[version] update

Created by: pukkandan

:ci skip all :ci run dl

.github/ISSUE_TEMPLATE/1_broken_site.yml

@@ -11,7 +11,7 @@ body:
       options:
         - label: I'm reporting a broken site
           required: true
-        - label: I've verified that I'm running yt-dlp version **2022.05.18** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
+        - label: I've verified that I'm running yt-dlp version **2022.06.29** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
           required: true
         - label: I've checked that all provided URLs are playable in a browser with the same IP and same login details
           required: true
@@ -51,12 +51,12 @@ body:
         [debug] Portable config file: yt-dlp.conf
         [debug] Portable config: ['-i']
         [debug] Encodings: locale cp1252, fs utf-8, stdout utf-8, stderr utf-8, pref cp1252
-        [debug] yt-dlp version 2022.05.18 (exe)
+        [debug] yt-dlp version 2022.06.29 (exe)
         [debug] Python version 3.8.8 (CPython 64bit) - Windows-10-10.0.19041-SP0
         [debug] exe versions: ffmpeg 3.0.1, ffprobe 3.0.1
         [debug] Optional libraries: Cryptodome, keyring, mutagen, sqlite, websockets
         [debug] Proxy map: {}
-        yt-dlp is up to date (2022.05.18)
+        yt-dlp is up to date (2022.06.29)
         <more lines>
       render: shell
     validations:

.github/ISSUE_TEMPLATE/2_site_support_request.yml

@@ -11,7 +11,7 @@ body:
       options:
         - label: I'm reporting a new site support request
           required: true
-        - label: I've verified that I'm running yt-dlp version **2022.05.18** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
+        - label: I've verified that I'm running yt-dlp version **2022.06.29** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
           required: true
         - label: I've checked that all provided URLs are playable in a browser with the same IP and same login details
           required: true
@@ -62,12 +62,12 @@ body:
         [debug] Portable config file: yt-dlp.conf
         [debug] Portable config: ['-i']
         [debug] Encodings: locale cp1252, fs utf-8, stdout utf-8, stderr utf-8, pref cp1252
-        [debug] yt-dlp version 2022.05.18 (exe)
+        [debug] yt-dlp version 2022.06.29 (exe)
         [debug] Python version 3.8.8 (CPython 64bit) - Windows-10-10.0.19041-SP0
         [debug] exe versions: ffmpeg 3.0.1, ffprobe 3.0.1
         [debug] Optional libraries: Cryptodome, keyring, mutagen, sqlite, websockets
         [debug] Proxy map: {}
-        yt-dlp is up to date (2022.05.18)
+        yt-dlp is up to date (2022.06.29)
         <more lines>
       render: shell
     validations:

.github/ISSUE_TEMPLATE/3_site_feature_request.yml

@@ -9,9 +9,9 @@ body:
       description: |
         Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
       options:
-        - label: I'm reporting a site feature request
+        - label: I'm requesting a site-specific feature
           required: true
-        - label: I've verified that I'm running yt-dlp version **2022.05.18** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
+        - label: I've verified that I'm running yt-dlp version **2022.06.29** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
           required: true
         - label: I've checked that all provided URLs are playable in a browser with the same IP and same login details
           required: true
@@ -60,12 +60,12 @@ body:
         [debug] Portable config file: yt-dlp.conf
         [debug] Portable config: ['-i']
         [debug] Encodings: locale cp1252, fs utf-8, stdout utf-8, stderr utf-8, pref cp1252
-        [debug] yt-dlp version 2022.05.18 (exe)
+        [debug] yt-dlp version 2022.06.29 (exe)
         [debug] Python version 3.8.8 (CPython 64bit) - Windows-10-10.0.19041-SP0
         [debug] exe versions: ffmpeg 3.0.1, ffprobe 3.0.1
         [debug] Optional libraries: Cryptodome, keyring, mutagen, sqlite, websockets
         [debug] Proxy map: {}
-        yt-dlp is up to date (2022.05.18)
+        yt-dlp is up to date (2022.06.29)
         <more lines>
       render: shell
     validations:

.github/ISSUE_TEMPLATE/4_bug_report.yml

@@ -11,7 +11,7 @@ body:
       options:
         - label: I'm reporting a bug unrelated to a specific site
           required: true
-        - label: I've verified that I'm running yt-dlp version **2022.05.18** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
+        - label: I've verified that I'm running yt-dlp version **2022.06.29** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
           required: true
         - label: I've checked that all provided URLs are playable in a browser with the same IP and same login details
           required: true
@@ -45,12 +45,12 @@ body:
         [debug] Portable config file: yt-dlp.conf
         [debug] Portable config: ['-i']
         [debug] Encodings: locale cp1252, fs utf-8, stdout utf-8, stderr utf-8, pref cp1252
-        [debug] yt-dlp version 2022.05.18 (exe)
+        [debug] yt-dlp version 2022.06.29 (exe)
         [debug] Python version 3.8.8 (CPython 64bit) - Windows-10-10.0.19041-SP0
         [debug] exe versions: ffmpeg 3.0.1, ffprobe 3.0.1
         [debug] Optional libraries: Cryptodome, keyring, mutagen, sqlite, websockets
         [debug] Proxy map: {}
-        yt-dlp is up to date (2022.05.18)
+        yt-dlp is up to date (2022.06.29)
         <more lines>
       render: shell
     validations:

.github/ISSUE_TEMPLATE/5_feature_request.yml

@@ -9,11 +9,11 @@ body:
       description: |
         Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
       options:
-        - label: I'm reporting a feature request
+        - label: I'm requesting a feature unrelated to a specific site
           required: true
         - label: I've looked through the [README](https://github.com/yt-dlp/yt-dlp#readme)
           required: true
-        - label: I've verified that I'm running yt-dlp version **2022.05.18** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
+        - label: I've verified that I'm running yt-dlp version **2022.06.29** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
           required: true
         - label: I've searched the [bugtracker](https://github.com/yt-dlp/yt-dlp/issues?q=) for similar issues including closed ones. DO NOT post duplicates
           required: true

.github/ISSUE_TEMPLATE/6_question.yml

@@ -9,13 +9,15 @@ body:
       description: |
         Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
       options:
-        - label: I'm asking a question and **not** reporting a bug/feature request
+        - label: I'm asking a question and **not** reporting a bug or requesting a feature
           required: true
         - label: I've looked through the [README](https://github.com/yt-dlp/yt-dlp#readme)
           required: true
-        - label: I've read the [guidelines for opening an issue](https://github.com/yt-dlp/yt-dlp/blob/master/CONTRIBUTING.md#opening-an-issue)
+        - label: I've verified that I'm running yt-dlp version **2022.06.29** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
           required: true
-        - label: I've searched the [bugtracker](https://github.com/yt-dlp/yt-dlp/issues?q=) for similar questions including closed ones
+        - label: I've searched the [bugtracker](https://github.com/yt-dlp/yt-dlp/issues?q=) for similar questions including closed ones. DO NOT post duplicates
+          required: true
+        - label: I've read the [guidelines for opening an issue](https://github.com/yt-dlp/yt-dlp/blob/master/CONTRIBUTING.md#opening-an-issue)
           required: true
   - type: textarea
     id: question

.github/ISSUE_TEMPLATE_tmpl/3_site_feature_request.yml

@@ -9,7 +9,7 @@ body:
       description: |
         Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
       options:
-        - label: I'm reporting a site feature request
+        - label: I'm requesting a site-specific feature
           required: true
         - label: I've verified that I'm running yt-dlp version **%(version)s** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
           required: true

.github/ISSUE_TEMPLATE_tmpl/5_feature_request.yml

@@ -9,7 +9,7 @@ body:
       description: |
         Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
       options:
-        - label: I'm reporting a feature request
+        - label: I'm requesting a feature unrelated to a specific site
           required: true
         - label: I've looked through the [README](https://github.com/yt-dlp/yt-dlp#readme)
           required: true

.github/ISSUE_TEMPLATE_tmpl/6_question.yml

@@ -9,13 +9,15 @@ body:
       description: |
         Carefully read and work through this check list in order to prevent the most common mistakes and misuse of yt-dlp:
       options:
-        - label: I'm asking a question and **not** reporting a bug/feature request
+        - label: I'm asking a question and **not** reporting a bug or requesting a feature
           required: true
         - label: I've looked through the [README](https://github.com/yt-dlp/yt-dlp#readme)
           required: true
-        - label: I've read the [guidelines for opening an issue](https://github.com/yt-dlp/yt-dlp/blob/master/CONTRIBUTING.md#opening-an-issue)
+        - label: I've verified that I'm running yt-dlp version **%(version)s** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
           required: true
-        - label: I've searched the [bugtracker](https://github.com/yt-dlp/yt-dlp/issues?q=) for similar questions including closed ones
+        - label: I've searched the [bugtracker](https://github.com/yt-dlp/yt-dlp/issues?q=) for similar questions including closed ones. DO NOT post duplicates
+          required: true
+        - label: I've read the [guidelines for opening an issue](https://github.com/yt-dlp/yt-dlp/blob/master/CONTRIBUTING.md#opening-an-issue)
           required: true
   - type: textarea
     id: question

.github/workflows/build.yml

@@ -2,27 +2,21 @@ name: Build
 on: workflow_dispatch
 
 jobs:
-  build_unix:
+  create_release:
     runs-on: ubuntu-latest
     outputs:
       version_suffix: ${{ steps.version_suffix.outputs.version_suffix }}
       ytdlp_version: ${{ steps.bump_version.outputs.ytdlp_version }}
       upload_url: ${{ steps.create_release.outputs.upload_url }}
-      sha256_bin: ${{ steps.sha256_bin.outputs.sha256_bin }}
-      sha512_bin: ${{ steps.sha512_bin.outputs.sha512_bin }}
-      sha256_tar: ${{ steps.sha256_tar.outputs.sha256_tar }}
-      sha512_tar: ${{ steps.sha512_tar.outputs.sha512_tar }}
-
+      release_id: ${{ steps.create_release.outputs.id }}
     steps:
     - uses: actions/checkout@v2
       with:
         fetch-depth: 0
-    - name: Set up Python
-      uses: actions/setup-python@v2
+    - uses: actions/setup-python@v2
       with:
-          python-version: '3.8'
-    - name: Install packages
-      run: sudo apt-get -y install zip pandoc man
+          python-version: '3.10'
+
     - name: Set version suffix
       id: version_suffix
       env:
@@ -34,55 +28,136 @@ jobs:
       run: |
         python devscripts/update-version.py ${{ steps.version_suffix.outputs.version_suffix }}
         make issuetemplates
+
     - name: Push to release
       id: push_release
       run: |
         git config --global user.name github-actions
         git config --global user.email github-actions@example.com
         git add -u
-        git commit -m "[version] update" -m "Created by: ${{ github.event.sender.login }}" -m ":ci skip all"
+        git commit -m "[version] update" -m "Created by: ${{ github.event.sender.login }}" -m ":ci skip all :ci run dl"
         git push origin --force ${{ github.event.ref }}:release
         echo ::set-output name=head_sha::$(git rev-parse HEAD)
     - name: Update master
-      id: push_master
       env:
         PUSH_VERSION_COMMIT: ${{ secrets.PUSH_VERSION_COMMIT }}
       if: "env.PUSH_VERSION_COMMIT != ''"
       run: git push origin ${{ github.event.ref }}
     - name: Get Changelog
-      id: get_changelog
       run: |
-        changelog=$(cat Changelog.md | grep -oPz '(?s)(?<=### ${{ steps.bump_version.outputs.ytdlp_version }}\n{2}).+?(?=\n{2,3}###)') || true
+        changelog=$(grep -oPz '(?s)(?<=### ${{ steps.bump_version.outputs.ytdlp_version }}\n{2}).+?(?=\n{2,3}###)' Changelog.md) || true
         echo "changelog<<EOF" >> $GITHUB_ENV
         echo "$changelog" >> $GITHUB_ENV
         echo "EOF" >> $GITHUB_ENV
 
