsh: completions/zsh/_yt-dlp
+lazy-extractors: yt_dlp/extractor/lazy_extractors.py
 
 PREFIX ?= /usr/local
 BINDIR ?= $(PREFIX)/bin
@@ -21,17 +37,12 @@ SYSCONFDIR = $(shell if [ $(PREFIX) = /usr -o $(PREFIX) = /usr/local ]; then ech
 # set markdown input format to "markdown-smart" for pandoc version 2 and to "markdown" for pandoc prior to version 2
 MARKDOWN = $(shell if [ `pandoc -v | head -n1 | cut -d" " -f2 | head -c1` = "2" ]; then echo markdown-smart; else echo markdown; fi)
 
-install: yt-dlp yt-dlp.1 yt-dlp.bash-completion yt-dlp.zsh yt-dlp.fish
+install: yt-dlp yt-dlp.1 completions
-	install -d $(DESTDIR)$(BINDIR)
+	install -Dm755 yt-dlp $(DESTDIR)$(BINDIR)
-	install -m 755 yt-dlp $(DESTDIR)$(BINDIR)
+	install -Dm644 yt-dlp.1 $(DESTDIR)$(MANDIR)/man1
-	install -d $(DESTDIR)$(MANDIR)/man1
+	install -Dm644 completions/bash/yt-dlp $(DESTDIR)$(SHAREDIR)/bash-completion/completions/yt-dlp
-	install -m 644 yt-dlp.1 $(DESTDIR)$(MANDIR)/man1
+	install -Dm644 completions/zsh/_yt-dlp $(DESTDIR)$(SHAREDIR)/zsh/site-functions/_yt-dlp
-	install -d $(DESTDIR)$(SYSCONFDIR)/bash_completion.d
+	install -Dm644 completions/fish/yt-dlp.fish $(DESTDIR)$(SHAREDIR)/fish/vendor_completions.d/yt-dlp.fish
-	install -m 644 yt-dlp.bash-completion $(DESTDIR)$(SYSCONFDIR)/bash_completion.d/yt-dlp
-	install -d $(DESTDIR)$(SHAREDIR)/zsh/site-functions
-	install -m 644 yt-dlp.zsh $(DESTDIR)$(SHAREDIR)/zsh/site-functions/_yt-dlp
-	install -d $(DESTDIR)$(SYSCONFDIR)/fish/completions
-	install -m 644 yt-dlp.fish $(DESTDIR)$(SYSCONFDIR)/fish/completions/yt-dlp.fish
 
 codetest:
 	flake8 .
@@ -41,8 +52,6 @@ test:
 	nosetests --verbose test
 	$(MAKE) codetest
 
-ot: offlinetest
-
 # Keep this list in sync with devscripts/run_tests.sh
 offlinetest: codetest
 	$(PYTHON) -m nose --verbose test \
@@ -57,12 +66,6 @@ offlinetest: codetest
 		--exclude test_youtube_signature.py \
 		--exclude test_post_hooks.py
 
-tar: yt-dlp.tar.gz
-
-.PHONY: all clean install test tar bash-completion pypi-files zsh-completion fish-completion ot offlinetest codetest supportedsites
-
-pypi-files: yt-dlp.bash-completion README.txt yt-dlp.1 yt-dlp.fish
-
 yt-dlp: yt_dlp/*.py yt_dlp/*/*.py
 	mkdir -p zip
 	for d in yt_dlp yt_dlp/downloader yt_dlp/extractor yt_dlp/postprocessor ; do \
@@ -92,7 +95,7 @@ issuetemplates: devscripts/make_issue_template.py .github/ISSUE_TEMPLATE_tmpl/1_
 	$(PYTHON) devscripts/make_issue_template.py .github/ISSUE_TEMPLATE_tmpl/5_feature_request.md .github/ISSUE_TEMPLATE/5_feature_request.md
 
 supportedsites:
-	$(PYTHON) devscripts/make_supportedsites.py docs/supportedsites.md
+	$(PYTHON) devscripts/make_supportedsites.py supportedsites.md
 
 README.txt: README.md
 	pandoc -f $(MARKDOWN) -t plain README.md -o README.txt
@@ -102,29 +105,24 @@ yt-dlp.1: README.md
 	pandoc -s -f $(MARKDOWN) -t man yt-dlp.1.temp.md -o yt-dlp.1
 	rm -f yt-dlp.1.temp.md
 
-yt-dlp.bash-completion: yt_dlp/*.py yt_dlp/*/*.py devscripts/bash-completion.in
+completions/bash/yt-dlp: yt_dlp/*.py yt_dlp/*/*.py devscripts/bash-completion.in
+	mkdir -p completions/bash
 	$(PYTHON) devscripts/bash-completion.py
 
-bash-completion: yt-dlp.bash-completion
+completions/zsh/_yt-dlp: yt_dlp/*.py yt_dlp/*/*.py devscripts/zsh-completion.in
-
+	mkdir -p completions/zsh
-yt-dlp.zsh: yt_dlp/*.py yt_dlp/*/*.py devscripts/zsh-completion.in
 	$(PYTHON) devscripts/zsh-completion.py
 
-zsh-completion: yt-dlp.zsh
+completions/fish/yt-dlp.fish: yt_dlp/*.py yt_dlp/*/*.py devscripts/fish-completion.in
-
+	mkdir -p completions/fish
-yt-dlp.fish: yt_dlp/*.py yt_dlp/*/*.py devscripts/fish-completion.in
 	$(PYTHON) devscripts/fish-completion.py
 
-fish-completion: yt-dlp.fish
-
-lazy-extractors: yt_dlp/extractor/lazy_extractors.py
-
 _EXTRACTOR_FILES = $(shell find yt_dlp/extractor -iname '*.py' -and -not -iname 'lazy_extractors.py')
 yt_dlp/extractor/lazy_extractors.py: devscripts/make_lazy_extractors.py devscripts/lazy_load_template.py $(_EXTRACTOR_FILES)
 	$(PYTHON) devscripts/make_lazy_extractors.py $@
 
-yt-dlp.tar.gz: yt-dlp README.md README.txt yt-dlp.1 yt-dlp.bash-completion yt-dlp.zsh yt-dlp.fish ChangeLog AUTHORS
+yt-dlp.tar.gz: README.md yt-dlp.1 completions Changelog.md AUTHORS
-	@tar -czf yt-dlp.tar.gz --transform "s|^|yt-dlp/|" --owner 0 --group 0 \
+	@tar -czf $(DESTDIR)/yt-dlp.tar.gz --transform "s|^|yt-dlp/|" --owner 0 --group 0 \
 		--exclude '*.DS_Store' \
 		--exclude '*.kate-swp' \
 		--exclude '*.pyc' \
@@ -134,8 +132,13 @@ yt-dlp.tar.gz: yt-dlp README.md README.txt yt-dlp.1 yt-dlp.bash-completion yt-dl
 		--exclude '.git' \
 		--exclude 'docs/_build' \
 		-- \
-		bin devscripts test yt_dlp docs \
+		devscripts test \
-		ChangeLog AUTHORS LICENSE README.md README.txt \
+		Changelog.md AUTHORS LICENSE README.md supportedsites.md \
-		Makefile MANIFEST.in yt-dlp.1 yt-dlp.bash-completion \
+		Makefile MANIFEST.in yt-dlp.1 completions \
-		yt-dlp.zsh yt-dlp.fish setup.py setup.cfg \
+		setup.py setup.cfg yt-dlp
-		yt-dlp
+
+AUTHORS: .mailmap
+	git shortlog -s -n | cut -f2 | sort > AUTHORS
+
+.mailmap:
+	git shortlog -s -e -n | awk '!(out[$$NF]++) { $$1="";sub(/^[ \t]+/,""); print}' > .mailmap

README.md

@@ -11,7 +11,7 @@
 [![PyPi Downloads](https://img.shields.io/pypi/dm/yt-dlp?label=PyPi)](https://pypi.org/project/yt-dlp)
 [![Doc Status](https://readthedocs.org/projects/yt-dlp/badge/?version=latest)](https://yt-dlp.readthedocs.io)
 
-A command-line program to download videos from youtube.com and many other [video platforms](docs/supportedsites.md)
+A command-line program to download videos from youtube.com and many other [video platforms](supportedsites.md)
 
 This is a fork of [youtube-dlc](https://github.com/blackjack4494/yt-dlc) which is inturn a fork of [youtube-dl](https://github.com/ytdl-org/youtube-dl)
 
@@ -57,7 +57,7 @@ The major new features from the latest release of [blackjack4494/yt-dlc](https:/
 
 * **[Format Sorting](#sorting-formats)**: The default format sorting options have been changed so that higher resolution and better codecs will be now preferred instead of simply using larger bitrate. Furthermore, you can now specify the sort order using `-S`. This allows for much easier format selection that what is possible by simply using `--format` ([examples](#format-selection-examples))
 
-* **Merged with youtube-dl v2021.02.10**: You get all the latest features and patches of [youtube-dl](https://github.com/ytdl-org/youtube-dl) in addition to all the features of [youtube-dlc](https://github.com/blackjack4494/yt-dlc)
+* **Merged with youtube-dl v2021.03.03**: You get all the latest features and patches of [youtube-dl](https://github.com/ytdl-org/youtube-dl) in addition to all the features of [youtube-dlc](https://github.com/blackjack4494/yt-dlc)
 
 * **Merged with animelover1984/youtube-dl**: You get most of the features and improvements from [animelover1984/youtube-dl](https://github.com/animelover1984/youtube-dl) including `--get-comments`, `BiliBiliSearch`, `BilibiliChannel`, Embedding thumbnail in mp4/ogg/opus, Playlist infojson etc. Note that the NicoNico improvements are not available. See [#31](https://github.com/yt-dlp/yt-dlp/pull/31) for details.
 
@@ -92,7 +92,7 @@ See [changelog](Changelog.md) or [commits](https://github.com/yt-dlp/yt-dlp/comm
 
 **PS**: Some of these changes are already in youtube-dlc, but are still unreleased. See [this](Changelog.md#unreleased-changes-in-blackjack4494yt-dlc) for details
 
-If you are coming from [youtube-dl](https://github.com/ytdl-org/youtube-dl), the amount of changes are very large. Compare [options](#options) and [supported sites](docs/supportedsites.md) with youtube-dl's to get an idea of the massive number of features/patches [youtube-dlc](https://github.com/blackjack4494/yt-dlc) has accumulated.
+If you are coming from [youtube-dl](https://github.com/ytdl-org/youtube-dl), the amount of changes are very large. Compare [options](#options) and [supported sites](supportedsites.md) with youtube-dl's to get an idea of the massive number of features/patches [youtube-dlc](https://github.com/blackjack4494/yt-dlc) has accumulated.
 
 
 # INSTALLATION
@@ -122,12 +122,12 @@ You can also build the executable without any version info or metadata by using:
 
 **For Unix**:
 You will need the required build tools: `python`, `make` (GNU), `pandoc`, `zip`, `nosetests`  
-Then simply run `make`. You can also run `make youtube_dlc` instead to compile only the binary without updating any of the additional files
+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
 
 # DESCRIPTION
-**yt-dlp** is a command-line program to download videos from youtube.com many other [video platforms](docs/supportedsites.md). It requires the Python interpreter, version 2.6, 2.7, or 3.2+, and it is not platform specific. It should work on your Unix box, on Windows or on macOS. It is released to the public domain, which means you can modify it, redistribute it or use it however you like.
+**yt-dlp** is a command-line program to download videos from youtube.com many other [video platforms](supportedsites.md). It requires the Python interpreter, version 2.6, 2.7, or 3.2+, and it is not platform specific. It should work on your Unix box, on Windows or on macOS. It is released to the public domain, which means you can modify it, redistribute it or use it however you like.
 
     yt-dlp [OPTIONS] [--] URL [URL...]
 
@@ -245,7 +245,7 @@ Then simply run `make`. You can also run `make youtube_dlc` instead to compile o
                                      "OUTPUT TEMPLATE" for a list of available
                                      keys) to match if the key is present, !key
                                      to check if the key is not present,
-                                     key>NUMBER (like "comment_count > 12", also
+                                     key>NUMBER (like "view_count > 12", also
                                      works with >=, <, <=, !=, =) to compare
                                      against a number, key = 'LITERAL' (like
                                      "uploader = 'Mike Smith'", also works with
@@ -317,13 +317,19 @@ Then simply run `make`. You can also run `make youtube_dlc` instead to compile o
                                      ffmpeg
     --hls-prefer-ffmpeg              Use ffmpeg instead of the native HLS
                                      downloader
-    --hls-use-mpegts                 Use the mpegts container for HLS videos,
+    --hls-use-mpegts                 Use the mpegts container for HLS videos;
-                                     allowing to play the video while
+                                     allowing some players to play the video
-                                     downloading (some players may not be able
+                                     while downloading, and reducing the chance
-                                     to play it)
+                                     of file corruption if download is
-    --external-downloader NAME       Use the specified external downloader.
+                                     interrupted. This is enabled by default for
-                                     Currently supports aria2c, avconv, axel,
+                                     live streams
-                                     curl, ffmpeg, httpie, wget
+    --no-hls-use-mpegts              Do not use the mpegts container for HLS
+                                     videos. This is default when not
+                                     downloading live streams
+    --external-downloader NAME       Name or path of the external downloader to
+                                     use. Currently supports aria2c, avconv,
+                                     axel, curl, ffmpeg, httpie, wget
+                                     (Recommended: aria2c)
     --downloader-args NAME:ARGS      Give these arguments to the external
                                      downloader. Specify the downloader name and
                                      the arguments separated by a colon ":". You
@@ -397,7 +403,9 @@ Then simply run `make`. You can also run `make youtube_dlc` instead to compile o
     --no-write-playlist-metafiles    Do not write playlist metadata when using
                                      --write-info-json, --write-description etc.
     --get-comments                   Retrieve video comments to be placed in the
-                                     .info.json file
+                                     .info.json file. The comments are fetched
+                                     even without this option if the extraction
+                                     is known to be quick
     --load-info-json FILE            JSON file containing the video information
                                      (created with the "--write-info-json"
                                      option)
@@ -485,6 +493,8 @@ Then simply run `make`. You can also run `make youtube_dlc` instead to compile o
     --bidi-workaround                Work around terminals that lack
                                      bidirectional text support. Requires bidiv
                                      or fribidi executable in PATH
+    --sleep-requests SECONDS         Number of seconds to sleep between requests
+                                     during data extraction
     --sleep-interval SECONDS         Number of seconds to sleep before each
                                      download when used alone or a lower bound
                                      of a range for randomized sleep before each
@@ -495,7 +505,8 @@ Then simply run `make`. You can also run `make youtube_dlc` instead to compile o
                                      before each download (maximum possible
                                      number of seconds to sleep). Must only be
                                      used along with --min-sleep-interval
-    --sleep-subtitles SECONDS        Enforce sleep interval on subtitles as well
+    --sleep-subtitles SECONDS        Number of seconds to sleep before each
+                                     subtitle download
 
 ## Video Format Options:
     -f, --format FORMAT              Video format code, see "FORMAT SELECTION"
@@ -641,7 +652,7 @@ Then simply run `make`. You can also run `make youtube_dlc` instead to compile o
                                      similar syntax to the output template can
                                      also be used. The parsed parameters replace
                                      any existing values and can be use in
-                                     output templateThis option can be used
+                                     output template. This option can be used
                                      multiple times. Example: --parse-metadata
                                      "title:%(artist)s - %(title)s" matches a
                                      title like "Coldplay - Paradise". Example
@@ -686,6 +697,8 @@ Then simply run `make`. You can also run `make youtube_dlc` instead to compile o
                                      directory
 
 ## Extractor Options:
+    --extractor-retries RETRIES      Number of retries for known extractor
+                                     errors (default is 3), or "infinite"
     --allow-dynamic-mpd              Process dynamic DASH manifests (default)
                                      (Alias: --no-ignore-dynamic-mpd)
     --ignore-dynamic-mpd             Do not process dynamic DASH manifests
@@ -805,7 +818,7 @@ The available fields are:
  - `dislike_count` (numeric): Number of negative ratings of the video
  - `repost_count` (numeric): Number of reposts of the video
  - `average_rating` (numeric): Average rating give by users, the scale used depends on the webpage
- - `comment_count` (numeric): Number of comments on the video
+ - `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)
  - `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

devscripts/bash-completion.py

@@ -8,7 +8,7 @@ import sys
 sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
 import yt_dlp
 