-    - name: Build lazy extractors
-      id: lazy_extractors
-      run: python devscripts/make_lazy_extractors.py
-    - name: Run Make
-      run: make all tar
-    - name: Get SHA2-256SUMS for yt-dlp
-      id: sha256_bin
-      run: echo "::set-output name=sha256_bin::$(sha256sum yt-dlp | awk '{print $1}')"
-    - name: Get SHA2-256SUMS for yt-dlp.tar.gz
-      id: sha256_tar
-      run: echo "::set-output name=sha256_tar::$(sha256sum yt-dlp.tar.gz | awk '{print $1}')"
-    - name: Get SHA2-512SUMS for yt-dlp
-      id: sha512_bin
-      run: echo "::set-output name=sha512_bin::$(sha512sum yt-dlp | awk '{print $1}')"
-    - name: Get SHA2-512SUMS for yt-dlp.tar.gz
-      id: sha512_tar
-      run: echo "::set-output name=sha512_tar::$(sha512sum yt-dlp.tar.gz | awk '{print $1}')"
-
-    - name: Install dependencies for pypi
+    - name: Create Release
+      id: create_release
+      uses: actions/create-release@v1
       env:
-        PYPI_TOKEN: ${{ secrets.PYPI_TOKEN }}
-      if: "env.PYPI_TOKEN != ''"
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        tag_name: ${{ steps.bump_version.outputs.ytdlp_version }}
+        release_name: yt-dlp ${{ steps.bump_version.outputs.ytdlp_version }}
+        commitish: ${{ steps.push_release.outputs.head_sha }}
+        draft: true
+        prerelease: false
+        body: |
+          #### [A description of the various files]((https://github.com/yt-dlp/yt-dlp#release-files)) are in the README
+
+          ---
+          <details open><summary><h3>Changelog</summary>
+          <p>
+
+          ${{ env.changelog }}
+
+          </p>
+          </details>
+
+
+  build_unix:
+    needs: create_release
+    runs-on: ubuntu-18.04  # Standalone executable should be built on minimum supported OS
+    outputs:
+      sha256_bin: ${{ steps.get_sha.outputs.sha256_bin }}
+      sha512_bin: ${{ steps.get_sha.outputs.sha512_bin }}
+      sha256_tar: ${{ steps.get_sha.outputs.sha256_tar }}
+      sha512_tar: ${{ steps.get_sha.outputs.sha512_tar }}
+      sha256_linux: ${{ steps.get_sha.outputs.sha256_linux }}
+      sha512_linux: ${{ steps.get_sha.outputs.sha512_linux }}
+      sha256_linux_zip: ${{ steps.get_sha.outputs.sha256_linux_zip }}
+      sha512_linux_zip: ${{ steps.get_sha.outputs.sha512_linux_zip }}
+
+    steps:
+    - uses: actions/checkout@v2
+    - uses: actions/setup-python@v2
+      with:
+          python-version: '3.10'
+    - name: Install Requirements
       run: |
-        python -m pip install --upgrade pip
-        pip install setuptools wheel twine
-    - name: Build and publish on pypi
+          sudo apt-get -y install zip pandoc man
+          python -m pip install --upgrade pip setuptools wheel twine
+          python -m pip install Pyinstaller -r requirements.txt
+
+    - name: Prepare
+      run: |
+          python devscripts/update-version.py ${{ needs.create_release.outputs.version_suffix }}
+          python devscripts/make_lazy_extractors.py
+    - name: Build Unix executables
+      run: |
+          make all tar
+          python pyinst.py --onedir
+          (cd ./dist/yt-dlp_linux && zip -r ../yt-dlp_linux.zip .)
+          python pyinst.py
+    - name: Get SHA2-SUMS
+      id: get_sha
+      run: |
+          echo "::set-output name=sha256_bin::$(sha256sum yt-dlp | awk '{print $1}')"
+          echo "::set-output name=sha512_bin::$(sha512sum yt-dlp | awk '{print $1}')"
+          echo "::set-output name=sha256_tar::$(sha256sum yt-dlp.tar.gz | awk '{print $1}')"
+          echo "::set-output name=sha512_tar::$(sha512sum yt-dlp.tar.gz | awk '{print $1}')"
+          echo "::set-output name=sha256_linux::$(sha256sum dist/yt-dlp_linux | awk '{print $1}')"
+          echo "::set-output name=sha512_linux::$(sha512sum dist/yt-dlp_linux | awk '{print $1}')"
+          echo "::set-output name=sha256_linux_zip::$(sha256sum dist/yt-dlp_linux.zip | awk '{print $1}')"
+          echo "::set-output name=sha512_linux_zip::$(sha512sum dist/yt-dlp_linux.zip | awk '{print $1}')"
+
+    - name: Upload zip binary
+      uses: actions/upload-release-asset@v1
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
+        asset_path: ./yt-dlp
+        asset_name: yt-dlp
+        asset_content_type: application/octet-stream
+    - name: Upload Source tar
+      uses: actions/upload-release-asset@v1
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
+        asset_path: ./yt-dlp.tar.gz
+        asset_name: yt-dlp.tar.gz
+        asset_content_type: application/gzip
+    - name: Upload standalone binary
+      uses: actions/upload-release-asset@v1
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
+        asset_path: ./dist/yt-dlp_linux
+        asset_name: yt-dlp_linux
+        asset_content_type: application/octet-stream
+    - name: Upload onedir binary
+      uses: actions/upload-release-asset@v1
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
+        asset_path: ./dist/yt-dlp_linux.zip
+        asset_name: yt-dlp_linux.zip
+        asset_content_type: application/zip
+
+    - name: Build and publish on PyPi
       env:
         TWINE_USERNAME: __token__
         TWINE_PASSWORD: ${{ secrets.PYPI_TOKEN }}
@@ -92,7 +167,7 @@ jobs:
         python setup.py sdist bdist_wheel
         twine upload dist/*
 
-    - name: Install SSH private key
+    - name: Install SSH private key for Homebrew
       env:
         BREW_TOKEN: ${{ secrets.BREW_TOKEN }}
       if: "env.BREW_TOKEN != ''"
@@ -105,309 +180,292 @@ jobs:
       if: "env.BREW_TOKEN != ''"
       run: |
         git clone git@github.com:yt-dlp/homebrew-taps taps/
-        python3 devscripts/update-formulae.py taps/Formula/yt-dlp.rb "${{ steps.bump_version.outputs.ytdlp_version }}"
+        python devscripts/update-formulae.py taps/Formula/yt-dlp.rb "${{ needs.create_release.outputs.ytdlp_version }}"
         git -C taps/ config user.name github-actions
         git -C taps/ config user.email github-actions@example.com
-        git -C taps/ commit -am 'yt-dlp: ${{ steps.bump_version.outputs.ytdlp_version }}'
+        git -C taps/ commit -am 'yt-dlp: ${{ needs.create_release.outputs.ytdlp_version }}'
         git -C taps/ push
 
-    - name: Create Release
-      id: create_release
-      uses: actions/create-release@v1
-      env:
-        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-      with:
-        tag_name: ${{ steps.bump_version.outputs.ytdlp_version }}
-        release_name: yt-dlp ${{ steps.bump_version.outputs.ytdlp_version }}
-        commitish: ${{ steps.push_release.outputs.head_sha }}
-        body: |
-          #### [A description of the various files]((https://github.com/yt-dlp/yt-dlp#release-files)) are in the README
-
-          ---
-
-          ### Changelog:
-          ${{ env.changelog }}
-        draft: false
-        prerelease: false
-    - name: Upload yt-dlp Unix binary
-      id: upload-release-asset
-      uses: actions/upload-release-asset@v1
-      env:
-        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-      with:
-        upload_url: ${{ steps.create_release.outputs.upload_url }}
-        asset_path: ./yt-dlp
-        asset_name: yt-dlp
-        asset_content_type: application/octet-stream
-    - name: Upload Source tar
-      uses: actions/upload-release-asset@v1
-      env:
-        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-      with:
-        upload_url: ${{ steps.create_release.outputs.upload_url }}
-        asset_path: ./yt-dlp.tar.gz
-        asset_name: yt-dlp.tar.gz
-        asset_content_type: application/gzip
 
   build_macos:
     runs-on: macos-11
-    needs: build_unix
+    needs: create_release
     outputs:
-      sha256_macos: ${{ steps.sha256_macos.outputs.sha256_macos }}
-      sha512_macos: ${{ steps.sha512_macos.outputs.sha512_macos }}
-      sha256_macos_zip: ${{ steps.sha256_macos_zip.outputs.sha256_macos_zip }}
-      sha512_macos_zip: ${{ steps.sha512_macos_zip.outputs.sha512_macos_zip }}
+      sha256_macos: ${{ steps.get_sha.outputs.sha256_macos }}
+      sha512_macos: ${{ steps.get_sha.outputs.sha512_macos }}
+      sha256_macos_zip: ${{ steps.get_sha.outputs.sha256_macos_zip }}
+      sha512_macos_zip: ${{ steps.get_sha.outputs.sha512_macos_zip }}
 
     steps:
     - uses: actions/checkout@v2
-    # In order to create a universal2 application, the version of python3 in /usr/bin has to be used
+    # NB: In order to create a universal2 application, the version of python3 in /usr/bin has to be used
     - name: Install Requirements
       run: |
           brew install coreutils
-          /usr/bin/python3 -m pip install -U --user pip Pyinstaller==4.10 -r requirements.txt
-    - name: Bump version
-      id: bump_version
-      run: /usr/bin/python3 devscripts/update-version.py
-    - name: Build lazy extractors
-      id: lazy_extractors
-      run: /usr/bin/python3 devscripts/make_lazy_extractors.py
-    - name: Run PyInstaller Script
-      run: /usr/bin/python3 pyinst.py --target-architecture universal2 --onefile
-    - name: Upload yt-dlp MacOS binary
-      id: upload-release-macos
+          /usr/bin/python3 -m pip install -U --user pip Pyinstaller -r requirements.txt
+
+    - name: Prepare
+      run: |
+          /usr/bin/python3 devscripts/update-version.py ${{ needs.create_release.outputs.version_suffix }}
+          /usr/bin/python3 devscripts/make_lazy_extractors.py
+    - name: Build
+      run: |
+          /usr/bin/python3 pyinst.py --target-architecture universal2 --onedir
+          (cd ./dist/yt-dlp_macos && zip -r ../yt-dlp_macos.zip .)
+          /usr/bin/python3 pyinst.py --target-architecture universal2
+    - name: Get SHA2-SUMS
+      id: get_sha
+      run: |
+          echo "::set-output name=sha256_macos::$(sha256sum dist/yt-dlp_macos | awk '{print $1}')"
+          echo "::set-output name=sha512_macos::$(sha512sum dist/yt-dlp_macos | awk '{print $1}')"
+          echo "::set-output name=sha256_macos_zip::$(sha256sum dist/yt-dlp_macos.zip | awk '{print $1}')"
+          echo "::set-output name=sha512_macos_zip::$(sha512sum dist/yt-dlp_macos.zip | awk '{print $1}')"
+
+    - name: Upload standalone binary
       uses: actions/upload-release-asset@v1
       env:
         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
       with:
-        upload_url: ${{ needs.build_unix.outputs.upload_url }}
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
         asset_path: ./dist/yt-dlp_macos
         asset_name: yt-dlp_macos
         asset_content_type: application/octet-stream
-    - name: Get SHA2-256SUMS for yt-dlp_macos
-      id: sha256_macos
-      run: echo "::set-output name=sha256_macos::$(sha256sum dist/yt-dlp_macos | awk '{print $1}')"
-    - name: Get SHA2-512SUMS for yt-dlp_macos
-      id: sha512_macos
-      run: echo "::set-output name=sha512_macos::$(sha512sum dist/yt-dlp_macos | awk '{print $1}')"
-
-    - name: Run PyInstaller Script with --onedir
-      run: |
-        /usr/bin/python3 pyinst.py --target-architecture universal2 --onedir
-        zip ./dist/yt-dlp_macos.zip ./dist/yt-dlp_macos
-    - name: Upload yt-dlp MacOS onedir
-      id: upload-release-macos-zip
+    - name: Upload onedir binary
       uses: actions/upload-release-asset@v1
       env:
         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
       with:
-        upload_url: ${{ needs.build_unix.outputs.upload_url }}
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
         asset_path: ./dist/yt-dlp_macos.zip
         asset_name: yt-dlp_macos.zip
         asset_content_type: application/zip
-    - name: Get SHA2-256SUMS for yt-dlp_macos.zip
-      id: sha256_macos_zip
-      run: echo "::set-output name=sha256_macos_zip::$(sha256sum dist/yt-dlp_macos.zip | awk '{print $1}')"
-    - name: Get SHA2-512SUMS for yt-dlp_macos.zip
-      id: sha512_macos_zip
-      run: echo "::set-output name=sha512_macos_zip::$(sha512sum dist/yt-dlp_macos.zip | awk '{print $1}')"
+
+
+  build_macos_legacy:
+    runs-on: macos-latest
+    needs: create_release
+    steps:
+    - uses: actions/checkout@v2
+    - name: Install Python
+      # We need the official Python, because the GA ones only support newer macOS versions
+      env:
+        PYTHON_VERSION: 3.10.5
+        MACOSX_DEPLOYMENT_TARGET: 10.9  # Used up by the Python build tools
+      run: |
+          # Hack to get the latest patch version. Uncomment if needed
+          #brew install python@3.10
+          #export PYTHON_VERSION=$( $(brew --prefix)/opt/python@3.10/bin/python3 --version | cut -d ' ' -f 2 )
+          curl https://www.python.org/ftp/python/${PYTHON_VERSION}/python-${PYTHON_VERSION}-macos11.pkg -o "python.pkg"
+          sudo installer -pkg python.pkg -target /
+          python3 --version
+    - name: Install Requirements
+      run: |
+          brew install coreutils
+          python3 -m pip install -U --user pip Pyinstaller -r requirements.txt
+
+    - name: Prepare
+      run: |
+          python3 devscripts/update-version.py ${{ needs.create_release.outputs.version_suffix }}
+          python3 devscripts/make_lazy_extractors.py
+    - name: Build
+      run: |
+          python3 pyinst.py
+    - name: Get SHA2-SUMS
+      id: get_sha
+      run: |
+          echo "::set-output name=sha256_macos_legacy::$(sha256sum dist/yt-dlp_macos | awk '{print $1}')"
+          echo "::set-output name=sha512_macos_legacy::$(sha512sum dist/yt-dlp_macos | awk '{print $1}')"
+
+    - name: Upload standalone binary
+      uses: actions/upload-release-asset@v1
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
+        asset_path: ./dist/yt-dlp_macos
+        asset_name: yt-dlp_macos_legacy
+        asset_content_type: application/octet-stream
+
 
   build_windows:
     runs-on: windows-latest
-    needs: build_unix
+    needs: create_release
     outputs:
-      sha256_win: ${{ steps.sha256_win.outputs.sha256_win }}
-      sha512_win: ${{ steps.sha512_win.outputs.sha512_win }}
-      sha256_py2exe: ${{ steps.sha256_py2exe.outputs.sha256_py2exe }}
-      sha512_py2exe: ${{ steps.sha512_py2exe.outputs.sha512_py2exe }}
-      sha256_win_zip: ${{ steps.sha256_win_zip.outputs.sha256_win_zip }}
-      sha512_win_zip: ${{ steps.sha512_win_zip.outputs.sha512_win_zip }}
+      sha256_win: ${{ steps.get_sha.outputs.sha256_win }}
+      sha512_win: ${{ steps.get_sha.outputs.sha512_win }}
+      sha256_py2exe: ${{ steps.get_sha.outputs.sha256_py2exe }}
+      sha512_py2exe: ${{ steps.get_sha.outputs.sha512_py2exe }}
+      sha256_win_zip: ${{ steps.get_sha.outputs.sha256_win_zip }}
+      sha512_win_zip: ${{ steps.get_sha.outputs.sha512_win_zip }}
 
     steps:
     - uses: actions/checkout@v2
-    # 3.8 is used for Win7 support
-    - name: Set up Python 3.8
-      uses: actions/setup-python@v2
-      with:
+    - uses: actions/setup-python@v2
+      with:  # 3.8 is used for Win7 support
           python-version: '3.8'
     - name: Install Requirements
-      # Custom pyinstaller built with https://github.com/yt-dlp/pyinstaller-builds
-      run: |
+      run: |  # Custom pyinstaller built with https://github.com/yt-dlp/pyinstaller-builds
           python -m pip install --upgrade pip setuptools wheel py2exe
           pip install "https://yt-dlp.github.io/Pyinstaller-Builds/x86_64/pyinstaller-4.10-py3-none-any.whl" -r requirements.txt
-    - name: Bump version
-      id: bump_version
-      env:
-        version_suffix: ${{ needs.build_unix.outputs.version_suffix }}
-      run: python devscripts/update-version.py ${{ env.version_suffix }}
-    - name: Build lazy extractors
-      id: lazy_extractors
-      run: python devscripts/make_lazy_extractors.py
-    - name: Run PyInstaller Script
-      run: python pyinst.py
-    - name: Upload yt-dlp.exe Windows binary
-      id: upload-release-windows
+
+    - name: Prepare
+      run: |
+          python devscripts/update-version.py ${{ needs.create_release.outputs.version_suffix }}
+          python devscripts/make_lazy_extractors.py
+    - name: Build
+      run: |
+          python setup.py py2exe
+          Move-Item ./dist/yt-dlp.exe ./dist/yt-dlp_min.exe
+          python pyinst.py
+          python pyinst.py --onedir
+          Compress-Archive -Path ./dist/yt-dlp/* -DestinationPath ./dist/yt-dlp_win.zip
+    - name: Get SHA2-SUMS
+      id: get_sha
+      run: |
+          echo "::set-output name=sha256_py2exe::$((Get-FileHash dist\yt-dlp_min.exe -Algorithm SHA256).Hash.ToLower())"
+          echo "::set-output name=sha512_py2exe::$((Get-FileHash dist\yt-dlp_min.exe -Algorithm SHA512).Hash.ToLower())"
+          echo "::set-output name=sha256_win::$((Get-FileHash dist\yt-dlp.exe -Algorithm SHA256).Hash.ToLower())"
+          echo "::set-output name=sha512_win::$((Get-FileHash dist\yt-dlp.exe -Algorithm SHA512).Hash.ToLower())"
+          echo "::set-output name=sha256_win_zip::$((Get-FileHash dist\yt-dlp_win.zip -Algorithm SHA256).Hash.ToLower())"
+          echo "::set-output name=sha512_win_zip::$((Get-FileHash dist\yt-dlp_win.zip -Algorithm SHA512).Hash.ToLower())"
+
+    - name: Upload py2exe binary
       uses: actions/upload-release-asset@v1
       env:
         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
       with:
-        upload_url: ${{ needs.build_unix.outputs.upload_url }}
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
+        asset_path: ./dist/yt-dlp_min.exe
+        asset_name: yt-dlp_min.exe
+        asset_content_type: application/vnd.microsoft.portable-executable
+    - name: Upload standalone binary
+      uses: actions/upload-release-asset@v1
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
         asset_path: ./dist/yt-dlp.exe
         asset_name: yt-dlp.exe
         asset_content_type: application/vnd.microsoft.portable-executable
-    - name: Get SHA2-256SUMS for yt-dlp.exe
-      id: sha256_win
-      run: echo "::set-output name=sha256_win::$((Get-FileHash dist\yt-dlp.exe -Algorithm SHA256).Hash.ToLower())"
-    - name: Get SHA2-512SUMS for yt-dlp.exe
-      id: sha512_win
-      run: echo "::set-output name=sha512_win::$((Get-FileHash dist\yt-dlp.exe -Algorithm SHA512).Hash.ToLower())"
-
-    - name: Run PyInstaller Script with --onedir
-      run: |
-        python pyinst.py --onedir
-        Compress-Archive -LiteralPath ./dist/yt-dlp -DestinationPath ./dist/yt-dlp_win.zip
-    - name: Upload yt-dlp Windows onedir
-      id: upload-release-windows-zip
+    - name: Upload onedir binary
       uses: actions/upload-release-asset@v1
       env:
         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
       with:
-        upload_url: ${{ needs.build_unix.outputs.upload_url }}
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
         asset_path: ./dist/yt-dlp_win.zip
         asset_name: yt-dlp_win.zip
         asset_content_type: application/zip
-    - name: Get SHA2-256SUMS for yt-dlp_win.zip
-      id: sha256_win_zip
-      run: echo "::set-output name=sha256_win_zip::$((Get-FileHash dist\yt-dlp_win.zip -Algorithm SHA256).Hash.ToLower())"
-    - name: Get SHA2-512SUMS for yt-dlp_win.zip
-      id: sha512_win_zip
-      run: echo "::set-output name=sha512_win_zip::$((Get-FileHash dist\yt-dlp_win.zip -Algorithm SHA512).Hash.ToLower())"
 
-    - name: Run py2exe Script
-      run: python setup.py py2exe
-    - name: Upload yt-dlp_min.exe Windows binary
-      id: upload-release-windows-py2exe
-      uses: actions/upload-release-asset@v1
-      env:
-        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-      with:
-        upload_url: ${{ needs.build_unix.outputs.upload_url }}
-        asset_path: ./dist/yt-dlp.exe
-        asset_name: yt-dlp_min.exe
-        asset_content_type: application/vnd.microsoft.portable-executable
-    - name: Get SHA2-256SUMS for yt-dlp_min.exe
-      id: sha256_py2exe
-      run: echo "::set-output name=sha256_py2exe::$((Get-FileHash dist\yt-dlp.exe -Algorithm SHA256).Hash.ToLower())"
-    - name: Get SHA2-512SUMS for yt-dlp_min.exe
-      id: sha512_py2exe
-      run: echo "::set-output name=sha512_py2exe::$((Get-FileHash dist\yt-dlp.exe -Algorithm SHA512).Hash.ToLower())"
 
   build_windows32:
     runs-on: windows-latest
-    needs: build_unix
-
+    needs: create_release
     outputs:
-      sha256_win32: ${{ steps.sha256_win32.outputs.sha256_win32 }}
-      sha512_win32: ${{ steps.sha512_win32.outputs.sha512_win32 }}
+      sha256_win32: ${{ steps.get_sha.outputs.sha256_win32 }}
+      sha512_win32: ${{ steps.get_sha.outputs.sha512_win32 }}
 