-BASH_COMPLETION_FILE = "yt-dlp.bash-completion"
+BASH_COMPLETION_FILE = "completions/bash/yt-dlp"
 BASH_COMPLETION_TEMPLATE = "devscripts/bash-completion.in"
 
 

devscripts/fish-completion.py

@@ -10,7 +10,7 @@ sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
 import yt_dlp
 from yt_dlp.utils import shell_quote
 
-FISH_COMPLETION_FILE = 'yt-dlp.fish'
+FISH_COMPLETION_FILE = 'completions/fish/yt-dlp.fish'
 FISH_COMPLETION_TEMPLATE = 'devscripts/fish-completion.in'
 
 EXTRA_ARGS = {

devscripts/release.sh

@@ -61,7 +61,7 @@ if ! type pandoc >/dev/null 2>/dev/null; then echo 'ERROR: pandoc is missing'; e
 if ! python3 -c 'import rsa' 2>/dev/null; then echo 'ERROR: python3-rsa is missing'; exit 1; fi
 if ! python3 -c 'import wheel' 2>/dev/null; then echo 'ERROR: wheel is missing'; exit 1; fi
 
-read -p "Is ChangeLog up to date? (y/n) " -n 1
+read -p "Is Changelog up to date? (y/n) " -n 1
 if [[ ! $REPLY =~ ^[Yy]$ ]]; then exit 1; fi
 
 /bin/echo -e "\n### First of all, testing..."
@@ -75,12 +75,12 @@ fi
 /bin/echo -e "\n### Changing version in version.py..."
 sed -i "s/__version__ = '.*'/__version__ = '$version'/" yt_dlp/version.py
 
-/bin/echo -e "\n### Changing version in ChangeLog..."
+/bin/echo -e "\n### Changing version in Changelog..."
-sed -i "s/<unreleased>/$version/" ChangeLog
+sed -i "s/<unreleased>/$version/" Changelog.md
 
 /bin/echo -e "\n### Committing documentation, templates and yt_dlp/version.py..."
 make README.md CONTRIBUTING.md issuetemplates supportedsites
-git add README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE/1_broken_site.md .github/ISSUE_TEMPLATE/2_site_support_request.md .github/ISSUE_TEMPLATE/3_site_feature_request.md .github/ISSUE_TEMPLATE/4_bug_report.md .github/ISSUE_TEMPLATE/5_feature_request.md .github/ISSUE_TEMPLATE/6_question.md docs/supportedsites.md yt_dlp/version.py ChangeLog
+git add README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE/1_broken_site.md .github/ISSUE_TEMPLATE/2_site_support_request.md .github/ISSUE_TEMPLATE/3_site_feature_request.md .github/ISSUE_TEMPLATE/4_bug_report.md .github/ISSUE_TEMPLATE/5_feature_request.md .github/ISSUE_TEMPLATE/6_question.md docs/supportedsites.md yt_dlp/version.py Changelog.md
 git commit $gpg_sign_commits -m "release $version"
 
 /bin/echo -e "\n### Now tagging, signing and pushing..."
@@ -111,7 +111,7 @@ RELEASE_FILES="yt-dlp yt-dlp.exe yt-dlp-$version.tar.gz"
 for f in $RELEASE_FILES; do gpg --passphrase-repeat 5 --detach-sig "build/$version/$f"; done
 
 ROOT=$(pwd)
-python devscripts/create-github-release.py ChangeLog $version "$ROOT/build/$version"
+python devscripts/create-github-release.py Changelog.md $version "$ROOT/build/$version"
 
 ssh ytdl@yt-dl.org "sh html/update_latest.sh $version"
 

devscripts/zsh-completion.py

@@ -8,7 +8,7 @@ import sys
 sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
 import yt_dlp
 
-ZSH_COMPLETION_FILE = "yt-dlp.zsh"
+ZSH_COMPLETION_FILE = "completions/zsh/_yt-dlp"
 ZSH_COMPLETION_TEMPLATE = "devscripts/zsh-completion.in"
 
 

docs/Changelog.md

@@ -0,0 +1,5 @@
+---
+orphan: true
+---
+```{include} ../Changelog.md
+```

docs/LICENSE.md

@@ -0,0 +1,6 @@
+---
+orphan: true
+---
+# LICENSE
+```{include} ../LICENSE
+```

docs/README.md

@@ -0,0 +1,2 @@
+```{include} ../README.md
+```

docs/conf.py

@@ -7,26 +7,21 @@ import os
 
 # Allows to import yt-dlp
 sys.path.insert(0, os.path.abspath('..'))
-from recommonmark.transform import AutoStructify
 
 # -- General configuration ------------------------------------------------
 
-# The suffix of source filenames.
-source_suffix = ['.rst', '.md']
-
 # Add any Sphinx extension module names here, as strings. They can be
 # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
 # ones.
 extensions = [
-    'sphinx.ext.autodoc',
+    'myst_parser',
-    'recommonmark',
 ]
 
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']
 
 # The master toctree document.
-master_doc = 'index'
+master_doc = 'README'
 
 # General information about the project.
 project = u'yt-dlp'
@@ -64,12 +59,10 @@ highlight_language = 'none'
 # so a file named "default.css" will overwrite the builtin "default.css".
 # html_static_path = ['_static']
 
+# Enable heading anchors
+myst_heading_anchors = 4
 
-def setup(app):
+# Suppress heading warnings
-    app.add_config_value('recommonmark_config', {
+suppress_warnings = [
-        'enable_math': False,
+    'myst.header',
-        'enable_inline_math': False,
+]
-        'enable_eval_rst': True,
-        'enable_auto_toc_tree': True,
-    }, True)
-    app.add_transform(AutoStructify)

docs/index.md

@@ -1 +0,0 @@
-../README.md

docs/requirements.txt

@@ -1,2 +1 @@
-recommonmark>=0.6.0
+myst-parser
-m2r2

docs/ytdlp_plugins.md

@@ -0,0 +1,6 @@
+---
+orphan: true
+---
+# ytdlp_plugins
+
+See [https://github.com/yt-dlp/yt-dlp/tree/master/ytdlp_plugins](https://github.com/yt-dlp/yt-dlp/tree/master/ytdlp_plugins).

setup.py

@@ -27,8 +27,9 @@ if len(sys.argv) >= 2 and sys.argv[1] == 'py2exe':
     print("inv")
 else:
     files_spec = [
-        ('etc/bash_completion.d', ['yt-dlp.bash-completion']),
+        ('share/bash-completion/completions', ['completions/bash/yt-dlp']),
-        ('etc/fish/completions', ['yt-dlp.fish']),
+        ('share/zsh/site-functions', ['completions/zsh/_yt-dlp']),
+        ('share/fish/vendor_completions.d', ['completions/fish/yt-dlp.fish']),
         ('share/doc/yt_dlp', ['README.txt']),
         ('share/man/man1', ['yt-dlp.1'])
     ]
@@ -38,7 +39,7 @@ else:
         resfiles = []
         for fn in files:
             if not os.path.exists(fn):
-                warnings.warn('Skipping file %s since it is not present. Type  make  to build all automatically generated files.' % fn)
+                warnings.warn('Skipping file %s since it is not present. Try running `make pypi-files` first.' % fn)
             else:
                 resfiles.append(fn)
         data_files.append((dirname, resfiles))
@@ -86,7 +87,7 @@ setup(
         #'Funding': 'https://donate.pypi.org',
     },
     classifiers=[
-	    "Topic :: Multimedia :: Video",
+        "Topic :: Multimedia :: Video",
         "Development Status :: 5 - Production/Stable",
         "Environment :: Console",
         "Programming Language :: Python",
@@ -110,7 +111,7 @@ setup(
         "Operating System :: OS Independent",
     ],
     python_requires='>=2.6',
-	
+
-	cmdclass={'build_lazy_extractors': build_lazy_extractors},
+    cmdclass={'build_lazy_extractors': build_lazy_extractors},
     **params
 )

yt_dlp/YoutubeDL.py

@@ -324,6 +324,8 @@ class YoutubeDL(object):
     source_address:    Client-side IP address to bind to.
     call_home:         Boolean, true iff we are allowed to contact the
                        yt-dlp servers for debugging. (BROKEN)
+    sleep_interval_requests: Number of seconds to sleep between requests
+                       during extraction
     sleep_interval:    Number of seconds to sleep before each download when
                        used alone or a lower bound of a range for randomized
                        sleep before each download (minimum possible number
@@ -334,6 +336,7 @@ class YoutubeDL(object):
                        Must only be used along with sleep_interval.
                        Actual sleep time will be a random float from range
                        [sleep_interval; max_sleep_interval].
+    sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
     listformats:       Print an overview of available video formats and exit.
     list_thumbnails:   Print a table of all thumbnails and exit.
     match_filter:      A function that gets called with the info_dict of
@@ -378,17 +381,18 @@ class YoutubeDL(object):
                         Use 'default' as the name for arguments to passed to all PP
 
     The following options are used by the extractors:
-    dynamic_mpd:        Whether to process dynamic DASH manifests (default: True)
+    extractor_retries: Number of times to retry for known errors
+    dynamic_mpd:       Whether to process dynamic DASH manifests (default: True)
     hls_split_discontinuity: Split HLS playlists to different formats at
-                        discontinuities such as ad breaks (default: False)
+                       discontinuities such as ad breaks (default: False)
     youtube_include_dash_manifest: If True (default), DASH manifests and related
-                        data will be downloaded and processed by extractor.
+                       data will be downloaded and processed by extractor.
-                        You can reduce network I/O by disabling it if you don't
+                       You can reduce network I/O by disabling it if you don't
-                        care about DASH. (only for youtube)
+                       care about DASH. (only for youtube)
     youtube_include_hls_manifest: If True (default), HLS manifests and related
-                        data will be downloaded and processed by extractor.
+                       data will be downloaded and processed by extractor.
-                        You can reduce network I/O by disabling it if you don't
+                       You can reduce network I/O by disabling it if you don't
-                        care about HLS. (only for youtube)
+                       care about HLS. (only for youtube)
     """
 
     _NUMERIC_FIELDS = set((
@@ -406,6 +410,7 @@ class YoutubeDL(object):
     _ies = []
     _pps = {'beforedl': [], 'aftermove': [], 'normal': []}
     __prepare_filename_warned = False
+    _first_webpage_request = True
     _download_retcode = None
     _num_downloads = None
     _playlist_level = 0
@@ -420,6 +425,7 @@ class YoutubeDL(object):
         self._ies_instances = {}
         self._pps = {'beforedl': [], 'aftermove': [], 'normal': []}
         self.__prepare_filename_warned = False
+        self._first_webpage_request = True
         self._post_hooks = []
         self._progress_hooks = []
         self._download_retcode = 0
@@ -2036,6 +2042,7 @@ class YoutubeDL(object):
             self.to_stdout(formatSeconds(info_dict['duration']))
         print_mandatory('format')
         if self.params.get('forcejson', False):
+            self.post_extract(info_dict)
             self.to_stdout(json.dumps(info_dict))
 
     def process_info(self, info_dict):
@@ -2059,6 +2066,7 @@ class YoutubeDL(object):
         if self._match_entry(info_dict, incomplete=False) is not None:
             return
 
+        self.post_extract(info_dict)
         self._num_downloads += 1
 
         info_dict = self.pre_process(info_dict)
@@ -2166,15 +2174,6 @@ class YoutubeDL(object):
                     else:
                         try:
                             dl(sub_filename, sub_info, subtitle=True)
-                            '''
-                            if self.params.get('sleep_interval_subtitles', False):
-                                dl(sub_filename, sub_info)
-                            else:
-                                sub_data = ie._request_webpage(
-                                    sub_info['url'], info_dict['id'], note=False).read()
-                                with io.open(encodeFilename(sub_filename), 'wb') as subfile:
-                                    subfile.write(sub_data)
-                            '''
                             files_to_move[sub_filename] = sub_filename_final
                         except (ExtractorError, IOError, OSError, ValueError, compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
                             self.report_warning('Unable to download subtitle for "%s": %s' %
@@ -2501,6 +2500,7 @@ class YoutubeDL(object):
                 raise
             else:
                 if self.params.get('dump_single_json', False):
+                    self.post_extract(res)
                     self.to_stdout(json.dumps(res))
 
         return self._download_retcode
@@ -2549,6 +2549,24 @@ class YoutubeDL(object):
                     del files_to_move[old_filename]
         return files_to_move, infodict
 
+    @staticmethod
+    def post_extract(info_dict):
+        def actual_post_extract(info_dict):
+            if info_dict.get('_type') in ('playlist', 'multi_video'):
+                for video_dict in info_dict.get('entries', {}):
+                    actual_post_extract(video_dict)
+                return
+
+            if '__post_extractor' not in info_dict:
+                return
+            post_extractor = info_dict['__post_extractor']
+            if post_extractor:
+                info_dict.update(post_extractor().items())
+            del info_dict['__post_extractor']
+            return
+
+        actual_post_extract(info_dict)
+
     def pre_process(self, ie_info):
         info = dict(ie_info)
         for pp in self._pps['beforedl']:
@@ -2940,7 +2958,7 @@ class YoutubeDL(object):
                 self.to_screen('[%s] %s: Thumbnail %sis already present' %
                                (info_dict['extractor'], info_dict['id'], thumb_display_id))
             else:
-                self.to_screen('[%s] %s: Downloading thumbnail %s...' %
+                self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
                                (info_dict['extractor'], info_dict['id'], thumb_display_id))
                 try:
                     uf = self.urlopen(t['url'])

yt_dlp/__init__.py

@@ -169,25 +169,33 @@ def _real_main(argv=None):
             parser.error('max sleep interval must be greater than or equal to min sleep interval')
     else:
         opts.max_sleep_interval = opts.sleep_interval
+    if opts.sleep_interval_subtitles is not None:
+        if opts.sleep_interval_subtitles < 0:
+            parser.error('subtitles sleep interval must be positive or 0')
+    if opts.sleep_interval_requests is not None:
+        if opts.sleep_interval_requests < 0:
+            parser.error('requests sleep interval must be positive or 0')
     if opts.ap_mso and opts.ap_mso not in MSO_INFO:
         parser.error('Unsupported TV Provider, use --ap-list-mso to get a list of supported TV Providers')
     if opts.overwrites:
         # --yes-overwrites implies --no-continue
         opts.continue_dl = False
 
-    def parse_retries(retries):
+    def parse_retries(retries, name=''):
         if retries in ('inf', 'infinite'):
             parsed_retries = float('inf')
         else:
             try:
                 parsed_retries = int(retries)
             except (TypeError, ValueError):
-                parser.error('invalid retry count specified')
+                parser.error('invalid %sretry count specified' % name)
         return parsed_retries
     if opts.retries is not None:
         opts.retries = parse_retries(opts.retries)
     if opts.fragment_retries is not None:
-        opts.fragment_retries = parse_retries(opts.fragment_retries)
+        opts.fragment_retries = parse_retries(opts.fragment_retries, 'fragment ')
+    if opts.extractor_retries is not None:
+        opts.extractor_retries = parse_retries(opts.extractor_retries, 'extractor ')
     if opts.buffersize is not None:
         numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
         if numeric_buffersize is None:
@@ -262,6 +270,11 @@ def _real_main(argv=None):
     any_printing = opts.print_json
     download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive
 