     steps:
     - uses: actions/checkout@v2
-    # 3.7 is used for Vista support. See https://github.com/yt-dlp/yt-dlp/issues/390
-    - name: Set up Python 3.7 32-Bit
-      uses: actions/setup-python@v2
-      with:
+    - uses: actions/setup-python@v2
+      with:  # 3.7 is used for Vista support. See https://github.com/yt-dlp/yt-dlp/issues/390
           python-version: '3.7'
           architecture: 'x86'
     - name: Install Requirements
       run: |
           python -m pip install --upgrade pip setuptools wheel
           pip install "https://yt-dlp.github.io/Pyinstaller-Builds/i686/pyinstaller-4.10-py3-none-any.whl" -r requirements.txt
-    - name: Bump version
-      id: bump_version
-      env:
-        version_suffix: ${{ needs.build_unix.outputs.version_suffix }}
-      run: python devscripts/update-version.py ${{ env.version_suffix }}
-    - name: Build lazy extractors
-      id: lazy_extractors
-      run: python devscripts/make_lazy_extractors.py
-    - name: Run PyInstaller Script for 32 Bit
-      run: python pyinst.py
-    - name: Upload Executable yt-dlp_x86.exe
-      id: upload-release-windows32
+
+    - name: Prepare
+      run: |
+          python devscripts/update-version.py ${{ needs.create_release.outputs.version_suffix }}
+          python devscripts/make_lazy_extractors.py
+    - name: Build
+      run: |
+          python pyinst.py
+    - name: Get SHA2-SUMS
+      id: get_sha
+      run: |
+          echo "::set-output name=sha256_win32::$((Get-FileHash dist\yt-dlp_x86.exe -Algorithm SHA256).Hash.ToLower())"
+          echo "::set-output name=sha512_win32::$((Get-FileHash dist\yt-dlp_x86.exe -Algorithm SHA512).Hash.ToLower())"
+
+    - name: Upload standalone binary
       uses: actions/upload-release-asset@v1
       env:
         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
       with:
-        upload_url: ${{ needs.build_unix.outputs.upload_url }}
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
         asset_path: ./dist/yt-dlp_x86.exe
         asset_name: yt-dlp_x86.exe
         asset_content_type: application/vnd.microsoft.portable-executable
-    - name: Get SHA2-256SUMS for yt-dlp_x86.exe
-      id: sha256_win32
-      run: echo "::set-output name=sha256_win32::$((Get-FileHash dist\yt-dlp_x86.exe -Algorithm SHA256).Hash.ToLower())"
-    - name: Get SHA2-512SUMS for yt-dlp_x86.exe
-      id: sha512_win32
-      run: echo "::set-output name=sha512_win32::$((Get-FileHash dist\yt-dlp_x86.exe -Algorithm SHA512).Hash.ToLower())"
+
 
   finish:
     runs-on: ubuntu-latest
-    needs: [build_unix, build_windows, build_windows32, build_macos]
+    needs: [create_release, build_unix, build_windows, build_windows32, build_macos, build_macos_legacy]
 
     steps:
-    - name: Make SHA2-256SUMS file
-      env:
-        SHA256_BIN: ${{ needs.build_unix.outputs.sha256_bin }}
-        SHA256_TAR: ${{ needs.build_unix.outputs.sha256_tar }}
-        SHA256_WIN: ${{ needs.build_windows.outputs.sha256_win }}
-        SHA256_PY2EXE: ${{ needs.build_windows.outputs.sha256_py2exe }}
-        SHA256_WIN_ZIP: ${{ needs.build_windows.outputs.sha256_win_zip }}
-        SHA256_WIN32: ${{ needs.build_windows32.outputs.sha256_win32 }}
-        SHA256_MACOS: ${{ needs.build_macos.outputs.sha256_macos }}
-        SHA256_MACOS_ZIP: ${{ needs.build_macos.outputs.sha256_macos_zip }}
+    - name: Make SHA2-SUMS files
       run: |
-        echo "${{ env.SHA256_BIN }}  yt-dlp" >> SHA2-256SUMS
-        echo "${{ env.SHA256_TAR }}  yt-dlp.tar.gz" >> SHA2-256SUMS
-        echo "${{ env.SHA256_WIN }}  yt-dlp.exe" >> SHA2-256SUMS
-        echo "${{ env.SHA256_PY2EXE }}  yt-dlp_min.exe" >> SHA2-256SUMS
-        echo "${{ env.SHA256_WIN32 }}  yt-dlp_x86.exe" >> SHA2-256SUMS
-        echo "${{ env.SHA256_WIN_ZIP }}  yt-dlp_win.zip" >> SHA2-256SUMS
-        echo "${{ env.SHA256_MACOS }}  yt-dlp_macos" >> SHA2-256SUMS
-        echo "${{ env.SHA256_MACOS_ZIP }}  yt-dlp_macos.zip" >> SHA2-256SUMS
-    - name: Upload 256SUMS file
-      id: upload-sums
+        echo "${{ needs.build_unix.outputs.sha256_bin }}  yt-dlp" >> SHA2-256SUMS
+        echo "${{ needs.build_unix.outputs.sha256_tar }}  yt-dlp.tar.gz" >> SHA2-256SUMS
+        echo "${{ needs.build_unix.outputs.sha256_linux }}  yt-dlp_linux" >> SHA2-256SUMS
+        echo "${{ needs.build_unix.outputs.sha256_linux_zip }}  yt-dlp_linux.zip" >> SHA2-256SUMS
+        echo "${{ needs.build_windows.outputs.sha256_win }}  yt-dlp.exe" >> SHA2-256SUMS
+        echo "${{ needs.build_windows.outputs.sha256_py2exe }}  yt-dlp_min.exe" >> SHA2-256SUMS
+        echo "${{ needs.build_windows32.outputs.sha256_win32 }}  yt-dlp_x86.exe" >> SHA2-256SUMS
+        echo "${{ needs.build_windows.outputs.sha256_win_zip }}  yt-dlp_win.zip" >> SHA2-256SUMS
+        echo "${{ needs.build_macos.outputs.sha256_macos }}  yt-dlp_macos" >> SHA2-256SUMS
+        echo "${{ needs.build_macos.outputs.sha256_macos_zip }}  yt-dlp_macos.zip" >> SHA2-256SUMS
+        echo "${{ needs.build_macos_legacy.outputs.sha256_macos_legacy }}  yt-dlp_macos_legacy" >> SHA2-256SUMS
+        echo "${{ needs.build_unix.outputs.sha512_bin }}  yt-dlp" >> SHA2-512SUMS
+        echo "${{ needs.build_unix.outputs.sha512_tar }}  yt-dlp.tar.gz" >> SHA2-512SUMS
+        echo "${{ needs.build_unix.outputs.sha512_linux }}  yt-dlp_linux" >> SHA2-512SUMS
+        echo "${{ needs.build_unix.outputs.sha512_linux_zip }}  yt-dlp_linux.zip" >> SHA2-512SUMS
+        echo "${{ needs.build_windows.outputs.sha512_win }}  yt-dlp.exe" >> SHA2-512SUMS
+        echo "${{ needs.build_windows.outputs.sha512_py2exe }}  yt-dlp_min.exe" >> SHA2-512SUMS
+        echo "${{ needs.build_windows32.outputs.sha512_win32 }}  yt-dlp_x86.exe" >> SHA2-512SUMS
+        echo "${{ needs.build_windows.outputs.sha512_win_zip }}  yt-dlp_win.zip" >> SHA2-512SUMS
+        echo "${{ needs.build_macos.outputs.sha512_macos }}  yt-dlp_macos" >> SHA2-512SUMS
+        echo "${{ needs.build_macos.outputs.sha512_macos_zip }}  yt-dlp_macos.zip" >> SHA2-512SUMS
+        echo "${{ needs.build_macos_legacy.outputs.sha512_macos_legacy }}  yt-dlp_macos_legacy" >> SHA2-512SUMS
+
+    - name: Upload SHA2-256SUMS file
       uses: actions/upload-release-asset@v1
       env:
         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
       with:
-        upload_url: ${{ needs.build_unix.outputs.upload_url }}
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
         asset_path: ./SHA2-256SUMS
         asset_name: SHA2-256SUMS
         asset_content_type: text/plain
-    - name: Make SHA2-512SUMS file
-      env:
-        SHA512_BIN: ${{ needs.build_unix.outputs.sha512_bin }}
-        SHA512_TAR: ${{ needs.build_unix.outputs.sha512_tar }}
-        SHA512_WIN: ${{ needs.build_windows.outputs.sha512_win }}
-        SHA512_PY2EXE: ${{ needs.build_windows.outputs.sha512_py2exe }}
-        SHA512_WIN_ZIP: ${{ needs.build_windows.outputs.sha512_win_zip }}
-        SHA512_WIN32: ${{ needs.build_windows32.outputs.sha512_win32 }}
-        SHA512_MACOS: ${{ needs.build_macos.outputs.sha512_macos }}
-        SHA512_MACOS_ZIP: ${{ needs.build_macos.outputs.sha512_macos_zip }}
-      run: |
-        echo "${{ env.SHA512_BIN }}  yt-dlp" >> SHA2-512SUMS
-        echo "${{ env.SHA512_TAR }}  yt-dlp.tar.gz" >> SHA2-512SUMS
-        echo "${{ env.SHA512_WIN }}  yt-dlp.exe" >> SHA2-512SUMS
-        echo "${{ env.SHA512_WIN_ZIP }}  yt-dlp_win.zip" >> SHA2-512SUMS
-        echo "${{ env.SHA512_PY2EXE }}  yt-dlp_min.exe" >> SHA2-512SUMS
-        echo "${{ env.SHA512_WIN32 }}  yt-dlp_x86.exe" >> SHA2-512SUMS
-        echo "${{ env.SHA512_MACOS }}  yt-dlp_macos" >> SHA2-512SUMS
-        echo "${{ env.SHA512_MACOS_ZIP }}  yt-dlp_macos.zip" >> SHA2-512SUMS
-    - name: Upload 512SUMS file
-      id: upload-512sums
+    - name: Upload SHA2-512SUMS file
       uses: actions/upload-release-asset@v1
       env:
         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
       with:
-        upload_url: ${{ needs.build_unix.outputs.upload_url }}
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
         asset_path: ./SHA2-512SUMS
         asset_name: SHA2-512SUMS
         asset_content_type: text/plain
+
+    - name: Make Update spec
+      run: |
+        echo "# This file is used for regulating self-update" >> _update_spec
+    - name: Upload update spec
+      uses: actions/upload-release-asset@v1
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
+        asset_path: ./_update_spec
+        asset_name: _update_spec
+        asset_content_type: text/plain
+
+    - name: Finalize release
+      env:
+        GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      run: |
+        gh api -X PATCH -H "Accept: application/vnd.github.v3+json" \
+          /repos/${{ github.repository }}/releases/${{ needs.create_release.outputs.release_id }} \
+          -F draft=false

.github/workflows/core.yml

@@ -10,12 +10,15 @@ jobs:
       matrix:
         os: [ubuntu-latest]
         # CPython 3.9 is in quick-test
-        python-version: ['3.6', '3.7', '3.10', 3.11-dev, pypy-3.6, pypy-3.7, pypy-3.8, pypy-3.9]
+        python-version: ['3.6', '3.7', '3.10', 3.11-dev, pypy-3.6, pypy-3.7, pypy-3.8]
         run-tests-ext: [sh]
         include:
-        # atleast one of the tests must be in windows
+        # atleast one of each CPython/PyPy tests must be in windows
         - os: windows-latest
-          python-version: 3.8
+          python-version: '3.8'
+          run-tests-ext: bat
+        - os: windows-latest
+          python-version: pypy-3.9
           run-tests-ext: bat
     steps:
     - uses: actions/checkout@v2

.github/workflows/download.yml

@@ -9,11 +9,15 @@ jobs:
       fail-fast: true
       matrix:
         os: [ubuntu-latest]
-        python-version: ['3.6', '3.7', '3.9', '3.10', 3.11-dev, pypy-3.6, pypy-3.7, pypy-3.8, pypy-3.9]
+        python-version: ['3.6', '3.7', '3.9', '3.10', 3.11-dev, pypy-3.6, pypy-3.7, pypy-3.8]
         run-tests-ext: [sh]
         include:
+        # atleast one of each CPython/PyPy tests must be in windows
         - os: windows-latest
-          python-version: 3.8
+          python-version: '3.8'
+          run-tests-ext: bat
+        - os: windows-latest
+          python-version: pypy-3.9
           run-tests-ext: bat
     steps:
     - uses: actions/checkout@v2