+    # If JSON is not printed anywhere, but comments are requested, save it to file
+    printing_json = opts.dumpjson or opts.print_json or opts.dump_single_json
+    if opts.getcomments and not printing_json:
+        opts.writeinfojson = True
+
     def report_conflict(arg1, arg2):
         write_string('WARNING: %s is ignored since %s was given\n' % (arg2, arg1), out=sys.stderr)
     if opts.remuxvideo and opts.recodevideo:
@@ -447,6 +460,7 @@ def _real_main(argv=None):
         'overwrites': opts.overwrites,
         'retries': opts.retries,
         'fragment_retries': opts.fragment_retries,
+        'extractor_retries': opts.extractor_retries,
         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
         'keep_fragments': opts.keep_fragments,
         'buffersize': opts.buffersize,
@@ -466,7 +480,7 @@ def _real_main(argv=None):
         'updatetime': opts.updatetime,
         'writedescription': opts.writedescription,
         'writeannotations': opts.writeannotations,
-        'writeinfojson': opts.writeinfojson or opts.getcomments,
+        'writeinfojson': opts.writeinfojson,
         'allow_playlist_files': opts.allow_playlist_files,
         'getcomments': opts.getcomments,
         'writethumbnail': opts.writethumbnail,
@@ -524,6 +538,7 @@ def _real_main(argv=None):
         'fixup': opts.fixup,
         'source_address': opts.source_address,
         'call_home': opts.call_home,
+        'sleep_interval_requests': opts.sleep_interval_requests,
         'sleep_interval': opts.sleep_interval,
         'max_sleep_interval': opts.max_sleep_interval,
         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
@@ -541,7 +556,6 @@ def _real_main(argv=None):
         'postprocessor_args': opts.postprocessor_args,
         'cn_verification_proxy': opts.cn_verification_proxy,
         'geo_verification_proxy': opts.geo_verification_proxy,
-        'config_location': opts.config_location,
         'geo_bypass': opts.geo_bypass,
         'geo_bypass_country': opts.geo_bypass_country,
         'geo_bypass_ip_block': opts.geo_bypass_ip_block,

yt_dlp/downloader/__init__.py

@@ -53,7 +53,7 @@ def get_suitable_downloader(info_dict, params={}, default=HttpFD):
     external_downloader = params.get('external_downloader')
     if external_downloader is not None:
         ed = get_external_downloader(external_downloader)
-        if ed.can_download(info_dict):
+        if ed.can_download(info_dict, external_downloader):
             return ed
 
     if protocol.startswith('m3u8'):

yt_dlp/downloader/common.py

@@ -312,7 +312,7 @@ class FileDownloader(object):
     def report_retry(self, err, count, retries):
         """Report retry in case of HTTP error 5xx"""
         self.to_screen(
-            '[download] Got server HTTP error: %s. Retrying (attempt %d of %s)...'
+            '[download] Got server HTTP error: %s. Retrying (attempt %d of %s) ...'
             % (error_to_compat_str(err), count, self.format_retries(retries)))
 
     def report_file_already_downloaded(self, file_name):
@@ -359,7 +359,7 @@ class FileDownloader(object):
                 max_sleep_interval = self.params.get('max_sleep_interval', min_sleep_interval)
                 sleep_interval = random.uniform(min_sleep_interval, max_sleep_interval)
                 self.to_screen(
-                    '[download] Sleeping %s seconds...' % (
+                    '[download] Sleeping %s seconds ...' % (
                         int(sleep_interval) if sleep_interval.is_integer()
                         else '%.2f' % sleep_interval))
                 time.sleep(sleep_interval)
@@ -369,7 +369,7 @@ class FileDownloader(object):
                 sleep_interval_sub = self.params.get('sleep_interval_subtitles')
             if sleep_interval_sub > 0:
                 self.to_screen(
-                    '[download] Sleeping %s seconds...' % (
+                    '[download] Sleeping %s seconds ...' % (
                         sleep_interval_sub))
                 time.sleep(sleep_interval_sub)
         return self.real_download(filename, info_dict), True

yt_dlp/downloader/external.py

@@ -85,16 +85,16 @@ class ExternalFD(FileDownloader):
         return self.params.get('external_downloader')
 
     @classmethod
-    def available(cls):
+    def available(cls, path=None):
-        return check_executable(cls.get_basename(), [cls.AVAILABLE_OPT])
+        return check_executable(path or cls.get_basename(), [cls.AVAILABLE_OPT])
 
     @classmethod
     def supports(cls, info_dict):
         return info_dict['protocol'] in cls.SUPPORTED_PROTOCOLS
 
     @classmethod
-    def can_download(cls, info_dict):
+    def can_download(cls, info_dict, path=None):
-        return cls.available() and cls.supports(info_dict)
+        return cls.available(path) and cls.supports(info_dict)
 
     def _option(self, command_option, param):
         return cli_option(self.params, command_option, param)
@@ -398,7 +398,10 @@ class FFmpegFD(ExternalFD):
             args += ['-fs', compat_str(self._TEST_FILE_SIZE)]
 
         if protocol in ('m3u8', 'm3u8_native'):
-            if self.params.get('hls_use_mpegts', False) or tmpfilename == '-':
+            use_mpegts = (tmpfilename == '-') or self.params.get('hls_use_mpegts')
+            if use_mpegts is None:
+                use_mpegts = info_dict.get('is_live')
+            if use_mpegts:
                 args += ['-f', 'mpegts']
             else:
                 args += ['-f', 'mp4']

yt_dlp/downloader/fragment.py

@@ -55,11 +55,11 @@ 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)...'
+            '[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):
-        self.to_screen('[download] Skipping fragment %d...' % frag_index)
+        self.to_screen('[download] Skipping fragment %d ...' % frag_index)
 
     def _prepare_url(self, info_dict, url):
         headers = info_dict.get('http_headers')
@@ -174,7 +174,7 @@ class FragmentFD(FileDownloader):
                         '.ytdl file is corrupt' if is_corrupt else
                         'Inconsistent state of incomplete fragment download')
                     self.report_warning(
-                        '%s. Restarting from the beginning...' % message)
+                        '%s. Restarting from the beginning ...' % message)
                     ctx['fragment_index'] = resume_len = 0
                     if 'ytdl_corrupt' in ctx:
                         del ctx['ytdl_corrupt']

yt_dlp/downloader/niconico.py

@@ -29,7 +29,7 @@ class NiconicoDmcFD(FileDownloader):
         heartbeat_url = heartbeat_info_dict['url']
         heartbeat_data = heartbeat_info_dict['data']
         heartbeat_interval = heartbeat_info_dict.get('interval', 30)
-        self.to_screen('[%s] Heartbeat with %s second interval...' % (self.FD_NAME, heartbeat_interval))
+        self.to_screen('[%s] Heartbeat with %s second interval ...' % (self.FD_NAME, heartbeat_interval))
 
         def heartbeat():
             try:

yt_dlp/extractor/bandaichannel.py

@@ -0,0 +1,37 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .brightcove import BrightcoveNewIE
+from ..utils import extract_attributes
+
+
+class BandaiChannelIE(BrightcoveNewIE):
+    IE_NAME = 'bandaichannel'
+    _VALID_URL = r'https?://(?:www\.)?b-ch\.com/titles/(?P<id>\d+/\d+)'
+    _TESTS = [{
+        'url': 'https://www.b-ch.com/titles/514/001',
+        'md5': 'a0f2d787baa5729bed71108257f613a4',
+        'info_dict': {
+            'id': '6128044564001',
+            'ext': 'mp4',
+            'title': 'メタルファイターMIKU 第1話',
+            'timestamp': 1580354056,
+            'uploader_id': '5797077852001',
+            'upload_date': '20200130',
+            'duration': 1387.733,
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        attrs = extract_attributes(self._search_regex(
+            r'(<video-js[^>]+\bid="bcplayer"[^>]*>)', webpage, 'player'))
+        bc = self._download_json(
+            'https://pbifcd.b-ch.com/v1/playbackinfo/ST/70/' + attrs['data-info'],
+            video_id, headers={'X-API-KEY': attrs['data-auth'].strip()})['bc']
+        return self._parse_brightcove_metadata(bc, bc['id'])

yt_dlp/extractor/bbc.py

@@ -5,10 +5,15 @@ import itertools
 import re
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_etree_Element,
+    compat_HTTPError,
+    compat_urlparse,
+)
 from ..utils import (
+    ExtractorError,
     clean_html,
     dict_get,
-    ExtractorError,
     float_or_none,
     get_element_by_class,
     int_or_none,
@@ -21,11 +26,6 @@ from ..utils import (
     urlencode_postdata,
     urljoin,
 )
-from ..compat import (
-    compat_etree_Element,
-    compat_HTTPError,
-    compat_urlparse,
-)
 
 
 class BBCCoUkIE(InfoExtractor):
@@ -793,6 +793,20 @@ class BBCIE(BBCCoUkIE):
             'description': 'Learn English words and phrases from this story',
         },
         'add_ie': [BBCCoUkIE.ie_key()],
+    }, {
+        # BBC Reel
+        'url': 'https://www.bbc.com/reel/video/p07c6sb6/how-positive-thinking-is-harming-your-happiness',
+        'info_dict': {
+            'id': 'p07c6sb9',
+            'ext': 'mp4',
+            'title': 'How positive thinking is harming your happiness',
+            'alt_title': 'The downsides of positive thinking',
+            'description': 'md5:fad74b31da60d83b8265954ee42d85b4',
+            'duration': 235,
+            'thumbnail': r're:https?://.+/p07c9dsr.jpg',
+            'upload_date': '20190604',
+            'categories': ['Psychology'],
+        },
     }]
 
     @classmethod
@@ -929,7 +943,7 @@ class BBCIE(BBCCoUkIE):
                                     else:
                                         entry['title'] = info['title']
                                         entry['formats'].extend(info['formats'])
-                                except Exception as e:
+                                except ExtractorError as e:
                                     # Some playlist URL may fail with 500, at the same time
                                     # the other one may work fine (e.g.
                                     # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
@@ -980,6 +994,37 @@ class BBCIE(BBCCoUkIE):
                 'subtitles': subtitles,
             }
 
+        # bbc reel (e.g. https://www.bbc.com/reel/video/p07c6sb6/how-positive-thinking-is-harming-your-happiness)
+        initial_data = self._parse_json(self._html_search_regex(
+            r'<script[^>]+id=(["\'])initial-data\1[^>]+data-json=(["\'])(?P<json>(?:(?!\2).)+)',
+            webpage, 'initial data', default='{}', group='json'), playlist_id, fatal=False)
+        if initial_data:
+            init_data = try_get(
+                initial_data, lambda x: x['initData']['items'][0], dict) or {}
+            smp_data = init_data.get('smpData') or {}
+            clip_data = try_get(smp_data, lambda x: x['items'][0], dict) or {}
+            version_id = clip_data.get('versionID')
+            if version_id:
+                title = smp_data['title']
+                formats, subtitles = self._download_media_selector(version_id)
+                self._sort_formats(formats)
+                image_url = smp_data.get('holdingImageURL')
+                display_date = init_data.get('displayDate')
+                topic_title = init_data.get('topicTitle')
+
+                return {
+                    'id': version_id,
+                    'title': title,
+                    'formats': formats,
+                    'alt_title': init_data.get('shortTitle'),
+                    'thumbnail': image_url.replace('$recipe', 'raw') if image_url else None,
+                    'description': smp_data.get('summary') or init_data.get('shortSummary'),
+                    'upload_date': display_date.replace('-', '') if display_date else None,
+                    'subtitles': subtitles,
+                    'duration': int_or_none(clip_data.get('duration')),
+                    'categories': [topic_title] if topic_title else None,
+                }
+
         # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
         # There are several setPayload calls may be present but the video
         # seems to be always related to the first one
@@ -1041,7 +1086,7 @@ class BBCIE(BBCCoUkIE):
                 thumbnail = None
                 image_url = current_programme.get('image_url')
                 if image_url:
-                    thumbnail = image_url.replace('{recipe}', '1920x1920')
+                    thumbnail = image_url.replace('{recipe}', 'raw')
                 return {
                     'id': programme_id,
                     'title': title,

yt_dlp/extractor/bilibili.py

@@ -255,10 +255,6 @@ class BiliBiliIE(InfoExtractor):
             info['uploader'] = self._html_search_meta(
                 'author', webpage, 'uploader', default=None)
 
-        comments = None
-        if self._downloader.params.get('getcomments', False):
-            comments = self._get_all_comment_pages(video_id)
-
         raw_danmaku = self._get_raw_danmaku(video_id, cid)
 
         raw_tags = self._get_tags(video_id)
@@ -266,11 +262,18 @@ class BiliBiliIE(InfoExtractor):
 
         top_level_info = {
             'raw_danmaku': raw_danmaku,
-            'comments': comments,
-            'comment_count': len(comments) if comments is not None else None,
             'tags': tags,
             'raw_tags': raw_tags,
         }
+        if self._downloader.params.get('getcomments', False):
+            def get_comments():
+                comments = self._get_all_comment_pages(video_id)
+                return {
+                    'comments': comments,
+                    'comment_count': len(comments)
+                }
+
+            top_level_info['__post_extractor'] = get_comments
 
         '''
         # Requires https://github.com/m13253/danmaku2ass which is licenced under GPL3
@@ -555,6 +558,7 @@ class BilibiliAudioIE(BilibiliAudioBaseIE):
         formats = [{
             'url': play_data['cdns'][0],
             'filesize': int_or_none(play_data.get('size')),
+            'vcodec': 'none'
         }]
 
         song = self._call_api('song/info', au_id)

yt_dlp/extractor/common.py

@@ -294,6 +294,14 @@ class InfoExtractor(object):
                     players on other sites. Can be True (=always allowed),
                     False (=never allowed), None (=unknown), or a string
                     specifying the criteria for embedability (Eg: 'whitelist').
+    __post_extractor: A function to be called just before the metadata is
+                    written to either disk, logger or console. The function
+                    must return a dict which will be added to the info_dict.
+                    This is usefull for additional information that is
+                    time-consuming to extract. Note that the fields thus
+                    extracted will not be available to output template and
+                    match_filter. So, only "comments" and "comment_count" are
+                    currently allowed to be extracted via this method.
 
     The following fields should only be used when the video belongs to some logical
     chapter or section:
@@ -606,6 +614,14 @@ class InfoExtractor(object):
 
         See _download_webpage docstring for arguments specification.
         """
+        if not self._downloader._first_webpage_request:
+            sleep_interval = float_or_none(self._downloader.params.get('sleep_interval_requests')) or 0
+            if sleep_interval > 0:
+                self.to_screen('Sleeping %s seconds ...' % sleep_interval)
+                time.sleep(sleep_interval)
+        else:
+            self._downloader._first_webpage_request = False
+
         if note is None:
             self.report_download_webpage(video_id)
         elif note is not False:
@@ -1888,13 +1904,16 @@ 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, m3u8_doc=None):
+        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=errnote or 'Failed to download m3u8 playlist information',
+                    errnote='Failed to download m3u8 playlist information',
-                    fatal=fatal, data=data, headers=headers, query=query)
+                    fatal=fatal, data=data, headers=headers)
 
                 if res is False:
                     return []
@@ -1928,7 +1947,7 @@ class InfoExtractor(object):
 
         if '#EXT-X-TARGETDURATION' in m3u8_doc:  # media playlist, return as is
 
-            playlist_formats = _extract_m3u8_playlist_formats(m3u8_doc, True)
+            playlist_formats = _extract_m3u8_playlist_formats(m3u8_doc=m3u8_doc)
 
             for format in playlist_formats:
                 format_id = []
@@ -1966,7 +1985,8 @@ class InfoExtractor(object):
             if media_url:
                 manifest_url = format_url(media_url)
                 format_id = []
-                playlist_formats = _extract_m3u8_playlist_formats(manifest_url)
+                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')
@@ -2027,7 +2047,8 @@ 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)
+                playlist_formats = _extract_m3u8_playlist_formats(manifest_url, video_id=video_id,
+                                                                  fatal=fatal, data=data, headers=headers)
 
                 for format in playlist_formats:
                     format_id = []

yt_dlp/extractor/dreisat.py

@@ -1,193 +1,43 @@
 from __future__ import unicode_literals
 
-import re
+from .zdf import ZDFIE
-
-from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    unified_strdate,
-    xpath_text,
-    determine_ext,
-    float_or_none,
-    ExtractorError,
-)
 
 
-class DreiSatIE(InfoExtractor):
+class DreiSatIE(ZDFIE):
     IE_NAME = '3sat'
-    _GEO_COUNTRIES = ['DE']
+    _VALID_URL = r'https?://(?:www\.)?3sat\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
-    _VALID_URL = r'https?://(?:www\.)?3sat\.de/mediathek/(?:(?:index|mediathek)\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)'
+    _TESTS = [{
-    _TESTS = [
+        # Same as https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html
-        {
+        'url': 'https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html',
-            'url': 'http://www.3sat.de/mediathek/index.php?mode=play&obj=45918',
+        'md5': '0aff3e7bc72c8813f5e0fae333316a1d',
-            'md5': 'be37228896d30a88f315b638900a026e',
+        'info_dict': {
-            'info_dict': {
+            'id': '141007_ab18_10wochensommer_film',
-                'id': '45918',
+            'ext': 'mp4',
-                'ext': 'mp4',
+            'title': 'Ab 18! - 10 Wochen Sommer',
-                'title': 'Waidmannsheil',
+            'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26',
-                'description': 'md5:cce00ca1d70e21425e72c86a98a56817',
+            'duration': 2660,
-                'uploader': 'SCHWEIZWEIT',
+            'timestamp': 1608604200,
-                'uploader_id': '100000210',
+            'upload_date': '20201222',
-                'upload_date': '20140913'
-            },
-            'params': {
-                'skip_download': True,  # m3u8 downloads
-            }
         },
-        {
+    }, {
-            'url': 'http://www.3sat.de/mediathek/mediathek.php?mode=play&obj=51066',
+        'url': 'https://www.3sat.de/gesellschaft/schweizweit/waidmannsheil-100.html',
-            'only_matching': True,
+        'info_dict': {
+            'id': '140913_sendung_schweizweit',
+            'ext': 'mp4',
+            'title': 'Waidmannsheil',
+            'description': 'md5:cce00ca1d70e21425e72c86a98a56817',
+            'timestamp': 1410623100,
+            'upload_date': '20140913'
         },
-    ]
+        'params': {
-
+            'skip_download': True,
-    def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
-        param_groups = {}
-        for param_group in smil.findall(self._xpath_ns('./head/paramGroup', namespace)):
-            group_id = param_group.get(self._xpath_ns(
-                'id', 'http://www.w3.org/XML/1998/namespace'))
-            params = {}
-            for param in param_group:
-                params[param.get('name')] = param.get('value')
-            param_groups[group_id] = params
-
-        formats = []
-        for video in smil.findall(self._xpath_ns('.//video', namespace)):
-            src = video.get('src')
-            if not src:
-                continue
-            bitrate = int_or_none(self._search_regex(r'_(\d+)k', src, 'bitrate', None)) or float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
-            group_id = video.get('paramGroup')
-            param_group = param_groups[group_id]
-            for proto in param_group['protocols'].split(','):
-                formats.append({
-                    'url': '%s://%s' % (proto, param_group['host']),
-                    'app': param_group['app'],
-                    'play_path': src,
-                    'ext': 'flv',
-                    'format_id': '%s-%d' % (proto, bitrate),
-                    'tbr': bitrate,
-                })
-        self._sort_formats(formats)
-        return formats
-
-    def extract_from_xml_url(self, video_id, xml_url):
-        doc = self._download_xml(
-            xml_url, video_id,
-            note='Downloading video info',
-            errnote='Failed to download video info')
-
-        status_code = xpath_text(doc, './status/statuscode')
-        if status_code and status_code != 'ok':
-            if status_code == 'notVisibleAnymore':
-                message = 'Video %s is not available' % video_id
-            else:
-                message = '%s returned error: %s' % (self.IE_NAME, status_code)
-            raise ExtractorError(message, expected=True)
-
-        title = xpath_text(doc, './/information/title', 'title', True)
-
-        urls = []
-        formats = []
-        for fnode in doc.findall('.//formitaeten/formitaet'):
-            video_url = xpath_text(fnode, 'url')
-            if not video_url or video_url in urls:
-                continue
-            urls.append(video_url)
-
-            is_available = 'http://www.metafilegenerator' not in video_url
-            geoloced = 'static_geoloced_online' in video_url
-            if not is_available or geoloced:
-                continue
-
-            format_id = fnode.attrib['basetype']
-            format_m = re.match(r'''(?x)
-                (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
-                (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
-            ''', format_id)
-
-            ext = determine_ext(video_url, None) or format_m.group('container')
-
-            if ext == 'meta':
-                continue
-            elif ext == 'smil':
-                formats.extend(self._extract_smil_formats(
-                    video_url, video_id, fatal=False))
-            elif ext == 'm3u8':
-                # the certificates are misconfigured (see
-                # https://github.com/ytdl-org/youtube-dl/issues/8665)
-                if video_url.startswith('https://'):
-                    continue
-                formats.extend(self._extract_m3u8_formats(
-                    video_url, video_id, 'mp4', 'm3u8_native',
-                    m3u8_id=format_id, fatal=False))
-            elif ext == 'f4m':
-                formats.extend(self._extract_f4m_formats(
-                    video_url, video_id, f4m_id=format_id, fatal=False))
-            else:
-                quality = xpath_text(fnode, './quality')
-                if quality:
-                    format_id += '-' + quality
-
-                abr = int_or_none(xpath_text(fnode, './audioBitrate'), 1000)
-                vbr = int_or_none(xpath_text(fnode, './videoBitrate'), 1000)
-
-                tbr = int_or_none(self._search_regex(
-                    r'_(\d+)k', video_url, 'bitrate', None))
-                if tbr and vbr and not abr:
-                    abr = tbr - vbr
-
-                formats.append({
-                    'format_id': format_id,
-                    'url': video_url,
-                    'ext': ext,
-                    'acodec': format_m.group('acodec'),
-                    'vcodec': format_m.group('vcodec'),
-                    'abr': abr,
-                    'vbr': vbr,
-                    'tbr': tbr,
-                    'width': int_or_none(xpath_text(fnode, './width')),
-                    'height': int_or_none(xpath_text(fnode, './height')),
-                    'filesize': int_or_none(xpath_text(fnode, './filesize')),
-                    'protocol': format_m.group('proto').lower(),
-                })
-
-        geolocation = xpath_text(doc, './/details/geolocation')
-        if not formats and geolocation and geolocation != 'none':
-            self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
-
-        self._sort_formats(formats)
-
-        thumbnails = []
-        for node in doc.findall('.//teaserimages/teaserimage'):
-            thumbnail_url = node.text
-            if not thumbnail_url:
-                continue
-            thumbnail = {
-                'url': thumbnail_url,
-            }
-            thumbnail_key = node.get('key')
-            if thumbnail_key:
-                m = re.match('^([0-9]+)x([0-9]+)$', thumbnail_key)
-                if m:
-                    thumbnail['width'] = int(m.group(1))
-                    thumbnail['height'] = int(m.group(2))
-            thumbnails.append(thumbnail)
-
-        upload_date = unified_strdate(xpath_text(doc, './/details/airtime'))
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': xpath_text(doc, './/information/detail'),
-            'duration': int_or_none(xpath_text(doc, './/details/lengthSec')),
-            'thumbnails': thumbnails,
-            'uploader': xpath_text(doc, './/details/originChannelTitle'),
-            'uploader_id': xpath_text(doc, './/details/originChannelId'),
-            'upload_date': upload_date,
-            'formats': formats,
         }
-
+    }, {
-    def _real_extract(self, url):
+        # Same as https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html
-        video_id = self._match_id(url)
+        'url': 'https://www.3sat.de/film/spielfilm/der-hauptmann-100.html',
-        details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?id=%s' % video_id
+        'only_matching': True,
-        return self.extract_from_xml_url(video_id, details_url)
+    }, {
+        # Same as https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
+        'url': 'https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html',
+        'only_matching': True,
+    }]

yt_dlp/extractor/extractors.py

@@ -103,6 +103,7 @@ from .awaan import (
 )
 from .azmedien import AZMedienIE
 from .baidu import BaiduVideoIE
+from .bandaichannel import BandaiChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE, BandcampWeeklyIE
 from .bbc import (
     BBCCoUkIE,
@@ -737,6 +738,7 @@ from .mtv import (
 )
 from .muenchentv import MuenchenTVIE
 from .mwave import MwaveIE, MwaveMeetGreetIE
+from .mxplayer import MxplayerIE
 from .mychannels import MyChannelsIE
 from .myspace import MySpaceIE, MySpaceAlbumIE
 from .myspass import MySpassIE

yt_dlp/extractor/hrfensehen.py

@@ -4,7 +4,7 @@ from __future__ import unicode_literals
 import json
 import re
 
-from yt_dlp.utils import int_or_none, unified_timestamp, unescapeHTML
+from ..utils import int_or_none, unified_timestamp, unescapeHTML
 from .common import InfoExtractor
 
 

yt_dlp/extractor/mtv.py

@@ -7,7 +7,6 @@ from .common import InfoExtractor
 from ..compat import (
     compat_str,
     compat_xpath,
-    compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
@@ -23,7 +22,6 @@ from ..utils import (
     unescapeHTML,
     update_url_query,
     url_basename,
-    get_domain,
     xpath_text,
 )
 
@@ -45,7 +43,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         # Remove the templates, like &device={device}
         return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url)
 
-    def _get_feed_url(self, uri, url=None):
+    def _get_feed_url(self, uri):
         return self._FEED_URL
 
     def _get_thumbnail_url(self, uri, itemdoc):
@@ -211,9 +209,9 @@ class MTVServicesInfoExtractor(InfoExtractor):
             data['lang'] = self._LANG
         return data
 
-    def _get_videos_info(self, uri, use_hls=True, url=None):
+    def _get_videos_info(self, uri, use_hls=True):
         video_id = self._id_from_uri(uri)
-        feed_url = self._get_feed_url(uri, url)
+        feed_url = self._get_feed_url(uri)
         info_url = update_url_query(feed_url, self._get_feed_query(uri))
         return self._get_videos_info_from_url(info_url, video_id, use_hls)
 
@@ -259,41 +257,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
     def _extract_child_with_type(parent, t):
         return next(c for c in parent['children'] if c.get('type') == t)
 
-    def _extract_new_triforce_mgid(self, webpage, url='', video_id=None):
+    def _extract_mgid(self, webpage):
-        if url == '':
-            return
-        domain = get_domain(url)
-        if domain is None:
-            raise ExtractorError(
-                '[%s] could not get domain' % self.IE_NAME,
-                expected=True)
-        url = url.replace("https://", "http://")
-        enc_url = compat_urlparse.quote(url, safe='')
-        _TRIFORCE_V8_TEMPLATE = 'https://%s/feeds/triforce/manifest/v8?url=%s'
-        triforce_manifest_url = _TRIFORCE_V8_TEMPLATE % (domain, enc_url)
-
-        manifest = self._download_json(triforce_manifest_url, video_id, fatal=False)
-        if manifest:
-            if manifest.get('manifest').get('type') == 'redirect':
-                self.to_screen('Found a redirect. Downloading manifest from new location')
-                new_loc = manifest.get('manifest').get('newLocation')
-                new_loc = new_loc.replace("https://", "http://")
-                enc_new_loc = compat_urlparse.quote(new_loc, safe='')
-                triforce_manifest_new_loc = _TRIFORCE_V8_TEMPLATE % (domain, enc_new_loc)
-                manifest = self._download_json(triforce_manifest_new_loc, video_id, fatal=False)
-
-        item_id = try_get(manifest, lambda x: x['manifest']['reporting']['itemId'], compat_str)
-        if not item_id:
-            self.to_screen('No id found!')
-            return
-
-        # 'episode' can be anything. 'content' is used often as well
-        _MGID_TEMPLATE = 'mgid:arc:episode:%s:%s'
-        mgid = _MGID_TEMPLATE % (domain, item_id)
-
-        return mgid
-
-    def _extract_mgid(self, webpage, url, title=None, data_zone=None):
         try:
             # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
             # or http://media.mtvnservices.com/{mgid}
@@ -304,21 +268,6 @@ class MTVServicesInfoExtractor(InfoExtractor):
         except RegexNotFoundError:
             mgid = None
 
-        if not title:
-            title = url_basename(url)
-
-        try:
-            window_data = self._parse_json(self._search_regex(
-                r'(?s)window.__DATA__ = (?P<json>{.+});', webpage,
-                'JSON Window Data', default=None, fatal=False, group='json'), title, fatal=False)
-            main_container = None
-            for i in range(len(window_data['children'])):
-                if window_data['children'][i]['type'] == 'MainContainer':
-                    main_container = window_data['children'][i]
-            mgid = main_container['children'][0]['props']['media']['video']['config']['uri']
-        except (KeyError, IndexError, TypeError):
-            pass
-
         if mgid is None or ':' not in mgid:
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject\.embedSWF\(".*?(mgid:.*?)"'],
@@ -331,10 +280,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
                 r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid', default=None)
 
         if not mgid:
-            mgid = self._extract_new_triforce_mgid(webpage, url)
+            mgid = self._extract_triforce_mgid(webpage)
-
-        if not mgid:
-            mgid = self._extract_triforce_mgid(webpage, data_zone)
 
         if not mgid:
             data = self._parse_json(self._search_regex(
@@ -348,8 +294,8 @@ class MTVServicesInfoExtractor(InfoExtractor):
     def _real_extract(self, url):
         title = url_basename(url)
         webpage = self._download_webpage(url, title)
-        mgid = self._extract_mgid(webpage, url, title=title)
+        mgid = self._extract_mgid(webpage)
-        videos_info = self._get_videos_info(mgid, url=url)
+        videos_info = self._get_videos_info(mgid)
         return videos_info
 
 

yt_dlp/extractor/mxplayer.py

@@ -0,0 +1,103 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    js_to_json,
+    url_or_none,
+    urljoin,
+)
+
+
+VALID_STREAMS = ('dash', )
+
+
+class MxplayerIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?mxplayer\.in/movie/(?P<slug>[a-z0-9]+(?:-[a-z0-9]+)*)'
+    _TEST = {
+        'url': 'https://www.mxplayer.in/movie/watch-knock-knock-hindi-dubbed-movie-online-b9fa28df3bfb8758874735bbd7d2655a?watch=true',
+        'info_dict': {
+            'id': 'b9fa28df3bfb8758874735bbd7d2655a',
+            'ext': 'mp4',
+            'title': 'Knock Knock Movie | Watch 2015 Knock Knock Full Movie Online- MX Player',
+            'description': 'md5:b195ba93ff1987309cfa58e2839d2a5b'
+        },
+        'params': {
+            'skip_download': True,
+            'format': 'bestvideo'
+        }
+    }
+
+    def _get_best_stream_url(self, stream):
+        best_stream = list(filter(None, [v for k, v in stream.items()]))
+        return best_stream.pop(0) if len(best_stream) else None
+
+    def _get_stream_urls(self, video_dict):
+        stream_dict = video_dict.get('stream', {'provider': {}})
+        stream_provider = stream_dict.get('provider')
+
+        if not stream_dict[stream_provider]:
+            message = 'No stream provider found'
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+
+        streams = []
+        for stream_name, v in stream_dict[stream_provider].items():
+            if stream_name in VALID_STREAMS:
+                stream_url = self._get_best_stream_url(v)
+                if stream_url is None:
+                    continue
+                streams.append((stream_name, stream_url))
+        return streams
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_slug = mobj.group('slug')
+
+        video_id = video_slug.split('-')[-1]
+
+        webpage = self._download_webpage(url, video_id)
+
+        window_state_json = self._html_search_regex(
+            r'(?s)<script>window\.state\s*[:=]\s(\{.+\})\n(\w+).*(</script>).*',
+            webpage, 'WindowState')
+
+        source = self._parse_json(js_to_json(window_state_json), video_id)
+        if not source:
+            raise ExtractorError('Cannot find source', expected=True)
+
+        config_dict = source['config']
+        video_dict = source['entities'][video_id]
+        stream_urls = self._get_stream_urls(video_dict)
+
+        title = self._og_search_title(webpage, fatal=True, default=video_dict['title'])
+
+        formats = []
+        headers = {'Referer': url}
+        for stream_name, stream_url in stream_urls:
+            if stream_name == 'dash':
+                format_url = url_or_none(urljoin(config_dict['videoCdnBaseUrl'], stream_url))
+                if not format_url:
+                    continue
+                formats.extend(self._extract_mpd_formats(
+                    format_url, video_id, mpd_id='dash', headers=headers))
+
+        self._sort_formats(formats)
+        info = {
+            'id': video_id,
+            'title': title,
+            'description': video_dict.get('description'),
+            'formats': formats
+        }
+
+        if video_dict.get('imageInfo'):
+            info['thumbnails'] = list(map(lambda i: dict(i, **{
+                'url': urljoin(config_dict['imageBaseUrl'], i['url'])
+            }), video_dict['imageInfo']))
+
+        if video_dict.get('webUrl'):
+            last_part = video_dict['webUrl'].split("/")[-1]
+            info['display_id'] = last_part.replace(video_id, "").rstrip("-")
+
+        return info

yt_dlp/extractor/nick.py

@@ -8,59 +8,66 @@ from ..utils import update_url_query
 
 
 class NickIE(MTVServicesInfoExtractor):
-    # None of videos on the website are still alive?
     IE_NAME = 'nick.com'
-    _VALID_URL = r'https?://(?P<domain>(?:(?:www|beta)\.)?nick(?:jr)?\.com)/(?:[^/]+/)?(?:videos/clip|[^/]+/videos)/(?P<id>[^/?#.]+)'
+    _VALID_URL = r'https?://(?P<domain>(?:www\.)?nick(?:jr)?\.com)/(?:[^/]+/)?(?P<type>videos/clip|[^/]+/videos|episodes/[^/]+)/(?P<id>[^/?#.]+)'
     _FEED_URL = 'http://udat.mtvnservices.com/service1/dispatch.htm'
     _GEO_COUNTRIES = ['US']
     _TESTS = [{
-        'url': 'http://www.nick.com/videos/clip/alvinnn-and-the-chipmunks-112-full-episode.html',
+        'url': 'https://www.nick.com/episodes/sq47rw/spongebob-squarepants-a-place-for-pets-lockdown-for-love-season-13-ep-1',
+        'info_dict': {
+            'description': 'md5:0650a9eb88955609d5c1d1c79292e234',
+            'title': 'A Place for Pets/Lockdown for Love',
+        },
         'playlist': [
             {
-                'md5': '6e5adc1e28253bbb1b28ab05403dd4d4',
+                'md5': 'cb8a2afeafb7ae154aca5a64815ec9d6',
                 'info_dict': {
-                    'id': 'be6a17b0-412d-11e5-8ff7-0026b9414f30',
+                    'id': '85ee8177-d6ce-48f8-9eee-a65364f8a6df',
                     'ext': 'mp4',
-                    'title': 'ALVINNN!!! and The Chipmunks: "Mojo Missing/Who\'s The Animal" S1',
+                    'title': 'SpongeBob SquarePants: "A Place for Pets/Lockdown for Love" S1',
-                    'description': 'Alvin is convinced his mojo was in a cap he gave to a fan, and must find a way to get his hat back before the Chipmunks’ big concert.\nDuring a costume visit to the zoo, Alvin finds himself mistaken for the real Tasmanian devil.',
+                    'description': 'A Place for Pets/Lockdown for Love: When customers bring pets into the Krusty Krab, Mr. Krabs realizes pets are more profitable than owners. Plankton ruins another date with Karen, so she puts the Chum Bucket on lockdown until he proves his affection.',
 