CONTRIBUTING.md

@@ -214,7 +214,7 @@ After you have ensured this site is distributing its content legally, you can fo
                 # TODO more properties (see yt_dlp/extractor/common.py)
             }
     ```
-1. Add an import in [`yt_dlp/extractor/extractors.py`](yt_dlp/extractor/extractors.py).
+1. Add an import in [`yt_dlp/extractor/_extractors.py`](yt_dlp/extractor/_extractors.py). Note that the class name must end with `IE`.
 1. Run `python test/test_download.py TestDownload.test_YourExtractor` (note that `YourExtractor` doesn't end with `IE`). This *should fail* at first, but you can continually re-run it until you're done. If you decide to add more than one test, the tests will then be named `TestDownload.test_YourExtractor`, `TestDownload.test_YourExtractor_1`, `TestDownload.test_YourExtractor_2`, etc. Note that tests with `only_matching` key in test's dict are not counted in. You can also run all the tests in one go with `TestDownload.test_YourExtractor_all`
 1. Make sure you have atleast one test for your extractor. Even if all videos covered by the extractor are expected to be inaccessible for automated testing, tests should still be added with a `skip` parameter indicating why the particular test is disabled from running.
 1. Have a look at [`yt_dlp/extractor/common.py`](yt_dlp/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should and may return](yt_dlp/extractor/common.py#L91-L426). Add tests and code for as many as you want.
@@ -225,7 +225,7 @@ After you have ensured this site is distributing its content legally, you can fo
 1. Make sure your code works under all [Python](https://www.python.org/) versions supported by yt-dlp, namely CPython and PyPy for Python 3.6 and above. Backward compatibility is not required for even older versions of Python.
 1. When the tests pass, [add](https://git-scm.com/docs/git-add) the new files, [commit](https://git-scm.com/docs/git-commit) them and [push](https://git-scm.com/docs/git-push) the result, like this:
 
-        $ git add yt_dlp/extractor/extractors.py
+        $ git add yt_dlp/extractor/_extractors.py
         $ git add yt_dlp/extractor/yourextractor.py
         $ git commit -m '[yourextractor] Add extractor'
         $ git push origin yourextractor
@@ -300,14 +300,10 @@ description = meta['summary']  # incorrect
 The latter will break extraction process with `KeyError` if `summary` disappears from `meta` at some later time but with the former approach extraction will just go ahead with `description` set to `None` which is perfectly fine (remember `None` is equivalent to the absence of data).
 
 
-If the data is nested, do not use `.get` chains, but instead make use of the utility functions `try_get` or `traverse_obj`
+If the data is nested, do not use `.get` chains, but instead make use of `traverse_obj`.
 
 Considering the above `meta` again, assume you want to extract `["user"]["name"]` and put it in the resulting info dict as `uploader`
 
-```python
-uploader = try_get(meta, lambda x: x['user']['name'])  # correct
-```
-or
 ```python
 uploader = traverse_obj(meta, ('user', 'name'))  # correct
 ```
@@ -321,6 +317,10 @@ or
 ```python
 uploader = meta.get('user', {}).get('name')  # incorrect
 ```
+or
+```python
+uploader = try_get(meta, lambda x: x['user']['name'])  # old utility
+```
 
 
 Similarly, you should pass `fatal=False` when extracting optional data from a webpage with `_search_regex`, `_html_search_regex` or similar methods, for instance:
@@ -346,25 +346,25 @@ On failure this code will silently continue the extraction with `description` se
 
 Another thing to remember is not to try to iterate over `None`
 
-Say you extracted a list of thumbnails into `thumbnail_data` using `try_get` and now want to iterate over them
+Say you extracted a list of thumbnails into `thumbnail_data` and want to iterate over them
 
 ```python
-thumbnail_data = try_get(...)
+thumbnail_data = data.get('thumbnails') or []
 thumbnails = [{
     'url': item['url']
-} for item in thumbnail_data or []]  # correct
+} for item in thumbnail_data]  # correct
 ```
 
 and not like:
 
 ```python
-thumbnail_data = try_get(...)
+thumbnail_data = data.get('thumbnails')
 thumbnails = [{
     'url': item['url']
 } for item in thumbnail_data]  # incorrect
 ```
 
-In the later case, `thumbnail_data` will be `None` if the field was not found and this will cause the loop `for item in thumbnail_data` to raise a fatal error. Using `for item in thumbnail_data or []` avoids this error and results in setting an empty list in `thumbnails` instead.
+In this case, `thumbnail_data` will be `None` if the field was not found and this will cause the loop `for item in thumbnail_data` to raise a fatal error. Using `or []` avoids this error and results in setting an empty list in `thumbnails` instead.
 
 
 ### Provide fallbacks
@@ -431,7 +431,7 @@ title = self._search_regex(  # correct
     r'<span[^>]+class="title"[^>]*>([^<]+)', webpage, 'title')
 ```
 
-Or even better:
+which tolerates potential changes in the `style` attribute's value. Or even better:
 
 ```python
 title = self._search_regex(  # correct
@@ -439,7 +439,7 @@ title = self._search_regex(  # correct
     webpage, 'title', group='title')
 ```
 
-Note how you tolerate potential changes in the `style` attribute's value or switch from using double quotes to single for `class` attribute: 
+which also handles both single quotes in addition to double quotes.
 
 The code definitely should not look like:
 
@@ -457,7 +457,42 @@ title = self._search_regex(  # incorrect
     webpage, 'title', group='title')
 ```
 
-Here the presence or absence of other attributes including `style` is irrelevent for the data we need, and so the regex must not depend on it
+Here the presence or absence of other attributes including `style` is irrelevant for the data we need, and so the regex must not depend on it
+
+
+#### Keep the regular expressions as simple as possible, but no simpler
+
+Since many extractors deal with unstructured data provided by websites, we will often need to use very complex regular expressions. You should try to use the *simplest* regex that can accomplish what you want. In other words, each part of the regex must have a reason for existing. If you can take out a symbol and the functionality does not change, the symbol should not be there.
+
+##### Example
+
+Correct:
+
+```python
+_VALID_URL = r'https?://(?:www\.)?website\.com/(?:[^/]+/){3,4}(?P<display_id>[^/]+)_(?P<id>\d+)'
+```
+
+Incorrect:
+
+```python
+_VALID_URL = r'https?:\/\/(?:www\.)?website\.com\/[^\/]+/[^\/]+/[^\/]+(?:\/[^\/]+)?\/(?P<display_id>[^\/]+)_(?P<id>\d+)'
+```
+
+#### Do not misuse `.` and use the correct quantifiers (`+*?`)
+
+Avoid creating regexes that over-match because of wrong use of quantifiers. Also try to avoid non-greedy matching (`?`) where possible since they could easily result in [catastrophic backtracking](https://www.regular-expressions.info/catastrophic.html)
+
+Correct:
+
+```python
+title = self._search_regex(r'<span\b[^>]+class="title"[^>]*>([^<]+)', webpage, 'title')
+```
+
+Incorrect:
+
+```python
+title = self._search_regex(r'<span\b.*class="title".*>(.+?)<', webpage, 'title')
+```
 
 
 ### Long lines policy
@@ -466,7 +501,7 @@ There is a soft limit to keep lines of code under 100 characters long. This mean
 
 For example, you should **never** split long string literals like URLs or some other often copied entities over multiple lines to fit this limit:
 
-Conversely, don't unecessarily split small lines further. As a rule of thumb, if removing the line split keeps the code under 80 characters, it should be a single line.
+Conversely, don't unnecessarily split small lines further. As a rule of thumb, if removing the line split keeps the code under 80 characters, it should be a single line.
 
 ##### Examples
 
@@ -521,19 +556,22 @@ formats = self._extract_m3u8_formats(m3u8_url,
 
 ### Quotes
 
-Always use single quotes for strings (even if the string has `'`) and double quotes for docstrings. Use `'''` only for multi-line strings. An exception can be made if a string has multiple single quotes in it and escaping makes it significantly harder to read. For f-strings, use you can use double quotes on the inside. But avoid f-strings that have too many quotes inside.
+Always use single quotes for strings (even if the string has `'`) and double quotes for docstrings. Use `'''` only for multi-line strings. An exception can be made if a string has multiple single quotes in it and escaping makes it *significantly* harder to read. For f-strings, use you can use double quotes on the inside. But avoid f-strings that have too many quotes inside.
 
 
 ### Inline values
 
 Extracting variables is acceptable for reducing code duplication and improving readability of complex expressions. However, you should avoid extracting variables used only once and moving them to opposite parts of the extractor file, which makes reading the linear flow difficult.
 
-#### Example
+#### Examples
 
 Correct:
 
 ```python
-title = self._html_search_regex(r'<h1>([^<]+)</h1>', webpage, 'title')
+return {
+    'title': self._html_search_regex(r'<h1>([^<]+)</h1>', webpage, 'title'),
+    # ...some lines of code...
+}
 ```
 
 Incorrect:
@@ -542,6 +580,11 @@ Incorrect:
 TITLE_RE = r'<h1>([^<]+)</h1>'
 # ...some lines of code...
 title = self._html_search_regex(TITLE_RE, webpage, 'title')
+# ...some lines of code...
+return {
+    'title': title,
+    # ...some lines of code...
+}
 ```
 
 
@@ -573,33 +616,32 @@ Methods supporting list of patterns are: `_search_regex`, `_html_search_regex`,
 
 ### Trailing parentheses
 
-Always move trailing parentheses used for grouping/functions after the last argument. On the other hand, literal list/tuple/dict/set should closed be in a new line. Generators and list/dict comprehensions may use either style
+Always move trailing parentheses used for grouping/functions after the last argument. On the other hand, multi-line literal list/tuple/dict/set should closed be in a new line. Generators and list/dict comprehensions may use either style
 
 #### Examples
 
 Correct:
 
 ```python
-url = try_get(
-    info,
-    lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
-    list)
+url = traverse_obj(info, (
+    'context', 'dispatcher', 'stores', 'VideoTitlePageStore', 'data', 'video', 0, 'VideoUrlSet', 'VideoUrl'), list)
 ```
 Correct:
 
 ```python
-url = try_get(info,
-              lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
-              list)
+url = traverse_obj(
+    info,
+    ('context', 'dispatcher', 'stores', 'VideoTitlePageStore', 'data', 'video', 0, 'VideoUrlSet', 'VideoUrl'),
+    list)
 ```
 
 Incorrect:
 