                 }
             },
             {
-                'md5': 'd7be441fc53a1d4882fa9508a1e5b3ce',
+                'md5': '839a04f49900a1fcbf517020d94e0737',
                 'info_dict': {
-                    'id': 'be6b8f96-412d-11e5-8ff7-0026b9414f30',
+                    'id': '2e2a9960-8fd4-411d-868b-28eb1beb7fae',
                     'ext': 'mp4',
-                    'title': 'ALVINNN!!! and The Chipmunks: "Mojo Missing/Who\'s The Animal" S2',
+                    'title': 'SpongeBob SquarePants: "A Place for Pets/Lockdown for Love" S2',
-                    'description': 'Alvin is convinced his mojo was in a cap he gave to a fan, and must find a way to get his hat back before the Chipmunks’ big concert.\nDuring a costume visit to the zoo, Alvin finds himself mistaken for the real Tasmanian devil.',
+                    'description': 'A Place for Pets/Lockdown for Love: When customers bring pets into the Krusty Krab, Mr. Krabs realizes pets are more profitable than owners. Plankton ruins another date with Karen, so she puts the Chum Bucket on lockdown until he proves his affection.',
 
                 }
             },
             {
-                'md5': 'efffe1728a234b2b0d2f2b343dd1946f',
+                'md5': 'f1145699f199770e2919ee8646955d46',
                 'info_dict': {
-                    'id': 'be6cf7e6-412d-11e5-8ff7-0026b9414f30',
+                    'id': 'dc91c304-6876-40f7-84a6-7aece7baa9d0',
                     'ext': 'mp4',
-                    'title': 'ALVINNN!!! and The Chipmunks: "Mojo Missing/Who\'s The Animal" S3',
+                    'title': 'SpongeBob SquarePants: "A Place for Pets/Lockdown for Love" S3',
-                    'description': 'Alvin is convinced his mojo was in a cap he gave to a fan, and must find a way to get his hat back before the Chipmunks’ big concert.\nDuring a costume visit to the zoo, Alvin finds himself mistaken for the real Tasmanian devil.',
+                    'description': 'A Place for Pets/Lockdown for Love: When customers bring pets into the Krusty Krab, Mr. Krabs realizes pets are more profitable than owners. Plankton ruins another date with Karen, so she puts the Chum Bucket on lockdown until he proves his affection.',
+
                 }
             },
             {
-                'md5': '1ec6690733ab9f41709e274a1d5c7556',
+                'md5': 'd463116875aee2585ee58de3b12caebd',
                 'info_dict': {
-                    'id': 'be6e3354-412d-11e5-8ff7-0026b9414f30',
+                    'id': '5d929486-cf4c-42a1-889a-6e0d183a101a',
                     'ext': 'mp4',
-                    'title': 'ALVINNN!!! and The Chipmunks: "Mojo Missing/Who\'s The Animal" S4',
+                    'title': 'SpongeBob SquarePants: "A Place for Pets/Lockdown for Love" S4',
-                    'description': 'Alvin is convinced his mojo was in a cap he gave to a fan, and must find a way to get his hat back before the Chipmunks’ big concert.\nDuring a costume visit to the zoo, Alvin finds himself mistaken for the real Tasmanian devil.',
+                    'description': 'A Place for Pets/Lockdown for Love: When customers bring pets into the Krusty Krab, Mr. Krabs realizes pets are more profitable than owners. Plankton ruins another date with Karen, so she puts the Chum Bucket on lockdown until he proves his affection.',
+
                 }
             },
         ],
     }, {
-        'url': 'http://www.nickjr.com/paw-patrol/videos/pups-save-a-goldrush-s3-ep302-full-episode/',
+        'url': 'http://www.nickjr.com/blues-clues-and-you/videos/blues-clues-and-you-original-209-imagination-station/',
-        'only_matching': True,
+        'info_dict': {
-    }, {
+            'id': '31631529-2fc5-430b-b2ef-6a74b4609abd',
-        'url': 'http://beta.nick.com/nicky-ricky-dicky-and-dawn/videos/nicky-ricky-dicky-dawn-301-full-episode/',
+            'ext': 'mp4',
-        'only_matching': True,
+            'description': 'md5:9d65a66df38e02254852794b2809d1cf',
+            'title': 'Blue\'s Imagination Station',
+        },
     }]
 
     def _get_feed_query(self, uri):
@@ -69,8 +76,14 @@ class NickIE(MTVServicesInfoExtractor):
             'mgid': uri,
         }
 
+    def _extract_mgid(self, webpage):
+        mgid = self._search_regex(r'"media":{"video":{"config":{"uri":"(mgid:.*?)"', webpage, 'mgid', default=None)
+        return mgid
+
     def _real_extract(self, url):
-        domain, display_id = re.match(self._VALID_URL, url).groups()
+        domain, video_type, display_id = re.match(self._VALID_URL, url).groups()
+        if video_type.startswith("episodes"):
+            return super()._real_extract(url)
         video_data = self._download_json(
             'http://%s/data/video.endLevel.json' % domain,
             display_id, query={

yt_dlp/extractor/ninecninemedia.py

@@ -23,11 +23,9 @@ class NineCNineMediaIE(InfoExtractor):
         destination_code, content_id = re.match(self._VALID_URL, url).groups()
         api_base_url = self._API_BASE_TEMPLATE % (destination_code, content_id)
         content = self._download_json(api_base_url, content_id, query={
-            '$include': '[Media,Season,ContentPackages]',
+            '$include': '[Media.Name,Season,ContentPackages.Duration,ContentPackages.Id]',
         })
         title = content['Name']
-        if len(content['ContentPackages']) > 1:
-            raise ExtractorError('multiple content packages')
         content_package = content['ContentPackages'][0]
         package_id = content_package['Id']
         content_package_url = api_base_url + 'contentpackages/%s/' % package_id

yt_dlp/extractor/phoenix.py

@@ -1,52 +1,128 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
+import re
-from ..utils import ExtractorError
+
+from .youtube import YoutubeIE
+from .zdf import ZDFBaseIE
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    merge_dicts,
+    unified_timestamp,
+    xpath_text,
+)
 
 
-class PhoenixIE(InfoExtractor):
+class PhoenixIE(ZDFBaseIE):
     IE_NAME = 'phoenix.de'
-    _VALID_URL = r'''https?://(?:www\.)?phoenix.de/\D+(?P<id>\d+)\.html'''
+    _VALID_URL = r'https?://(?:www\.)?phoenix\.de/(?:[^/]+/)*[^/?#&]*-a-(?P<id>\d+)\.html'
-    _TESTS = [
+    _TESTS = [{
-        {
+        # Same as https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html
-            'url': 'https://www.phoenix.de/sendungen/dokumentationen/unsere-welt-in-zukunft---stadt-a-1283620.html',
+        'url': 'https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html',
-            'md5': '5e765e838aa3531c745a4f5b249ee3e3',
+        'md5': '34ec321e7eb34231fd88616c65c92db0',
-            'info_dict': {
+        'info_dict': {
-                'id': '0OB4HFc43Ns',
+            'id': '210222_phx_nachgehakt_corona_protest',
-                'ext': 'mp4',
+            'ext': 'mp4',
-                'title': 'Unsere Welt in Zukunft - Stadt',
+            'title': 'Wohin führt der Protest in der Pandemie?',
-                'description': 'md5:9bfb6fd498814538f953b2dcad7ce044',
+            'description': 'md5:7d643fe7f565e53a24aac036b2122fbd',
-                'upload_date': '20190912',
+            'duration': 1691,
-                'uploader': 'phoenix',
+            'timestamp': 1613906100,
-                'uploader_id': 'phoenix',
+            'upload_date': '20210221',
-            }
+            'uploader': 'Phoenix',
+            'channel': 'corona nachgehakt',
         },
-        {
+    }, {
-            'url': 'https://www.phoenix.de/drohnenangriffe-in-saudi-arabien-a-1286995.html?ref=aktuelles',
+        # Youtube embed
-            'only_matching': True,
+        'url': 'https://www.phoenix.de/sendungen/gespraeche/phoenix-streitgut-brennglas-corona-a-1965505.html',
+        'info_dict': {
+            'id': 'hMQtqFYjomk',
+            'ext': 'mp4',
+            'title': 'phoenix streitgut: Brennglas Corona - Wie gerecht ist unsere Gesellschaft?',
+            'description': 'md5:ac7a02e2eb3cb17600bc372e4ab28fdd',
+            'duration': 3509,
+            'upload_date': '20201219',
+            'uploader': 'phoenix',
+            'uploader_id': 'phoenix',
         },
-        # an older page: https://www.phoenix.de/sendungen/gespraeche/phoenix-persoenlich/im-dialog-a-177727.html
+        'params': {
-        # seems to not have an embedded video, even though it's uploaded on youtube: https://www.youtube.com/watch?v=4GxnoUHvOkM
+            'skip_download': True,
-    ]
+        },
-
+    }, {
-    def extract_from_json_api(self, video_id, api_url):
+        'url': 'https://www.phoenix.de/entwicklungen-in-russland-a-2044720.html',
-        doc = self._download_json(
+        'only_matching': True,
-            api_url, video_id,
+    }, {
-            note="Downloading webpage metadata",
+        # no media
-            errnote="Failed to load webpage metadata")
+        'url': 'https://www.phoenix.de/sendungen/dokumentationen/mit-dem-jumbo-durch-die-nacht-a-89625.html',
-
+        'only_matching': True,
-        for a in doc["absaetze"]:
+    }, {
-            if a["typ"] == "video-youtube":
+        # Same as https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html
-                return {
+        'url': 'https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche',
-                    '_type': 'url_transparent',
+        'only_matching': True,
-                    'id': a["id"],
+    }]
-                    'title': doc["titel"],
-                    'url': "https://www.youtube.com/watch?v=%s" % a["id"],
-                    'ie_key': 'Youtube',
-                }
-        raise ExtractorError("No downloadable video found", expected=True)
 
     def _real_extract(self, url):
-        page_id = self._match_id(url)
+        article_id = self._match_id(url)
-        api_url = 'https://www.phoenix.de/response/id/%s' % page_id
+
-        return self.extract_from_json_api(page_id, api_url)
+        article = self._download_json(
+            'https://www.phoenix.de/response/id/%s' % article_id, article_id,
+            'Downloading article JSON')
+
+        video = article['absaetze'][0]
+        title = video.get('titel') or article.get('subtitel')
+
+        if video.get('typ') == 'video-youtube':
+            video_id = video['id']
+            return self.url_result(
+                video_id, ie=YoutubeIE.ie_key(), video_id=video_id,
+                video_title=title)
+
+        video_id = compat_str(video.get('basename') or video.get('content'))
+
+        details = self._download_xml(
+            'https://www.phoenix.de/php/mediaplayer/data/beitrags_details.php',
+            video_id, 'Downloading details XML', query={
+                'ak': 'web',
+                'ptmd': 'true',
+                'id': video_id,
+                'profile': 'player2',
+            })
+
+        title = title or xpath_text(
+            details, './/information/title', 'title', fatal=True)
+        content_id = xpath_text(
+            details, './/video/details/basename', 'content id', fatal=True)
+
+        info = self._extract_ptmd(
+            'https://tmd.phoenix.de/tmd/2/ngplayer_2_3/vod/ptmd/phoenix/%s' % content_id,
+            content_id, None, url)
+
+        timestamp = unified_timestamp(xpath_text(details, './/details/airtime'))
+
+        thumbnails = []
+        for node in details.findall('.//teaserimages/teaserimage'):
+            thumbnail_url = node.text
+            if not thumbnail_url:
+                continue
+            thumbnail = {
+                'url': thumbnail_url,
+            }
+            thumbnail_key = node.get('key')
+            if thumbnail_key:
+                m = re.match('^([0-9]+)x([0-9]+)$', thumbnail_key)
+                if m:
+                    thumbnail['width'] = int(m.group(1))
+                    thumbnail['height'] = int(m.group(2))
+            thumbnails.append(thumbnail)
+
+        return merge_dicts(info, {
+            'id': content_id,
+            'title': title,
+            'description': xpath_text(details, './/information/detail'),
+            'duration': int_or_none(xpath_text(details, './/details/lengthSec')),
+            'thumbnails': thumbnails,
+            'timestamp': timestamp,
+            'uploader': xpath_text(details, './/details/channel'),
+            'uploader_id': xpath_text(details, './/details/originChannelId'),
+            'channel': xpath_text(details, './/details/originChannelTitle'),
+        })

yt_dlp/extractor/rds.py

@@ -15,17 +15,17 @@ class RDSIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?rds\.ca/vid(?:[eé]|%C3%A9)os/(?:[^/]+/)*(?P<id>[^/]+)-\d+\.\d+'
 
     _TESTS = [{
-        'url': 'http://www.rds.ca/videos/football/nfl/fowler-jr-prend-la-direction-de-jacksonville-3.1132799',
+        # has two 9c9media ContentPackages, the web player selects the first ContentPackage
+        'url': 'https://www.rds.ca/videos/Hockey/NationalHockeyLeague/teams/9/forum-du-5-a-7-jesperi-kotkaniemi-de-retour-de-finlande-3.1377606',
         'info_dict': {
-            'id': '604333',
+            'id': '2083309',
-            'display_id': 'fowler-jr-prend-la-direction-de-jacksonville',
+            'display_id': 'forum-du-5-a-7-jesperi-kotkaniemi-de-retour-de-finlande',
             'ext': 'flv',
-            'title': 'Fowler Jr. prend la direction de Jacksonville',
+            'title': 'Forum du 5 à 7 : Kotkaniemi de retour de Finlande',
-            'description': 'Dante Fowler Jr. est le troisième choix du repêchage 2015 de la NFL. ',
+            'description': 'md5:83fa38ecc4a79b19e433433254077f25',
-            'timestamp': 1430397346,
+            'timestamp': 1606129030,
-            'upload_date': '20150430',
+            'upload_date': '20201123',
-            'duration': 154.354,
+            'duration': 773.039,
-            'age_limit': 0,
         }
     }, {
         'url': 'http://www.rds.ca/vid%C3%A9os/un-voyage-positif-3.877934',

yt_dlp/extractor/rts.py

@@ -6,11 +6,12 @@ import re
 from .srgssr import SRGSSRIE
 from ..compat import compat_str
 from ..utils import (
+    determine_ext,
     int_or_none,
     parse_duration,
     parse_iso8601,
     unescapeHTML,
-    determine_ext,
+    urljoin,
 )
 
 
@@ -21,7 +22,7 @@ class RTSIE(SRGSSRIE):
     _TESTS = [
         {
             'url': 'http://www.rts.ch/archives/tv/divers/3449373-les-enfants-terribles.html',
-            'md5': 'ff7f8450a90cf58dacb64e29707b4a8e',
+            'md5': '753b877968ad8afaeddccc374d4256a5',
             'info_dict': {
                 'id': '3449373',
                 'display_id': 'les-enfants-terribles',
@@ -35,6 +36,7 @@ class RTSIE(SRGSSRIE):
                 'thumbnail': r're:^https?://.*\.image',
                 'view_count': int,
             },
+            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
         },
         {
             'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html',
@@ -63,11 +65,12 @@ class RTSIE(SRGSSRIE):
                 # m3u8 download
                 'skip_download': True,
             },
+            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
             'skip': 'Blocked outside Switzerland',
         },
         {
             'url': 'http://www.rts.ch/video/info/journal-continu/5745356-londres-cachee-par-un-epais-smog.html',
-            'md5': '1bae984fe7b1f78e94abc74e802ed99f',
+            'md5': '9bb06503773c07ce83d3cbd793cebb91',
             'info_dict': {
                 'id': '5745356',
                 'display_id': 'londres-cachee-par-un-epais-smog',
@@ -81,6 +84,7 @@ class RTSIE(SRGSSRIE):
                 'thumbnail': r're:^https?://.*\.image',
                 'view_count': int,
             },
+            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
         },
         {
             'url': 'http://www.rts.ch/audio/couleur3/programmes/la-belle-video-de-stephane-laurenceau/5706148-urban-hippie-de-damien-krisl-03-04-2014.html',
@@ -160,7 +164,7 @@ class RTSIE(SRGSSRIE):
         media_type = 'video' if 'video' in all_info else 'audio'
 
         # check for errors
-        self.get_media_data('rts', media_type, media_id)
+        self._get_media_data('rts', media_type, media_id)
 
         info = all_info['video']['JSONinfo'] if 'video' in all_info else all_info['audio']
 
@@ -194,6 +198,7 @@ class RTSIE(SRGSSRIE):
                     'tbr': extract_bitrate(format_url),
                 })
 
+        download_base = 'http://rtsww%s-d.rts.ch/' % ('-a' if media_type == 'audio' else '')
         for media in info.get('media', []):
             media_url = media.get('url')
             if not media_url or re.match(r'https?://', media_url):
@@ -205,7 +210,7 @@ class RTSIE(SRGSSRIE):
                 format_id += '-%dk' % rate
             formats.append({
                 'format_id': format_id,
-                'url': 'http://download-video.rts.ch/' + media_url,
+                'url': urljoin(download_base, media_url),
                 'tbr': rate or extract_bitrate(media_url),
             })
 

yt_dlp/extractor/srgssr.py

@@ -4,16 +4,32 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlparse
 from ..utils import (
     ExtractorError,
+    float_or_none,
+    int_or_none,
     parse_iso8601,
     qualities,
+    try_get,
 )
 
 
 class SRGSSRIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn|srgssr):(?P<bu>srf|rts|rsi|rtr|swi):(?:[^:]+:)?(?P<type>video|audio):(?P<id>[0-9a-f\-]{36}|\d+)'
+    _VALID_URL = r'''(?x)
+                    (?:
+                        https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn|
+                        srgssr
+                    ):
+                    (?P<bu>
+                        srf|rts|rsi|rtr|swi
+                    ):(?:[^:]+:)?
+                    (?P<type>
+                        video|audio
+                    ):
+                    (?P<id>
+                        [0-9a-f\-]{36}|\d+
+                    )
+                    '''
     _GEO_BYPASS = False
     _GEO_COUNTRIES = ['CH']
 
@@ -25,25 +41,39 @@ class SRGSSRIE(InfoExtractor):
         'LEGAL': 'The video cannot be transmitted for legal reasons.',
         'STARTDATE': 'This video is not yet available. Please try again later.',
     }
+    _DEFAULT_LANGUAGE_CODES = {
+        'srf': 'de',
+        'rts': 'fr',
+        'rsi': 'it',
+        'rtr': 'rm',
+        'swi': 'en',
+    }
 
     def _get_tokenized_src(self, url, video_id, format_id):
-        sp = compat_urllib_parse_urlparse(url).path.split('/')
         token = self._download_json(
-            'http://tp.srgssr.ch/akahd/token?acl=/%s/%s/*' % (sp[1], sp[2]),
+            'http://tp.srgssr.ch/akahd/token?acl=*',
             video_id, 'Downloading %s token' % format_id, fatal=False) or {}
-        auth_params = token.get('token', {}).get('authparams')
+        auth_params = try_get(token, lambda x: x['token']['authparams'])
         if auth_params:
-            url += '?' + auth_params
+            url += ('?' if '?' not in url else '&') + auth_params
         return url
 
-    def get_media_data(self, bu, media_type, media_id):
+    def _get_media_data(self, bu, media_type, media_id):
-        media_data = self._download_json(
+        query = {'onlyChapters': True} if media_type == 'video' else {}
-            'http://il.srgssr.ch/integrationlayer/1.0/ue/%s/%s/play/%s.json' % (bu, media_type, media_id),
+        full_media_data = self._download_json(
-            media_id)[media_type.capitalize()]
+            'https://il.srgssr.ch/integrationlayer/2.0/%s/mediaComposition/%s/%s.json'
+            % (bu, media_type, media_id),
+            media_id, query=query)['chapterList']
+        try:
+            media_data = next(
+                x for x in full_media_data if x.get('id') == media_id)
+        except StopIteration:
+            raise ExtractorError('No media information found')
 
-        if media_data.get('block') and media_data['block'] in self._ERRORS:
+        block_reason = media_data.get('blockReason')
-            message = self._ERRORS[media_data['block']]
+        if block_reason and block_reason in self._ERRORS:
-            if media_data['block'] == 'GEOBLOCK':
+            message = self._ERRORS[block_reason]
+            if block_reason == 'GEOBLOCK':
                 self.raise_geo_restricted(
                     msg=message, countries=self._GEO_COUNTRIES)
             raise ExtractorError(
@@ -53,53 +83,75 @@ class SRGSSRIE(InfoExtractor):
 
     def _real_extract(self, url):
         bu, media_type, media_id = re.match(self._VALID_URL, url).groups()
+        media_data = self._get_media_data(bu, media_type, media_id)
+        title = media_data['title']
 
-        media_data = self.get_media_data(bu, media_type, media_id)
-
-        metadata = media_data['AssetMetadatas']['AssetMetadata'][0]
-        title = metadata['title']
-        description = metadata.get('description')
-        created_date = media_data.get('createdDate') or metadata.get('createdDate')
-        timestamp = parse_iso8601(created_date)
-
-        thumbnails = [{
-            'id': image.get('id'),
-            'url': image['url'],
-        } for image in media_data.get('Image', {}).get('ImageRepresentations', {}).get('ImageRepresentation', [])]
-
-        preference = qualities(['LQ', 'MQ', 'SD', 'HQ', 'HD'])
         formats = []
-        for source in media_data.get('Playlists', {}).get('Playlist', []) + media_data.get('Downloads', {}).get('Download', []):
+        q = qualities(['SD', 'HD'])
-            protocol = source.get('@protocol')
+        for source in (media_data.get('resourceList') or []):
-            for asset in source['url']:
+            format_url = source.get('url')
-                asset_url = asset['text']
+            if not format_url:
-                quality = asset['@quality']
+                continue
-                format_id = '%s-%s' % (protocol, quality)
+            protocol = source.get('protocol')
-                if protocol.startswith('HTTP-HDS') or protocol.startswith('HTTP-HLS'):
+            quality = source.get('quality')
-                    asset_url = self._get_tokenized_src(asset_url, media_id, format_id)
+            format_id = []
-                    if protocol.startswith('HTTP-HDS'):
+            for e in (protocol, source.get('encoding'), quality):
-                        formats.extend(self._extract_f4m_formats(
+                if e:
-                            asset_url + ('?' if '?' not in asset_url else '&') + 'hdcore=3.4.0',
+                    format_id.append(e)
-                            media_id, f4m_id=format_id, fatal=False))
+            format_id = '-'.join(format_id)
-                    elif protocol.startswith('HTTP-HLS'):
+
-                        formats.extend(self._extract_m3u8_formats(
+            if protocol in ('HDS', 'HLS'):
-                            asset_url, media_id, 'mp4', 'm3u8_native',
+                if source.get('tokenType') == 'AKAMAI':
-                            m3u8_id=format_id, fatal=False))
+                    format_url = self._get_tokenized_src(
-                else:
+                        format_url, media_id, format_id)
-                    formats.append({
+                    formats.extend(self._extract_akamai_formats(
-                        'format_id': format_id,
+                        format_url, media_id))
-                        'url': asset_url,
+                elif protocol == 'HLS':
-                        'quality': preference(quality),
+                    formats.extend(self._extract_m3u8_formats(
-                        'ext': 'flv' if protocol == 'RTMP' else None,
+                        format_url, media_id, 'mp4', 'm3u8_native',
-                    })
+                        m3u8_id=format_id, fatal=False))
+            elif protocol in ('HTTP', 'HTTPS'):
+                formats.append({
+                    'format_id': format_id,
+                    'url': format_url,
+                    'quality': q(quality),
+                })
+
+        # This is needed because for audio medias the podcast url is usually
+        # always included, even if is only an audio segment and not the
+        # whole episode.
+        if int_or_none(media_data.get('position')) == 0:
+            for p in ('S', 'H'):
+                podcast_url = media_data.get('podcast%sdUrl' % p)
+                if not podcast_url:
+                    continue
+                quality = p + 'D'
+                formats.append({
+                    'format_id': 'PODCAST-' + quality,
+                    'url': podcast_url,
+                    'quality': q(quality),
+                })
         self._sort_formats(formats)
 
+        subtitles = {}
+        if media_type == 'video':
+            for sub in (media_data.get('subtitleList') or []):
+                sub_url = sub.get('url')
+                if not sub_url:
+                    continue
+                lang = sub.get('locale') or self._DEFAULT_LANGUAGE_CODES[bu]
+                subtitles.setdefault(lang, []).append({
+                    'url': sub_url,
+                })
+
         return {
             'id': media_id,
             'title': title,
-            'description': description,
+            'description': media_data.get('description'),
-            'timestamp': timestamp,
+            'timestamp': parse_iso8601(media_data.get('date')),
-            'thumbnails': thumbnails,
+            'thumbnail': media_data.get('imageUrl'),
+            'duration': float_or_none(media_data.get('duration'), 1000),
+            'subtitles': subtitles,
             'formats': formats,
         }
 
@@ -119,26 +171,17 @@ class SRGSSRPlayIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
-        'md5': 'da6b5b3ac9fa4761a942331cef20fcb3',
+        'md5': '6db2226ba97f62ad42ce09783680046c',
         'info_dict': {
             'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
             'ext': 'mp4',
             'upload_date': '20130701',
             'title': 'Snowden beantragt Asyl in Russland',
-            'timestamp': 1372713995,
+            'timestamp': 1372708215,
-        }
+            'duration': 113.827,
-    }, {
+            'thumbnail': r're:^https?://.*1383719781\.png$',
-        # No Speichern (Save) button
-        'url': 'http://www.srf.ch/play/tv/top-gear/video/jaguar-xk120-shadow-und-tornado-dampflokomotive?id=677f5829-e473-4823-ac83-a1087fe97faa',
-        'md5': '0a274ce38fda48c53c01890651985bc6',
-        'info_dict': {
-            'id': '677f5829-e473-4823-ac83-a1087fe97faa',
-            'ext': 'flv',
-            'upload_date': '20130710',
-            'title': 'Jaguar XK120, Shadow und Tornado-Dampflokomotive',
-            'description': 'md5:88604432b60d5a38787f152dec89cd56',
-            'timestamp': 1373493600,
         },
+        'expected_warnings': ['Unable to download f4m manifest'],
     }, {
         'url': 'http://www.rtr.ch/play/radio/actualitad/audio/saira-tujetsch-tuttina-cuntinuar-cun-sedrun-muster-turissem?id=63cb0778-27f8-49af-9284-8c7a8c6d15fc',
         'info_dict': {
@@ -146,7 +189,8 @@ class SRGSSRPlayIE(InfoExtractor):
             'ext': 'mp3',
             'upload_date': '20151013',
             'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem',
-            'timestamp': 1444750398,
+            'timestamp': 1444709160,
+            'duration': 336.816,
         },
         'params': {
             # rtmp download
@@ -159,19 +203,32 @@ class SRGSSRPlayIE(InfoExtractor):
             'id': '6348260',
             'display_id': '6348260',
             'ext': 'mp4',
-            'duration': 1796,
+            'duration': 1796.76,
             'title': 'Le 19h30',
-            'description': '',
-            'uploader': '19h30',
             'upload_date': '20141201',
             'timestamp': 1417458600,
             'thumbnail': r're:^https?://.*\.image',
-            'view_count': int,
         },
         'params': {
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        'url': 'http://play.swissinfo.ch/play/tv/business/video/why-people-were-against-tax-reforms?id=42960270',
+        'info_dict': {
+            'id': '42960270',
+            'ext': 'mp4',
+            'title': 'Why people were against tax reforms',
+            'description': 'md5:7ac442c558e9630e947427469c4b824d',
+            'duration': 94.0,
+            'upload_date': '20170215',
+            'timestamp': 1487173560,
+            'thumbnail': r're:https?://www\.swissinfo\.ch/srgscalableimage/42961964',
+            'subtitles': 'count:9',
+        },
+        'params': {
+            'skip_download': True,
+        }
     }, {
         'url': 'https://www.srf.ch/play/tv/popupvideoplayer?id=c4dba0ca-e75b-43b2-a34f-f708a4932e01',
         'only_matching': True,
@@ -181,6 +238,10 @@ class SRGSSRPlayIE(InfoExtractor):
     }, {
         'url': 'https://www.rts.ch/play/tv/19h30/video/le-19h30?urn=urn:rts:video:6348260',
         'only_matching': True,
+    }, {
+        # audio segment, has podcastSdUrl of the full episode
+        'url': 'https://www.srf.ch/play/radio/popupaudioplayer?id=50b20dc8-f05b-4972-bf03-e438ff2833eb',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -188,5 +249,4 @@ class SRGSSRPlayIE(InfoExtractor):
         bu = mobj.group('bu')
         media_type = mobj.group('type') or mobj.group('type_2')
         media_id = mobj.group('id')
-        # other info can be extracted from url + '&layout=json'
         return self.url_result('srgssr:%s:%s:%s' % (bu[:3], media_type, media_id), 'SRGSSR')

yt_dlp/extractor/stretchinternet.py

@@ -1,7 +1,6 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import int_or_none
 
 
 class StretchInternetIE(InfoExtractor):
@@ -11,22 +10,28 @@ class StretchInternetIE(InfoExtractor):
         'info_dict': {
             'id': '573272',
             'ext': 'mp4',
-            'title': 'University of Mary Wrestling vs. Upper Iowa',
+            'title': 'UNIVERSITY OF MARY WRESTLING VS UPPER IOWA',
-            'timestamp': 1575668361,
+            # 'timestamp': 1575668361,
-            'upload_date': '20191206',
+            # 'upload_date': '20191206',
+            'uploader_id': '99997',
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
+        media_url = self._download_json(
+            'https://core.stretchlive.com/trinity/event/tcg/' + video_id,
+            video_id)[0]['media'][0]['url']
         event = self._download_json(
-            'https://api.stretchinternet.com/trinity/event/tcg/' + video_id,
+            'https://neo-client.stretchinternet.com/portal-ws/getEvent.json',
-            video_id)[0]
+            video_id, query={'eventID': video_id, 'token': 'asdf'})['event']
 
         return {
             'id': video_id,
             'title': event['title'],
-            'timestamp': int_or_none(event.get('dateCreated'), 1000),
+            # TODO: parse US timezone abbreviations
-            'url': 'https://' + event['media'][0]['url'],
+            # 'timestamp': event.get('dateTimeString'),
+            'url': 'https://' + media_url,
+            'uploader_id': event.get('ownerID'),
         }

yt_dlp/extractor/tennistv.py

@@ -86,6 +86,7 @@ class TennisTVIE(InfoExtractor):
             'https://www.tennistv.com/api/users/v1/entitlementchecknondiva',
             video_id, note='Checking video authorization', headers=headers, data=check_json)
         formats = self._extract_m3u8_formats(check_result['contentUrl'], video_id, ext='mp4')
+        self._sort_formats(formats)
 