 ```python
-url = try_get(
+url = traverse_obj(
     info,
-    lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
-    list,
+    ('context', 'dispatcher', 'stores', 'VideoTitlePageStore', 'data', 'video', 0, 'VideoUrlSet', 'VideoUrl'),
+    list
 )
 ```
 
@@ -648,21 +690,17 @@ Use `unified_strdate` for uniform `upload_date` or any `YYYYMMDD` meta field ext
 
 Explore [`yt_dlp/utils.py`](yt_dlp/utils.py) for more useful convenience functions.
 
-#### More examples
+#### Examples
 
-##### Safely extract optional description from parsed JSON
 ```python
 description = traverse_obj(response, ('result', 'video', 'summary'), expected_type=str)
-```
-
-##### Safely extract more optional metadata
-```python
+thumbnails = traverse_obj(response, ('result', 'thumbnails', ..., 'url'), expected_type=url_or_none)
 video = traverse_obj(response, ('result', 'video', 0), default={}, expected_type=dict)
-description = video.get('summary')
 duration = float_or_none(video.get('durationMs'), scale=1000)
 view_count = int_or_none(video.get('views'))
 ```
 
+
 # My pull request is labeled pending-fixes
 
 The `pending-fixes` label is added when there are changes requested to a PR. When the necessary changes are made, the label should be removed. However, despite our best efforts, it may sometimes happen that the maintainer did not see the changes or forgot to remove the label. If your PR is still marked as `pending-fixes` a few days after all requested changes have been made, feel free to ping the maintainer who labeled your issue and ask them to re-review and remove the label.

CONTRIBUTORS

@@ -248,3 +248,27 @@ rand-net
 vertan
 Wikidepia
 Yipten
+moench-tegeder
+christoph-heinrich
+HobbyistDev
+LunarFang416
+sbor23
+aurelg
+adamanldo
+gamer191
+vkorablin
+Burve
+mnn
+ZhymabekRoman
+mozbugbox
+aejdl
+ping
+sqrtNOT
+bubbleguuum
+darkxex
+miseran
+StefanLobbenmeier
+crazymoose77756
+nomevi
+Brett824
+pingiun

Changelog.md

@@ -11,6 +11,170 @@
 -->
 
 
+### 2022.06.29
+
+* Fix `--downloader native`
+* Fix `section_end` of clips
+* Fix playlist error handling
+* Sanitize `chapters`
+* [extractor] Fix `_create_request` when headers is None
+* [extractor] Fix empty `BaseURL` in MPD
+* [ffmpeg] Write full output to debug on error
+* [hls] Warn user when trying to download live HLS
+* [options] Fix `parse_known_args` for `--`
+* [utils] Fix inconsistent default handling between HTTP and HTTPS requests by [coletdjnz](https://github.com/coletdjnz)
+* [build] Draft release until complete
+* [build] Fix release tag commit
+* [build] Standalone x64 builds for MacOS 10.9 by [StefanLobbenmeier](https://github.com/StefanLobbenmeier)
+* [update] Ability to set a maximum version for specific variants
+* [compat] Fix `compat.WINDOWS_VT_MODE`
+* [compat] Remove deprecated functions from core code
+* [compat] Remove more functions
+* [cleanup, extractor] Reduce direct use of `_downloader`
+* [cleanup] Consistent style for file heads
+* [cleanup] Fix some typos by [crazymoose77756](https://github.com/crazymoose77756)
+* [cleanup] Misc fixes and cleanup
+* [extractor/Scrolller] Add extractor by [LunarFang416](https://github.com/LunarFang416)
+* [extractor/ViMP] Add playlist extractor by [FestplattenSchnitzel](https://github.com/FestplattenSchnitzel)
+* [extractor/fuyin] Add extractor by [HobbyistDev](https://github.com/HobbyistDev)
+* [extractor/livestreamfails] Add extractor by [nomevi](https://github.com/nomevi)
+* [extractor/premiershiprugby] Add extractor by [HobbyistDev](https://github.com/HobbyistDev)
+* [extractor/steam] Add broadcast extractor by [HobbyistDev](https://github.com/HobbyistDev)
+* [extractor/youtube] Mark videos as fully watched by [Brett824](https://github.com/Brett824)
+* [extractor/CWTV] Extract thumbnail by [ischmidt20](https://github.com/ischmidt20)
+* [extractor/ViMP] Add thumbnail and support more sites by [FestplattenSchnitzel](https://github.com/FestplattenSchnitzel)
+* [extractor/dropout] Support cookies and login only as needed by [pingiun](https://github.com/pingiun), [pukkandan](https://github.com/pukkandan)
+* [extractor/ertflix] Improve `_VALID_URL`
+* [extractor/lbry] Use HEAD request for redirect URL by [flashdagger](https://github.com/flashdagger)
+* [extractor/mediaset] Improve `_VALID_URL`
+* [extractor/npr] Implement [e50c350](https://github.com/yt-dlp/yt-dlp/commit/e50c3500b43d80e4492569c4b4523c4379c6fbb2) differently
+* [extractor/tennistv] Rewrite extractor by [pukkandan](https://github.com/pukkandan), [zenerdi0de](https://github.com/zenerdi0de)
+
+### 2022.06.22.1
+
+* [build] Fix updating homebrew formula
+
+### 2022.06.22
+
+* [**Deprecate support for Python 3.6**](https://github.com/yt-dlp/yt-dlp/issues/3764#issuecomment-1154051119)
+* **Add option `--download-sections` to download video partially**
+    * Chapter regex and time ranges are accepted (Eg: `--download-sections *1:10-2:20`)
+* Add option `--alias`
+* Add option `--lazy-playlist` to process entries as they are received
+* Add option `--retry-sleep`
+* Add slicing notation to `--playlist-items`
+    * Adds support for negative indices and step
+    * Add `-I` as alias for `--playlist-index`
+    * Makes `--playlist-start`, `--playlist-end`, `--playlist-reverse`, `--no-playlist-reverse` redundant
+* `--config-location -` to provide options interactively
+* [build] Add Linux standalone builds
+* [update] Self-restart after update
+* Merge youtube-dl: Upto [commit/8a158a9](https://github.com/ytdl-org/youtube-dl/commit/8a158a9)
+* Add `--no-update`
+* Allow extractors to specify section_start/end for clips
+* Do not print progress to `stderr` with `-q`
+* Ensure pre-processor errors do not block video download
+* Fix `--simulate --max-downloads`
+* Improve error handling of bad config files
+* Return an error code if update fails
+* Fix bug in [3a408f9](https://github.com/yt-dlp/yt-dlp/commit/3a408f9d199127ca2626359e21a866a09ab236b3)
+* [ExtractAudio] Allow conditional conversion
+* [ModifyChapters] Fix repeated removal of small segments
+* [ThumbnailsConvertor] Allow conditional conversion
+* [cookies] Detect profiles for cygwin/BSD by [moench-tegeder](https://github.com/moench-tegeder)
+* [dash] Show fragment count with `--live-from-start` by [flashdagger](https://github.com/flashdagger)
+* [extractor] Add `_search_json` by [coletdjnz](https://github.com/coletdjnz), [pukkandan](https://github.com/pukkandan)
+* [extractor] Add `default` parameter to `_search_json` by [coletdjnz](https://github.com/coletdjnz), [pukkandan](https://github.com/pukkandan)
+* [extractor] Add dev option `--load-pages`
+* [extractor] Handle `json_ld` with multiple `@type`s
+* [extractor] Import `_ALL_CLASSES` lazily
+* [extractor] Recognize `src` attribute from HTML5 media elements by [Lesmiscore](https://github.com/Lesmiscore)
+* [extractor/generic] Revert e6ae51c123897927eb3c9899923d8ffd31c7f85d
+* [f4m] Bugfix
+* [ffmpeg] Check version lazily
+* [jsinterp] Some optimizations and refactoring by [dirkf](https://github.com/dirkf), [pukkandan](https://github.com/pukkandan)
+* [utils] Improve performance using `functools.cache`
+* [utils] Send HTTP/1.1 ALPN extension by [coletdjnz](https://github.com/coletdjnz)
+* [utils] `ExtractorError`: Fix `exc_info`
+* [utils] `ISO3166Utils`: Add `EU` and `AP`
+* [utils] `Popen`: Refactor to use contextmanager
+* [utils] `locked_file`: Fix for PyPy on Windows
+* [update] Expose more functionality to API
+* [update] Use `.git` folder to distinguish `source`/`unknown`
+* [compat] Add `functools.cached_property`
+* [test] Fix `FakeYDL` signatures by [coletdjnz](https://github.com/coletdjnz)
+* [docs] Improvements
+* [cleanup, ExtractAudio] Refactor
+* [cleanup, downloader] Refactor `report_progress`
+* [cleanup, extractor] Refactor `_download_...` methods
+* [cleanup, extractor] Rename `extractors.py` to `_extractors.py`
+* [cleanup, utils] Don't use kwargs for `format_field`
+* [cleanup, build] Refactor
+* [cleanup, docs] Re-indent "Usage and Options" section
+* [cleanup] Deprecate `YoutubeDL.parse_outtmpl`
+* [cleanup] Misc fixes and cleanup by [Lesmiscore](https://github.com/Lesmiscore), [MrRawes](https://github.com/MrRawes), [christoph-heinrich](https://github.com/christoph-heinrich), [flashdagger](https://github.com/flashdagger), [gamer191](https://github.com/gamer191), [kwconder](https://github.com/kwconder), [pukkandan](https://github.com/pukkandan)
+* [extractor/DailyWire] Add extractors by [HobbyistDev](https://github.com/HobbyistDev), [pukkandan](https://github.com/pukkandan)
+* [extractor/fourzerostudio] Add extractors by [Lesmiscore](https://github.com/Lesmiscore)
+* [extractor/GoogleDrive] Add folder extractor by [evansp](https://github.com/evansp), [pukkandan](https://github.com/pukkandan)
+* [extractor/MirrorCoUK] Add extractor by [LunarFang416](https://github.com/LunarFang416), [pukkandan](https://github.com/pukkandan)
+* [extractor/atscaleconfevent] Add extractor by [Ashish0804](https://github.com/Ashish0804)
+* [extractor/freetv] Add extractor by [elyse0](https://github.com/elyse0)
+* [extractor/ixigua] Add Extractor by [HobbyistDev](https://github.com/HobbyistDev)
+* [extractor/kicker.de] Add extractor by [HobbyistDev](https://github.com/HobbyistDev)
+* [extractor/netverse] Add extractors by [HobbyistDev](https://github.com/HobbyistDev), [pukkandan](https://github.com/pukkandan)
+* [extractor/playsuisse] Add extractor by [pukkandan](https://github.com/pukkandan), [sbor23](https://github.com/sbor23)
+* [extractor/substack] Add extractor by [elyse0](https://github.com/elyse0)
+* [extractor/youtube] **Support downloading clips**
+* [extractor/youtube] Add `innertube_host` and `innertube_key` extractor args by [coletdjnz](https://github.com/coletdjnz)
+* [extractor/youtube] Add warning for PostLiveDvr
+* [extractor/youtube] Bring back `_extract_chapters_from_description`
+* [extractor/youtube] Extract `comment_count` from webpage
+* [extractor/youtube] Fix `:ytnotifications` extractor by [coletdjnz](https://github.com/coletdjnz)
+* [extractor/youtube] Fix initial player response extraction by [coletdjnz](https://github.com/coletdjnz), [pukkandan](https://github.com/pukkandan)
+* [extractor/youtube] Fix live chat for videos with content warning by [coletdjnz](https://github.com/coletdjnz)
+* [extractor/youtube] Make signature extraction non-fatal
+* [extractor/youtube:tab] Detect `videoRenderer` in `_post_thread_continuation_entries`
+* [extractor/BiliIntl] Fix metadata extraction
+* [extractor/BiliIntl] Fix subtitle extraction by [HobbyistDev](https://github.com/HobbyistDev)
+* [extractor/FranceCulture] Fix extractor by [aurelg](https://github.com/aurelg), [pukkandan](https://github.com/pukkandan)
+* [extractor/PokemonSoundLibrary] Remove extractor by [Lesmiscore](https://github.com/Lesmiscore)
+* [extractor/StreamCZ] Fix extractor by [adamanldo](https://github.com/adamanldo), [dirkf](https://github.com/dirkf)
+* [extractor/WatchESPN] Support free videos and BAM_DTC by [ischmidt20](https://github.com/ischmidt20)
+* [extractor/animelab] Remove extractor by [gamer191](https://github.com/gamer191)
+* [extractor/bloomberg] Change playback endpoint by [m4tu4g](https://github.com/m4tu4g)
+* [extractor/ccc] Extract view_count by [vkorablin](https://github.com/vkorablin)
+* [extractor/crunchyroll:beta] Fix extractor after API change by [Burve](https://github.com/Burve), [tejing1](https://github.com/tejing1)
+* [extractor/curiositystream] Get `auth_token` from cookie by [mnn](https://github.com/mnn)
+* [extractor/digitalconcerthall] Fix extractor by [ZhymabekRoman](https://github.com/ZhymabekRoman)
+* [extractor/dropbox] Extract the correct `mountComponent`
+* [extractor/dropout] Login is not mandatory
+* [extractor/duboku] Fix for hostname change by [mozbugbox](https://github.com/mozbugbox)
+* [extractor/espn] Add `WatchESPN` extractor by [ischmidt20](https://github.com/ischmidt20), [pukkandan](https://github.com/pukkandan)
+* [extractor/expressen] Fix extractor by [aejdl](https://github.com/aejdl)
+* [extractor/foxnews] Update embed extraction by [elyse0](https://github.com/elyse0)
+* [extractor/ina] Fix extractor by [elyse0](https://github.com/elyse0)
+* [extractor/iwara:user] Make paging better by [Lesmiscore](https://github.com/Lesmiscore)
+* [extractor/jwplatform] Look for `data-video-jw-id`
+* [extractor/lbry] Update livestream API by [flashdagger](https://github.com/flashdagger)
+* [extractor/mediaset] Improve `_VALID_URL`
+* [extractor/naver] Add `navernow` extractor by [ping](https://github.com/ping)
+* [extractor/niconico:series] Fix extractor by [sqrtNOT](https://github.com/sqrtNOT)
+* [extractor/npr] Use stream url from json-ld by [r5d](https://github.com/r5d)
+* [extractor/pornhub] Extract `uploader_id` field by [Lesmiscore](https://github.com/Lesmiscore)
+* [extractor/radiofrance] Add more radios by [bubbleguuum](https://github.com/bubbleguuum)
+* [extractor/rumble] Detect JS embed
+* [extractor/rumble] Extract subtitles by [fstirlitz](https://github.com/fstirlitz)
+* [extractor/southpark] Add `southpark.lat` extractor by [darkxex](https://github.com/darkxex)
+* [extractor/spotify:show] Fix extractor
+* [extractor/tiktok] Detect embeds
+* [extractor/tiktok] Extract `SIGI_STATE` by [dirkf](https://github.com/dirkf), [pukkandan](https://github.com/pukkandan), [sulyi](https://github.com/sulyi)
+* [extractor/tver] Fix extractor by [Lesmiscore](https://github.com/Lesmiscore)
+* [extractor/vevo] Fix extractor by [Lesmiscore](https://github.com/Lesmiscore)
+* [extractor/yahoo:gyao] Fix extractor
+* [extractor/zattoo] Fix live streams by [miseran](https://github.com/miseran)
+* [extractor/zdf] Improve format sorting by [elyse0](https://github.com/elyse0)
+
+
 ### 2022.05.18
 
 * Add support for SSL client certificate authentication by [coletdjnz](https://github.com/coletdjnz), [dirkf](https://github.com/dirkf)
@@ -419,7 +583,7 @@
 * [downloader/ffmpeg] Handle unknown formats better
 * [outtmpl] Handle `-o ""` better
 * [outtmpl] Handle hard-coded file extension better
-* [extractor] Add convinience function `_yes_playlist`
+* [extractor] Add convenience function `_yes_playlist`
 * [extractor] Allow non-fatal `title` extraction
 * [extractor] Extract video inside `Article` json_ld
 * [generic] Allow further processing of json_ld URL
@@ -1156,7 +1320,7 @@
 * [build] Automate more of the release process by [animelover1984](https://github.com/animelover1984), [pukkandan](https://github.com/pukkandan)
 * [build] Fix sha256 by [nihil-admirari](https://github.com/nihil-admirari)
 * [build] Bring back brew taps by [nao20010128nao](https://github.com/nao20010128nao)
-* [build] Provide `--onedir` zip for windows by [pukkandan](https://github.com/pukkandan)
+* [build] Provide `--onedir` zip for windows
 * [cleanup,docs] Add deprecation warning in docs for some counter intuitive behaviour
 * [cleanup] Fix line endings for `nebula.py` by [glenn-slayden](https://github.com/glenn-slayden)
 * [cleanup] Improve `make clean-test` by [sulyi](https://github.com/sulyi)
@@ -1553,7 +1717,7 @@
 * [utils] Generalize `traverse_dict` to `traverse_obj`
 * [downloader/ffmpeg] Hide FFmpeg banner unless in verbose mode by [fstirlitz](https://github.com/fstirlitz)
 * [build] Release `yt-dlp.tar.gz`
-* [build,update] Add GNU-style SHA512 and prepare updater for simlar SHA256 by [nihil-admirari](https://github.com/nihil-admirari)
+* [build,update] Add GNU-style SHA512 and prepare updater for similar SHA256 by [nihil-admirari](https://github.com/nihil-admirari)
 * [pyinst] Show Python version in exe metadata by [nihil-admirari](https://github.com/nihil-admirari)
 * [docs] Improve documentation of dependencies
 * [cleanup] Mark unused files

Makefile

@@ -9,7 +9,8 @@ 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/*
+pypi-files: AUTHORS Changelog.md LICENSE README.md README.txt supportedsites \
+	        completions yt-dlp.1 requirements.txt setup.cfg devscripts/* test/*
 
 .PHONY: all clean install test tar pypi-files completions ot offlinetest codetest supportedsites
 
@@ -42,7 +43,7 @@ PYTHON ?= /usr/bin/env python3
 SYSCONFDIR = $(shell if [ $(PREFIX) = /usr -o $(PREFIX) = /usr/local ]; then echo /etc; else echo $(PREFIX)/etc; fi)
 
 # 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)
+MARKDOWN = $(shell if [ `pandoc -v | head -n1 | cut -d" " -f2 | head -c1` = "2" ]; then echo markdown-smart; else echo markdown; fi)
 
 install: lazy-extractors yt-dlp yt-dlp.1 completions
 	mkdir -p $(DESTDIR)$(BINDIR)
@@ -91,10 +92,10 @@ yt-dlp: yt_dlp/*.py yt_dlp/*/*.py
 	rm yt-dlp.zip
 	chmod a+x yt-dlp
 
-README.md: yt_dlp/*.py yt_dlp/*/*.py
-	COLUMNS=80 $(PYTHON) yt_dlp/__main__.py --help | $(PYTHON) devscripts/make_readme.py
+README.md: yt_dlp/*.py yt_dlp/*/*.py devscripts/make_readme.py
+	COLUMNS=80 $(PYTHON) yt_dlp/__main__.py --ignore-config --help | $(PYTHON) devscripts/make_readme.py
 
-CONTRIBUTING.md: README.md
+CONTRIBUTING.md: README.md devscripts/make_contributing.py
 	$(PYTHON) devscripts/make_contributing.py README.md CONTRIBUTING.md
 
 issuetemplates: devscripts/make_issue_template.py .github/ISSUE_TEMPLATE_tmpl/1_broken_site.yml .github/ISSUE_TEMPLATE_tmpl/2_site_support_request.yml .github/ISSUE_TEMPLATE_tmpl/3_site_feature_request.yml .github/ISSUE_TEMPLATE_tmpl/4_bug_report.yml .github/ISSUE_TEMPLATE_tmpl/5_feature_request.yml yt_dlp/version.py
@@ -111,7 +112,7 @@ supportedsites:
 README.txt: README.md
 	pandoc -f $(MARKDOWN) -t plain README.md -o README.txt
 
-yt-dlp.1: README.md
+yt-dlp.1: README.md devscripts/prepare_manpage.py
 	$(PYTHON) devscripts/prepare_manpage.py yt-dlp.1.temp.md
 	pandoc -s -f $(MARKDOWN) -t man yt-dlp.1.temp.md -o yt-dlp.1
 	rm -f yt-dlp.1.temp.md
@@ -128,7 +129,7 @@ completions/fish/yt-dlp.fish: yt_dlp/*.py yt_dlp/*/*.py devscripts/fish-completi
 	mkdir -p completions/fish
 	$(PYTHON) devscripts/fish-completion.py
 
-_EXTRACTOR_FILES = $(shell find yt_dlp/extractor -iname '*.py' -and -not -iname 'lazy_extractors.py')
+_EXTRACTOR_FILES = $(shell find yt_dlp/extractor -name '*.py' -and -not -name '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 $@
 
@@ -147,7 +148,7 @@ yt-dlp.tar.gz: all
 		CONTRIBUTING.md Collaborators.md CONTRIBUTORS AUTHORS \
 		Makefile MANIFEST.in yt-dlp.1 README.txt completions \
 		setup.py setup.cfg yt-dlp yt_dlp requirements.txt \
-		devscripts test tox.ini pytest.ini
+		devscripts test
 
 AUTHORS: .mailmap
 	git shortlog -s -n | cut -f2 | sort > AUTHORS

devscripts/bash-completion.py

@@ -1,9 +1,12 @@
 #!/usr/bin/env python3
+
+# Allow direct execution
 import os
 import sys
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
 import yt_dlp
 
 BASH_COMPLETION_FILE = "completions/bash/yt-dlp"

devscripts/check-porn.py

@@ -13,9 +13,11 @@ import sys
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import gettestcases
 
-from yt_dlp.utils import compat_urllib_parse_urlparse, compat_urllib_request
+import urllib.parse
+import urllib.request
+
+from test.helper import gettestcases
 
 if len(sys.argv) > 1:
     METHOD = 'LIST'
@@ -26,7 +28,7 @@ else:
 for test in gettestcases():
     if METHOD == 'EURISTIC':
         try:
-            webpage = compat_urllib_request.urlopen(test['url'], timeout=10).read()
+            webpage = urllib.request.urlopen(test['url'], timeout=10).read()
         except Exception:
             print('\nFail: {}'.format(test['name']))
             continue
@@ -36,7 +38,7 @@ for test in gettestcases():
         RESULT = 'porn' in webpage.lower()
 
     elif METHOD == 'LIST':
-        domain = compat_urllib_parse_urlparse(test['url']).netloc
+        domain = urllib.parse.urlparse(test['url']).netloc
         if not domain:
             print('\nFail: {}'.format(test['name']))
             continue

devscripts/fish-completion.py

@@ -1,10 +1,14 @@
 #!/usr/bin/env python3
-import optparse
+
+# Allow direct execution
 import os
 import sys
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import optparse
+
 import yt_dlp
 from yt_dlp.utils import shell_quote
 

devscripts/generate_aes_testdata.py

@@ -1,11 +1,15 @@
 #!/usr/bin/env python3
-import codecs
+
+# Allow direct execution
 import os
-import subprocess
 import sys
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import codecs
+import subprocess
+
 from yt_dlp.aes import aes_encrypt, key_expansion
 from yt_dlp.utils import intlist_to_bytes
 

devscripts/make_contributing.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 import optparse
 import re
 

devscripts/make_issue_template.py

@@ -1,5 +1,12 @@
 #!/usr/bin/env python3
-import io
+
+# Allow direct execution
+import os
+import sys
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
 import optparse
 
 
@@ -8,7 +15,7 @@ def read(fname):
         return f.read()
 
 
-# Get the version from yt_dlp/version.py without importing the package
+# Get the version without importing the package
 def read_version(fname):
     exec(compile(read(fname), fname, 'exec'))
     return locals()['__version__']

devscripts/make_lazy_extractors.py

@@ -1,12 +1,15 @@
 #!/usr/bin/env python3
-import optparse
+
+# Allow direct execution
 import os
 import sys
-from inspect import getsource
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
+import optparse
+from inspect import getsource
+
 NO_ATTR = object()
 STATIC_CLASS_PROPERTIES = ['IE_NAME', 'IE_DESC', 'SEARCH_KEY', '_WORKING', '_NETRC_MACHINE', 'age_limit']
 CLASS_METHODS = [
@@ -53,7 +56,7 @@ def get_all_ies():
     if os.path.exists(PLUGINS_DIRNAME):
         os.rename(PLUGINS_DIRNAME, BLOCKED_DIRNAME)
     try:
-        from yt_dlp.extractor import _ALL_CLASSES