         vdata = self._download_json(
             'https://www.tennistv.com/api/en/v2/none/common/video/%s' % video_id,

yt_dlp/extractor/urplay.py

@@ -21,6 +21,11 @@ class URPlayIE(InfoExtractor):
             'description': 'md5:5344508a52aa78c1ced6c1b8b9e44e9a',
             'timestamp': 1513292400,
             'upload_date': '20171214',
+            'series': 'UR Samtiden - Livet, universum och rymdens märkliga musik',
+            'duration': 2269,
+            'categories': ['Kultur & historia'],
+            'tags': ['Kritiskt tänkande', 'Vetenskap', 'Vetenskaplig verksamhet'],
+            'episode': 'Om vetenskap, kritiskt tänkande och motstånd',
         },
     }, {
         'url': 'https://urskola.se/Produkter/190031-Tripp-Trapp-Trad-Sovkudde',
@@ -31,6 +36,10 @@ class URPlayIE(InfoExtractor):
             'description': 'md5:b86bffdae04a7e9379d1d7e5947df1d1',
             'timestamp': 1440086400,
             'upload_date': '20150820',
+            'series': 'Tripp, Trapp, Träd',
+            'duration': 865,
+            'tags': ['Sova'],
+            'episode': 'Sovkudde',
         },
     }, {
         'url': 'http://urskola.se/Produkter/155794-Smasagor-meankieli-Grodan-i-vida-varlden',
@@ -41,9 +50,11 @@ class URPlayIE(InfoExtractor):
         video_id = self._match_id(url)
         url = url.replace('skola.se/Produkter', 'play.se/program')
         webpage = self._download_webpage(url, video_id)
-        urplayer_data = self._parse_json(self._html_search_regex(
+        vid = int(video_id)
+        accessible_episodes = self._parse_json(self._html_search_regex(
             r'data-react-class="routes/Product/components/ProgramContainer/ProgramContainer"[^>]+data-react-props="({.+?})"',
-            webpage, 'urplayer data'), video_id)['accessibleEpisodes'][0]
+            webpage, 'urplayer data'), video_id)['accessibleEpisodes']
+        urplayer_data = next(e for e in accessible_episodes if e.get('id') == vid)
         episode = urplayer_data['title']
 
         host = self._download_json('http://streaming-loadbalancer.ur.se/loadbalancer.json', video_id)['redirect']

yt_dlp/extractor/viki.py

@@ -255,15 +255,8 @@ class VikiIE(VikiBaseIE):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        resp = self._download_json(
+        video = self._call_api(
-            'https://www.viki.com/api/videos/' + video_id,
+            'videos/%s.json' % video_id, video_id, 'Downloading video JSON')
-            video_id, 'Downloading video JSON', headers={
-                'x-client-user-agent': std_headers['User-Agent'],
-                'x-viki-as-id': self._APP,
-                'x-viki-app-ver': self._APP_VERSION,
-            })
-        video = resp['video']
-
         self._check_errors(video)
 
         title = self.dict_selection(video.get('titles', {}), 'en', allow_fallback=False)
@@ -286,30 +279,12 @@ class VikiIE(VikiBaseIE):
             })
 
         subtitles = {}
-        try:
+        for subtitle_lang, _ in (video.get('subtitle_completions') or {}).items():
-            # New way to fetch subtitles
+            subtitles[subtitle_lang] = [{
-            new_video = self._download_json(
+                'ext': subtitles_format,
-                'https://www.viki.com/api/videos/%s' % video_id, video_id,
+                'url': self._prepare_call(
-                'Downloading new video JSON to get subtitles', fatal=False,
+                    'videos/%s/subtitles/%s.%s' % (video_id, subtitle_lang, subtitles_format)),
-                headers={
+            } for subtitles_format in ('srt', 'vtt')]
-                    'x-client-user-agent': std_headers['User-Agent'],
-                    'x-viki-as-id': self._APP,
-                    'x-viki-app-ver': self._APP_VERSION,
-                })
-            for sub in new_video.get('streamSubtitles').get('dash'):
-                subtitles[sub.get('srclang')] = [{
-                    'ext': 'vtt',
-                    'url': sub.get('src'),
-                    'completion': sub.get('percentage'),
-                }]
-        except AttributeError:
-            # fall-back to the old way if there isn't a streamSubtitles attribute
-            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')]
 
         result = {
             'id': video_id,
@@ -386,23 +361,20 @@ class VikiIE(VikiBaseIE):
                     'filesize': int_or_none(urlh.headers.get('Content-Length')),
                 })
 
-        for format_id, format_dict in (resp.get('streams') or {}).items():
+        streams = self._call_api(
-            add_format(format_id, format_dict)
+            'videos/%s/streams.json' % video_id, video_id,
-        if not formats:
+            'Downloading video streams JSON')
-            streams = self._call_api(
-                'videos/%s/streams.json' % video_id, video_id,
-                'Downloading video streams JSON')
 
-            if 'external' in streams:
+        if 'external' in streams:
-                result.update({
+            result.update({
-                    '_type': 'url_transparent',
+                '_type': 'url_transparent',
-                    'url': streams['external']['url'],
+                'url': streams['external']['url'],
-                })
+            })
-                return result
+            return result
 
-            for format_id, stream_dict in streams.items():
+        for format_id, stream_dict in streams.items():
-                for protocol, format_dict in stream_dict.items():
+            for protocol, format_dict in stream_dict.items():
-                    add_format(format_id, format_dict, protocol)
+                add_format(format_id, format_dict, protocol)
         self._sort_formats(formats)
 
         result['formats'] = formats

yt_dlp/extractor/vvvvid.py

@@ -75,12 +75,15 @@ class VVVVIDIE(InfoExtractor):
             'https://www.vvvvid.it/user/login',
             None, headers=self.geo_verification_headers())['data']['conn_id']
 
-    def _download_info(self, show_id, path, video_id, fatal=True):
+    def _download_info(self, show_id, path, video_id, fatal=True, query=None):
+        q = {
+            'conn_id': self._conn_id,
+        }
+        if query:
+            q.update(query)
         response = self._download_json(
             'https://www.vvvvid.it/vvvvid/ondemand/%s/%s' % (show_id, path),
-            video_id, headers=self.geo_verification_headers(), query={
+            video_id, headers=self.geo_verification_headers(), query=q, fatal=fatal)
-                'conn_id': self._conn_id,
-            }, fatal=fatal)
         if not (response or fatal):
             return
         if response.get('result') == 'error':
@@ -98,7 +101,8 @@ class VVVVIDIE(InfoExtractor):
         show_id, season_id, video_id = re.match(self._VALID_URL, url).groups()
 
         response = self._download_info(
-            show_id, 'season/%s' % season_id, video_id)
+            show_id, 'season/%s' % season_id,
+            video_id, query={'video_id': video_id})
 
         vid = int(video_id)
         video_data = list(filter(
@@ -247,9 +251,13 @@ class VVVVIDShowIE(VVVVIDIE):
         show_info = self._download_info(
             show_id, 'info/', show_title, fatal=False)
 
+        if not show_title:
+            base_url += "/title"
+
         entries = []
         for season in (seasons or []):
             episodes = season.get('episodes') or []
+            playlist_title = season.get('name') or show_info.get('title')
             for episode in episodes:
                 if episode.get('playable') is False:
                     continue
@@ -259,12 +267,13 @@ class VVVVIDShowIE(VVVVIDIE):
                     continue
                 info = self._extract_common_video_info(episode)
                 info.update({
-                    '_type': 'url',
+                    '_type': 'url_transparent',
                     'ie_key': VVVVIDIE.ie_key(),
                     'url': '/'.join([base_url, season_id, video_id]),
                     'title': episode.get('title'),
                     'description': episode.get('description'),
                     'season_id': season_id,
+                    'playlist_title': playlist_title,
                 })
                 entries.append(info)
 

yt_dlp/extractor/youtube.py

@@ -2,6 +2,7 @@
 
 from __future__ import unicode_literals
 
+import hashlib
 import itertools
 import json
 import os.path
@@ -274,7 +275,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         'context': {
             'client': {
                 'clientName': 'WEB',
-                'clientVersion': '2.20201021.03.00',
+                'clientVersion': '2.20210301.08.00',
             }
         },
     }
@@ -283,15 +284,28 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
     _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
     _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
 
-    def _call_api(self, ep, query, video_id, fatal=True):
+    def _generate_sapisidhash_header(self):
+        sapisid_cookie = self._get_cookies('https://www.youtube.com').get('SAPISID')
+        if sapisid_cookie is None:
+            return
+        time_now = round(time.time())
+        sapisidhash = hashlib.sha1((str(time_now) + " " + sapisid_cookie.value + " " + "https://www.youtube.com").encode("utf-8")).hexdigest()
+        return "SAPISIDHASH %s_%s" % (time_now, sapisidhash)
+
+    def _call_api(self, ep, query, video_id, fatal=True, headers=None,
+                  note='Downloading API JSON', errnote='Unable to download API page'):
         data = self._DEFAULT_API_DATA.copy()
         data.update(query)
+        headers = headers or {}
+        headers.update({'content-type': 'application/json'})
+        auth = self._generate_sapisidhash_header()
+        if auth is not None:
+            headers.update({'Authorization': auth, 'X-Origin': 'https://www.youtube.com'})
 
         return self._download_json(
-            'https://www.youtube.com/youtubei/v1/%s' % ep, video_id=video_id,
+            'https://www.youtube.com/youtubei/v1/%s' % ep,
-            note='Downloading API JSON', errnote='Unable to download API page',
+            video_id=video_id, fatal=fatal, note=note, errnote=errnote,
-            data=json.dumps(data).encode('utf8'), fatal=fatal,
+            data=json.dumps(data).encode('utf8'), headers=headers,
-            headers={'content-type': 'application/json'},
             query={'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'})
 
     def _extract_yt_initial_data(self, video_id, webpage):
@@ -1452,8 +1466,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         url, smuggled_data = unsmuggle_url(url, {})
         video_id = self._match_id(url)
         base_url = self.http_scheme() + '//www.youtube.com/'
-        webpage_url = base_url + 'watch?v=' + video_id + '&has_verified=1&bpctr=9999999999'
+        webpage_url = base_url + 'watch?v=' + video_id
-        webpage = self._download_webpage(webpage_url, video_id, fatal=False)
+        webpage = self._download_webpage(
+            webpage_url + '&has_verified=1&bpctr=9999999999',
+            video_id, fatal=False)
 
         player_response = None
         if webpage:
@@ -2010,9 +2026,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         # Get comments
         # TODO: Refactor and move to seperate function
-        if get_comments:
+        def extract_comments():
             expected_video_comment_count = 0
             video_comments = []
+            comment_xsrf = xsrf_token
 
             def find_value(html, key, num_chars=2, separator='"'):
                 pos_begin = html.find(key) + len(key) + num_chars
@@ -2081,7 +2098,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             self.to_screen('Downloading comments')
             while continuations:
                 continuation = continuations.pop()
-                comment_response = get_continuation(continuation, xsrf_token)
+                comment_response = get_continuation(continuation, comment_xsrf)
                 if not comment_response:
                     continue
                 if list(search_dict(comment_response, 'externalErrorMessage')):
@@ -2092,7 +2109,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     continue
                 # not sure if this actually helps
                 if 'xsrf_token' in comment_response:
-                    xsrf_token = comment_response['xsrf_token']
+                    comment_xsrf = comment_response['xsrf_token']
 
                 item_section = comment_response['response']['continuationContents']['itemSectionContinuation']
                 if first_continuation:
@@ -2121,7 +2138,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     while reply_continuations:
                         time.sleep(1)
                         continuation = reply_continuations.pop()
-                        replies_data = get_continuation(continuation, xsrf_token, True)
+                        replies_data = get_continuation(continuation, comment_xsrf, True)
                         if not replies_data or 'continuationContents' not in replies_data[1]['response']:
                             continue
 
@@ -2150,10 +2167,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 time.sleep(1)
 
             self.to_screen('Total comments downloaded: %d of ~%d' % (len(video_comments), expected_video_comment_count))
-            info.update({
+            return {
                 'comments': video_comments,
                 'comment_count': expected_video_comment_count
-            })
+            }
+
+        if get_comments:
+            info['__post_extractor'] = extract_comments
 
         self.mark_watched(video_id, player_response)
 
@@ -2693,7 +2713,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             ctp = continuation_ep.get('clickTrackingParams')
             return YoutubeTabIE._build_continuation_query(continuation, ctp)
 
-    def _entries(self, tab, identity_token):
+    def _entries(self, tab, identity_token, item_id):
 
         def extract_entries(parent_renderer):  # this needs to called again for continuation to work with feeds
             contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
@@ -2756,27 +2776,39 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
         for page_num in itertools.count(1):
             if not continuation:
                 break
-            count = 0
+            retries = self._downloader.params.get('extractor_retries', 3)
-            retries = 3
+            count = -1
-            while count <= retries:
+            last_error = None
+            while count < retries:
+                count += 1
+                if last_error:
+                    self.report_warning('%s. Retrying ...' % last_error)
                 try:
-                    # Downloading page may result in intermittent 5xx HTTP error
+                    response = self._call_api(
-                    # that is usually worked around with a retry
+                        ep="browse", fatal=True, headers=headers,
-                    browse = self._download_json(
+                        video_id='%s page %s' % (item_id, page_num),
-                        'https://www.youtube.com/browse_ajax', None,
+                        query={
-                        'Downloading page %d%s'
+                            'continuation': continuation['continuation'],
-                        % (page_num, ' (retry #%d)' % count if count else ''),
+                            'clickTracking': {'clickTrackingParams': continuation['itct']},
-                        headers=headers, query=continuation)
+                        },
-                    break
+                        note='Downloading API JSON%s' % (' (retry #%d)' % count if count else ''))
                 except ExtractorError as e:
-                    if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
+                    if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404):
-                        count += 1
+                        # Downloading page may result in intermittent 5xx HTTP error
-                        if count <= retries:
+                        # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
+                        last_error = 'HTTP Error %s' % e.cause.code
+                        if count < retries:
                             continue
                     raise
-            if not browse:
+                else:
-                break
+                    # Youtube sometimes sends incomplete data
-            response = try_get(browse, lambda x: x[1]['response'], dict)
+                    # See: https://github.com/ytdl-org/youtube-dl/issues/28194
+                    if response.get('continuationContents') or response.get('onResponseReceivedActions'):
+                        break
+                    last_error = 'Incomplete data recieved'
+                    if count >= retries:
+                        self._downloader.report_error(last_error)
+
             if not response:
                 break
 
@@ -2920,7 +2952,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'channel_id': metadata['uploader_id'],
             'channel_url': metadata['uploader_url']})
         return self.playlist_result(
-            self._entries(selected_tab, identity_token),
+            self._entries(selected_tab, identity_token, playlist_id),
             **metadata)
 
     def _extract_from_playlist(self, item_id, url, data, playlist):
@@ -2998,19 +3030,35 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
             self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
 
-        webpage = self._download_webpage(url, item_id)
+        retries = self._downloader.params.get('extractor_retries', 3)
-        identity_token = self._extract_identity_token(webpage, item_id)
+        count = -1
-        data = self._extract_yt_initial_data(item_id, webpage)
+        last_error = 'Incomplete yt initial data recieved'
-        err_msg = None
+        while count < retries:
-        for alert_type, alert_message in self._extract_alerts(data):
+            count += 1
-            if alert_type.lower() == 'error':
+            # Sometimes youtube returns a webpage with incomplete ytInitialData
-                if err_msg:
+            # See: https://github.com/yt-dlp/yt-dlp/issues/116
-                    self._downloader.report_warning('YouTube said: %s - %s' % ('ERROR', err_msg))
+            if count:
-                err_msg = alert_message
+                self.report_warning('%s. Retrying ...' % last_error)
-            else:
+            webpage = self._download_webpage(
-                self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
+                url, item_id,
-        if err_msg:
+                'Downloading webpage%s' % ' (retry #%d)' % count if count else '')
-            raise ExtractorError('YouTube said: %s' % err_msg, expected=True)
+            identity_token = self._extract_identity_token(webpage, item_id)
+            data = self._extract_yt_initial_data(item_id, webpage)
+            err_msg = None
+            for alert_type, alert_message in self._extract_alerts(data):
+                if alert_type.lower() == 'error':
+                    if err_msg:
+                        self._downloader.report_warning('YouTube said: %s - %s' % ('ERROR', err_msg))
+                    err_msg = alert_message
+                else:
+                    self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
+            if err_msg:
+                raise ExtractorError('YouTube said: %s' % err_msg, expected=True)
+            if data.get('contents') or data.get('currentVideoEndpoint'):
+                break
+            if count >= retries:
+                self._downloader.report_error(last_error)
+
         tabs = try_get(
             data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
         if tabs:
@@ -3191,26 +3239,14 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
     _TESTS = []
 
     def _entries(self, query, n):
-        data = {
+        data = {'query': query}
-            'context': {
-                'client': {
-                    'clientName': 'WEB',
-                    'clientVersion': '2.20201021.03.00',
-                }
-            },
-            'query': query,
-        }
         if self._SEARCH_PARAMS:
             data['params'] = self._SEARCH_PARAMS
         total = 0
         for page_num in itertools.count(1):
-            search = self._download_json(
+            search = self._call_api(
-                'https://www.youtube.com/youtubei/v1/search?key=AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
+                ep='search', video_id='query "%s"' % query, fatal=False,
-                video_id='query "%s"' % query,
+                note='Downloading page %s' % page_num, query=data)
-                note='Downloading page %s' % page_num,
-                errnote='Unable to download API page', fatal=False,
-                data=json.dumps(data).encode('utf8'),
-                headers={'content-type': 'application/json'})
             if not search:
                 break
             slr_contents = try_get(
@@ -3362,8 +3398,8 @@ class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
 
 
 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
-    IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
+    IE_DESC = 'Youtube watch history, ":ythis" for short (requires authentication)'
-    _VALID_URL = r':ythistory'
+    _VALID_URL = r':ythis(?:tory)?'
     _FEED_NAME = 'history'
     _TESTS = [{
         'url': ':ythistory',

yt_dlp/extractor/zdf.py

@@ -7,7 +7,9 @@ from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
     determine_ext,
+    float_or_none,
     int_or_none,
+    merge_dicts,
     NO_DEFAULT,
     orderedSet,
     parse_codecs,
@@ -21,61 +23,17 @@ from ..utils import (
 
 
 class ZDFBaseIE(InfoExtractor):
-    def _call_api(self, url, player, referrer, video_id, item):
-        return self._download_json(
-            url, video_id, 'Downloading JSON %s' % item,
-            headers={
-                'Referer': referrer,
-                'Api-Auth': 'Bearer %s' % player['apiToken'],
-            })
-
-    def _extract_player(self, webpage, video_id, fatal=True):
-        return self._parse_json(
-            self._search_regex(
-                r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
-                'player JSON', default='{}' if not fatal else NO_DEFAULT,
-                group='json'),
-            video_id)
-
-
-class ZDFIE(ZDFBaseIE):
-    IE_NAME = "ZDF-3sat"
-    _VALID_URL = r'https?://www\.(zdf|3sat)\.de/(?:[^/]+/)*(?P<id>[^/?]+)\.html'
-    _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')
     _GEO_COUNTRIES = ['DE']
+    _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')
 
-    _TESTS = [{
+    def _call_api(self, url, video_id, item, api_token=None, referrer=None):
-        'url': 'https://www.3sat.de/wissen/wissenschaftsdoku/luxusgut-lebensraum-100.html',
+        headers = {}
-        'info_dict': {
+        if api_token:
-            'id': 'luxusgut-lebensraum-100',
+            headers['Api-Auth'] = 'Bearer %s' % api_token
-            'ext': 'mp4',
+        if referrer:
-            'title': 'Luxusgut Lebensraum',
+            headers['Referer'] = referrer
-            'description': 'md5:5c09b2f45ac3bc5233d1b50fc543d061',
+        return self._download_json(
-            'duration': 2601,
+            url, video_id, 'Downloading JSON %s' % item, headers=headers)
-            'timestamp': 1566497700,
-            'upload_date': '20190822',
-        }
-    }, {
-        'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
-        'info_dict': {
-            'id': 'die-magie-der-farben-von-koenigspurpur-und-jeansblau-100',
-            'ext': 'mp4',
-            'title': 'Die Magie der Farben (2/2)',
-            'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
-            'duration': 2615,
-            'timestamp': 1465021200,
-            'upload_date': '20160604',
-        },
-    }, {
-        'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
-        'only_matching': True,
-    }]
 
     @staticmethod
     def _extract_subtitles(src):
@@ -121,20 +79,11 @@ class ZDFIE(ZDFBaseIE):
             })
             formats.append(f)
 
-    def _extract_entry(self, url, player, content, video_id):
+    def _extract_ptmd(self, ptmd_url, video_id, api_token, referrer):
-        title = content.get('title') or content['teaserHeadline']
-
-        t = content['mainVideoContent']['http://zdf.de/rels/target']
-
-        ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')
-
-        if not ptmd_path:
-            ptmd_path = t[
-                'http://zdf.de/rels/streams/ptmd-template'].replace(
-                '{playerId}', 'ngplayer_2_4')
-
         ptmd = self._call_api(
-            urljoin(url, ptmd_path), player, url, video_id, 'metadata')
+            ptmd_url, video_id, 'metadata', api_token, referrer)
+
+        content_id = ptmd.get('basename') or ptmd_url.split('/')[-1]
 
         formats = []
         track_uris = set()
@@ -152,7 +101,7 @@ class ZDFIE(ZDFBaseIE):
                         continue
                     for track in tracks:
                         self._extract_format(
-                            video_id, formats, track_uris, {
+                            content_id, formats, track_uris, {
                                 'url': track.get('uri'),
                                 'type': f.get('type'),
                                 'mimeType': f.get('mimeType'),
@@ -161,6 +110,103 @@ class ZDFIE(ZDFBaseIE):
                             })
         self._sort_formats(formats)
 
+        duration = float_or_none(try_get(
+            ptmd, lambda x: x['attributes']['duration']['value']), scale=1000)
+
+        return {
+            'extractor_key': ZDFIE.ie_key(),
+            'id': content_id,
+            'duration': duration,
+            'formats': formats,
+            'subtitles': self._extract_subtitles(ptmd),
+        }
+
+    def _extract_player(self, webpage, video_id, fatal=True):
+        return self._parse_json(
+            self._search_regex(
+                r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
+                'player JSON', default='{}' if not fatal else NO_DEFAULT,
+                group='json'),
+            video_id)
+
+
+class ZDFIE(ZDFBaseIE):
+    _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
+    _TESTS = [{
+        # Same as https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html
+        'url': 'https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html',
+        'md5': '34ec321e7eb34231fd88616c65c92db0',
+        'info_dict': {
+            'id': '210222_phx_nachgehakt_corona_protest',
+            'ext': 'mp4',
+            'title': 'Wohin führt der Protest in der Pandemie?',
+            'description': 'md5:7d643fe7f565e53a24aac036b2122fbd',
+            'duration': 1691,
+            'timestamp': 1613948400,
+            'upload_date': '20210221',
+        },
+    }, {
+        # Same as https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html
+        'url': 'https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html',
+        'md5': '0aff3e7bc72c8813f5e0fae333316a1d',
+        'info_dict': {
+            'id': '141007_ab18_10wochensommer_film',
+            'ext': 'mp4',
+            'title': 'Ab 18! - 10 Wochen Sommer',
+            'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26',
+            'duration': 2660,
+            'timestamp': 1608604200,
+            'upload_date': '20201222',
+        },
+    }, {
+        'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
+        'info_dict': {
+            'id': '151025_magie_farben2_tex',
+            'ext': 'mp4',
+            'title': 'Die Magie der Farben (2/2)',
+            'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
+            'duration': 2615,
+            'timestamp': 1465021200,
+            'upload_date': '20160604',
+        },
+    }, {
+        # Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche
+        'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html',
+        'only_matching': True,
+    }, {
+        # Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html
+        'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html',
+        'only_matching': True,
+    }, {
+        # Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
+        'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
+        'only_matching': True,
+    }]
+
+    def _extract_entry(self, url, player, content, video_id):
+        title = content.get('title') or content['teaserHeadline']
+
+        t = content['mainVideoContent']['http://zdf.de/rels/target']
+
+        ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')
+
+        if not ptmd_path:
+            ptmd_path = t[
+                'http://zdf.de/rels/streams/ptmd-template'].replace(
+                '{playerId}', 'ngplayer_2_4')
+
+        info = self._extract_ptmd(
+            urljoin(url, ptmd_path), video_id, player['apiToken'], url)
+
         thumbnails = []
         layouts = try_get(
             content, lambda x: x['teaserImageRef']['layouts'], dict)
@@ -181,33 +227,33 @@ class ZDFIE(ZDFBaseIE):
                     })
                 thumbnails.append(thumbnail)
 
-        return {
+        return merge_dicts(info, {
-            'id': video_id,
             'title': title,
             'description': content.get('leadParagraph') or content.get('teasertext'),
             'duration': int_or_none(t.get('duration')),
             'timestamp': unified_timestamp(content.get('editorialDate')),
             'thumbnails': thumbnails,
-            'subtitles': self._extract_subtitles(ptmd),
+        })
-            'formats': formats,
-        }
 
     def _extract_regular(self, url, player, video_id):
         content = self._call_api(
-            player['content'], player, url, video_id, 'content')
+            player['content'], video_id, 'content', player['apiToken'], url)
         return self._extract_entry(player['content'], player, content, video_id)
 
     def _extract_mobile(self, video_id):
-        document = self._download_json(
+        video = self._download_json(
             'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
-            video_id)['document']
+            video_id)
+
+        document = video['document']
 
         title = document['titel']
+        content_id = document['basename']
 
         formats = []
         format_urls = set()
         for f in document['formitaeten']:
-            self._extract_format(video_id, formats, format_urls, f)
+            self._extract_format(content_id, formats, format_urls, f)
         self._sort_formats(formats)
 
         thumbnails = []
@@ -225,12 +271,12 @@ class ZDFIE(ZDFBaseIE):
                     })
 
         return {
-            'id': video_id,
+            'id': content_id,
             'title': title,
             'description': document.get('beschreibung'),
             'duration': int_or_none(document.get('length')),
-            'timestamp': unified_timestamp(try_get(
+            'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
-                document, lambda x: x['meta']['editorialDate'], compat_str)),
+                try_get(video, lambda x: x['meta']['editorialDate'], compat_str)),
             'thumbnails': thumbnails,
             'subtitles': self._extract_subtitles(document),
             'formats': formats,

yt_dlp/options.py

@@ -347,7 +347,7 @@ def parseOpts(overrideArguments=None):
             'Specify any key (see "OUTPUT TEMPLATE" for a list of available keys) to '
             'match if the key is present, '
             '!key to check if the key is not present, '
-            'key>NUMBER (like "comment_count > 12", also works with '
+            'key>NUMBER (like "view_count > 12", also works with '
             '>=, <, <=, !=, =) to compare against a number, '
             'key = \'LITERAL\' (like "uploader = \'Mike Smith\'", also works with !=) '
             'to match against a string literal '
@@ -369,7 +369,7 @@ def parseOpts(overrideArguments=None):
         help='Download only the video, if the URL refers to a video and a playlist')
     selection.add_option(
         '--yes-playlist',
-        action='store_false', dest='noplaylist', default=False,
+        action='store_false', dest='noplaylist',
         help='Download the playlist, if the URL refers to a video and a playlist')
     selection.add_option(
         '--age-limit',
@@ -634,16 +634,24 @@ def parseOpts(overrideArguments=None):
         help='Use ffmpeg instead of the native HLS downloader')
     downloader.add_option(
         '--hls-use-mpegts',
-        dest='hls_use_mpegts', action='store_true',
+        dest='hls_use_mpegts', action='store_true', default=None,
         help=(
-            'Use the mpegts container for HLS videos, allowing to play the '
+            'Use the mpegts container for HLS videos; '
-            'video while downloading (some players may not be able to play it)'))
+            'allowing some players to play the video while downloading, '
+            'and reducing the chance of file corruption if download is interrupted. '
+            'This is enabled by default for live streams'))
+    downloader.add_option(
+        '--no-hls-use-mpegts',
+        dest='hls_use_mpegts', action='store_false',
+        help=(
+            'Do not use the mpegts container for HLS videos. '
+            'This is default when not downloading live streams'))
     downloader.add_option(
         '--external-downloader',
         dest='external_downloader', metavar='NAME',
         help=(
-            'Use the specified external downloader. '
+            'Name or path of the external downloader to use. '
-            'Currently supports %s' % ', '.join(list_external_downloaders())))
+            'Currently supports %s (Recommended: aria2c)' % ', '.join(list_external_downloaders())))
     downloader.add_option(
         '--downloader-args', '--external-downloader-args',
         metavar='NAME:ARGS', dest='external_downloader_args', default={}, type='str',
@@ -688,6 +696,10 @@ def parseOpts(overrideArguments=None):
         '--bidi-workaround',
         dest='bidi_workaround', action='store_true',
         help='Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
+    workarounds.add_option(
+        '--sleep-requests', metavar='SECONDS',
+        dest='sleep_interval_requests', type=float,
+        help='Number of seconds to sleep between requests during data extraction')
     workarounds.add_option(
         '--sleep-interval', '--min-sleep-interval', metavar='SECONDS',
         dest='sleep_interval', type=float,
@@ -706,7 +718,7 @@ def parseOpts(overrideArguments=None):
     workarounds.add_option(
         '--sleep-subtitles', metavar='SECONDS',
         dest='sleep_interval_subtitles', default=0, type=int,
-        help='Enforce sleep interval on subtitles as well')
+        help='Number of seconds to sleep before each subtitle download')
 
     verbosity = optparse.OptionGroup(parser, 'Verbosity and Simulation Options')
     verbosity.add_option(
@@ -973,7 +985,9 @@ def parseOpts(overrideArguments=None):
     filesystem.add_option(
         '--get-comments',
         action='store_true', dest='getcomments', default=False,
-        help='Retrieve video comments to be placed in the .info.json file')
+        help=(
+            'Retrieve video comments to be placed in the .info.json file. '
+            'The comments are fetched even without this option if the extraction is known to be quick'))
     filesystem.add_option(
         '--load-info-json', '--load-info',
         dest='load_info_filename', metavar='FILE',
@@ -1129,7 +1143,7 @@ def parseOpts(overrideArguments=None):
             'Give field name to extract data from, and format of the field seperated by a ":". '
             'Either regular expression with named capture groups or a '
             'similar syntax to the output template can also be used. '
-            'The parsed parameters replace any existing values and can be use in output template'
+            'The parsed parameters replace any existing values and can be use in output template. '
             'This option can be used multiple times. '
             'Example: --parse-metadata "title:%(artist)s - %(title)s" matches a title like '
             '"Coldplay - Paradise". '
@@ -1204,6 +1218,10 @@ def parseOpts(overrideArguments=None):
         help=optparse.SUPPRESS_HELP)
 
     extractor = optparse.OptionGroup(parser, 'Extractor Options')
+    extractor.add_option(
+        '--extractor-retries',
+        dest='extractor_retries', metavar='RETRIES', default=3,
+        help='Number of retries for known extractor errors (default is %default), or "infinite"')
     extractor.add_option(
         '--allow-dynamic-mpd', '--no-ignore-dynamic-mpd',
         action='store_true', dest='dynamic_mpd', default=True,

yt_dlp/postprocessor/embedthumbnail.py

@@ -193,4 +193,6 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
                 info['__thumbnail_filename'], os.path.splitext(original_thumbnail)[1][1:])
             if original_thumbnail == thumbnail_filename:
                 files_to_delete = []
+        elif original_thumbnail != thumbnail_filename:
+            files_to_delete.append(original_thumbnail)
         return files_to_delete, info

yt_dlp/utils.py

@@ -5945,9 +5945,13 @@ def make_dir(path, to_screen=None):
 
 
 def get_executable_path():
-    path = os.path.dirname(sys.argv[0])
+    from zipimport import zipimporter
-    if os.path.basename(sys.argv[0]) == '__main__':  # Running from source
+    if hasattr(sys, 'frozen'):  # Running from PyInstaller
-        path = os.path.join(path, '..')
+        path = os.path.dirname(sys.executable)
+    elif isinstance(globals().get('__loader__'), zipimporter):  # Running from ZIP
+        path = os.path.join(os.path.dirname(__file__), '../..')
+    else:
+        path = os.path.join(os.path.dirname(__file__), '..')
     return os.path.abspath(path)
 
 

yt_dlp/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2021.02.19'
+__version__ = '2021.03.01'