+        from yt_dlp.extractor.extractors import _ALL_CLASSES
     finally:
         if os.path.exists(BLOCKED_DIRNAME):
             os.rename(BLOCKED_DIRNAME, PLUGINS_DIRNAME)

devscripts/make_readme.py

@@ -1,7 +1,12 @@
 #!/usr/bin/env python3
 
-# yt-dlp --help | make_readme.py
-# This must be run in a console of correct width
+"""
+yt-dlp --help | make_readme.py
+This must be run in a console of correct width
+"""
+
+
+import functools
 import re
 import sys
 
@@ -10,21 +15,60 @@ README_FILE = 'README.md'
 OPTIONS_START = 'General Options:'
 OPTIONS_END = 'CONFIGURATION'
 EPILOG_START = 'See full documentation'
+ALLOWED_OVERSHOOT = 2
+
+DISABLE_PATCH = object()
 
 
-helptext = sys.stdin.read()
-if isinstance(helptext, bytes):
-    helptext = helptext.decode()
+def take_section(text, start=None, end=None, *, shift=0):
+    return text[
+        text.index(start) + shift if start else None:
+        text.index(end) + shift if end else None
+    ]
 
-start, end = helptext.index(f'\n  {OPTIONS_START}'), helptext.index(f'\n{EPILOG_START}')
-options = re.sub(r'(?m)^  (\w.+)$', r'## \1', helptext[start + 1: end + 1])
+
+def apply_patch(text, patch):
+    return text if patch[0] is DISABLE_PATCH else re.sub(*patch, text)
+
+
+options = take_section(sys.stdin.read(), f'\n  {OPTIONS_START}', f'\n{EPILOG_START}', shift=1)
+
+max_width = max(map(len, options.split('\n')))
+switch_col_width = len(re.search(r'(?m)^\s{5,}', options).group())
+delim = f'\n{" " * switch_col_width}'
+
+PATCHES = (
+    (  # Headings
+        r'(?m)^  (\w.+\n)(    (?=\w))?',
+        r'## \1'
+    ),
+    (  # Do not split URLs
+        rf'({delim[:-1]})? (?P<label>\[\S+\] )?(?P<url>https?({delim})?:({delim})?/({delim})?/(({delim})?\S+)+)\s',
+        lambda mobj: ''.join((delim, mobj.group('label') or '', re.sub(r'\s+', '', mobj.group('url')), '\n'))
+    ),
+    (  # Do not split "words"
+        rf'(?m)({delim}\S+)+$',
+        lambda mobj: ''.join((delim, mobj.group(0).replace(delim, '')))
+    ),
+    (  # Allow overshooting last line
+        rf'(?m)^(?P<prev>.+)${delim}(?P<current>.+)$(?!{delim})',
+        lambda mobj: (mobj.group().replace(delim, ' ')
+                      if len(mobj.group()) - len(delim) + 1 <= max_width + ALLOWED_OVERSHOOT
+                      else mobj.group())
+    ),
+    (  # Avoid newline when a space is available b/w switch and description
+        DISABLE_PATCH,  # This creates issues with prepare_manpage
+        r'(?m)^(\s{4}-.{%d})(%s)' % (switch_col_width - 6, delim),
+        r'\1 '
+    ),
+)
 
 with open(README_FILE, encoding='utf-8') as f:
     readme = f.read()
 
-header = readme[:readme.index(f'## {OPTIONS_START}')]
-footer = readme[readme.index(f'# {OPTIONS_END}'):]
-
 with open(README_FILE, 'w', encoding='utf-8') as f:
-    for part in (header, options, footer):
-        f.write(part)
+    f.write(''.join((
+        take_section(readme, end=f'## {OPTIONS_START}'),
+        functools.reduce(apply_patch, PATCHES, options),
+        take_section(readme, f'# {OPTIONS_END}'),
+    )))

devscripts/make_supportedsites.py

@@ -1,10 +1,14 @@
 #!/usr/bin/env python3
-import optparse
+
+# Allow direct execution
 import os
 import sys
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import optparse
+
 from yt_dlp.extractor import list_extractor_classes
 
 

devscripts/prepare_manpage.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 import optparse
 import os.path
 import re
@@ -23,7 +24,7 @@ yt\-dlp \- A youtube-dl fork with additional features and patches
 
 def main():
     parser = optparse.OptionParser(usage='%prog OUTFILE.md')
-    options, args = parser.parse_args()
+    _, args = parser.parse_args()
     if len(args) != 1:
         parser.error('Expected an output filename')
 

devscripts/run_tests.sh

@@ -1,4 +1,4 @@
-#!/bin/sh
+#!/usr/bin/env sh
 
 if [ -z $1 ]; then
     test_set='test'

devscripts/update-formulae.py

@@ -1,12 +1,15 @@
 #!/usr/bin/env python3
-import json
+
+# Allow direct execution
 import os
-import re
 import sys
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from yt_dlp.compat import compat_urllib_request
+
+import json
+import re
+import urllib.request
 
 # usage: python3 ./devscripts/update-formulae.py <path-to-formulae-rb> <version>
 # version can be either 0-aligned (yt-dlp version) or normalized (PyPl version)
@@ -15,7 +18,7 @@ filename, version = sys.argv[1:]
 
 normalized_version = '.'.join(str(int(x)) for x in version.split('.'))
 
-pypi_release = json.loads(compat_urllib_request.urlopen(
+pypi_release = json.loads(urllib.request.urlopen(
     'https://pypi.org/pypi/yt-dlp/%s/json' % normalized_version
 ).read().decode())
 

devscripts/update-version.py

@@ -1,4 +1,12 @@
 #!/usr/bin/env python3
+
+# Allow direct execution
+import os
+import sys
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
 import subprocess
 import sys
 from datetime import datetime

devscripts/zsh-completion.py

@@ -1,9 +1,12 @@
 #!/usr/bin/env python3
+
+# Allow direct execution
 import os
 import sys
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
 import yt_dlp
 
 ZSH_COMPLETION_FILE = "completions/zsh/_yt-dlp"

pyinst.py

@@ -1,28 +1,12 @@
 #!/usr/bin/env python3
+
 import os
 import platform
 import sys
 
 from PyInstaller.__main__ import run as run_pyinstaller
 
-OS_NAME = platform.system()
-if OS_NAME == 'Windows':
-    from PyInstaller.utils.win32.versioninfo import (
-        FixedFileInfo,
-        SetVersion,
-        StringFileInfo,
-        StringStruct,
-        StringTable,
-        VarFileInfo,
-        VarStruct,
-        VSVersionInfo,
-    )
-elif OS_NAME == 'Darwin':
-    pass
-else:
-    raise Exception(f'{OS_NAME} is not supported')
-
-ARCH = platform.architecture()[0][:2]
+OS_NAME, ARCH = sys.platform, platform.architecture()[0][:2]
 
 
 def main():
@@ -33,10 +17,7 @@ def main():
     if not onedir and '-F' not in opts and '--onefile' not in opts:
         opts.append('--onefile')
 
-    name = 'yt-dlp%s' % ('_macos' if OS_NAME == 'Darwin' else '_x86' if ARCH == '32' else '')
-    final_file = ''.join((
-        'dist/', f'{name}/' if onedir else '', name, '.exe' if OS_NAME == 'Windows' else ''))
-
+    name, final_file = exe(onedir)
     print(f'Building yt-dlp v{version} {ARCH}bit for {OS_NAME} with options {opts}')
     print('Remember to update the version using  "devscripts/update-version.py"')
     if not os.path.isfile('yt_dlp/extractor/lazy_extractors.py'):
@@ -63,7 +44,7 @@ def main():
 
 
 def parse_options():
-    # Compatability with older arguments
+    # Compatibility with older arguments
     opts = sys.argv[1:]
     if opts[0:1] in (['32'], ['64']):
         if ARCH != opts[0]:
@@ -79,6 +60,21 @@ def read_version(fname):
         return locals()['__version__']
 
 
+def exe(onedir):
+    """@returns (name, path)"""
+    name = '_'.join(filter(None, (
+        'yt-dlp',
+        {'win32': '', 'darwin': 'macos'}.get(OS_NAME, OS_NAME),
+        ARCH == '32' and 'x86'
+    )))
+    return name, ''.join(filter(None, (
+        'dist/',
+        onedir and f'{name}/',
+        name,
+        OS_NAME == 'win32' and '.exe'
+    )))
+
+
 def version_to_list(version):
     version_list = version.split('.')
     return list(map(int, version_list)) + [0] * (4 - len(version_list))
@@ -109,11 +105,22 @@ def pycryptodome_module():
 
 
 def set_version_info(exe, version):
-    if OS_NAME == 'Windows':
+    if OS_NAME == 'win32':
         windows_set_version(exe, version)
 
 
 def windows_set_version(exe, version):
+    from PyInstaller.utils.win32.versioninfo import (
+        FixedFileInfo,
+        SetVersion,
+        StringFileInfo,
+        StringStruct,
+        StringTable,
+        VarFileInfo,
+        VarStruct,
+        VSVersionInfo,
+    )
+
     version_list = version_to_list(version)
     suffix = '_x86' if ARCH == '32' else ''
     SetVersion(exe, VSVersionInfo(

pytest.ini

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

requirements.txt

@@ -3,4 +3,4 @@ pycryptodomex
 websockets
 brotli; platform_python_implementation=='CPython'
 brotlicffi; platform_python_implementation!='CPython'
-certifi
+certifi

setup.cfg

@@ -1,6 +1,41 @@
 [wheel]
-universal = True
+universal = true
+
 
 [flake8]
-exclude = devscripts/lazy_load_template.py,devscripts/make_issue_template.py,setup.py,build,.git,venv
+exclude = build,venv,.tox,.git,.pytest_cache
 ignore = E402,E501,E731,E741,W503
+max_line_length = 120
+per_file_ignores =
+    devscripts/lazy_load_template.py: F401
+
+
+[tool:pytest]
+addopts = -ra -v --strict-markers
+markers =
+    download
+
+
+[tox:tox]
+skipsdist = true
+envlist = py{36,37,38,39,310},pypy{36,37,38,39}
+skip_missing_interpreters = true
+
+[testenv]  # tox
+deps =
+   pytest
+commands = pytest {posargs:"-m not download"}
+passenv = HOME  # For test_compat_expanduser
+setenv =
+    # PYTHONWARNINGS = error  # Catches PIP's warnings too
+
+
+[isort]
+py_version = 36
+multi_line_output = VERTICAL_HANGING_INDENT
+line_length = 80
+reverse_relative = true
+ensure_newline_before_comments = true
+include_trailing_comma = true
+known_first_party =
+    test

setup.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 import os.path
 import sys
 import warnings
@@ -36,7 +37,7 @@ REQUIREMENTS = read('requirements.txt').splitlines()
 
 
 if sys.argv[1:2] == ['py2exe']:
-    import py2exe
+    import py2exe  # noqa: F401
     warnings.warn(
         'py2exe builds do not support pycryptodomex and needs VC++14 to run. '
         'The recommended way is to use "pyinst.py" to build using pyinstaller')
@@ -140,6 +141,9 @@ setup(
         'Programming Language :: Python :: 3.6',
         'Programming Language :: Python :: 3.7',
         'Programming Language :: Python :: 3.8',
+        'Programming Language :: Python :: 3.9',
+        'Programming Language :: Python :: 3.10',
+        'Programming Language :: Python :: 3.11',
         'Programming Language :: Python :: Implementation',
         'Programming Language :: Python :: Implementation :: CPython',
         'Programming Language :: Python :: Implementation :: PyPy',

supportedsites.md

@@ -1,4 +1,6 @@
 # Supported sites
+ - **0000studio:archive**
+ - **0000studio:clip**
  - **17live**
  - **17live:clip**
  - **1tv**: Первый канал
@@ -60,8 +62,6 @@
  - **AmHistoryChannel**
  - **anderetijden**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **AnimalPlanet**
- - **AnimeLab**: [<abbr title="netrc machine"><em>animelab</em></abbr>]
- - **AnimeLabShows**: [<abbr title="netrc machine"><em>animelab</em></abbr>]
  - **AnimeOnDemand**: [<abbr title="netrc machine"><em>animeondemand</em></abbr>]
  - **ant1newsgr:article**: ant1news.gr articles
  - **ant1newsgr:embed**: ant1news.gr embedded videos
@@ -89,6 +89,7 @@
  - **AsianCrush**
  - **AsianCrushPlaylist**
  - **AtresPlayer**: [<abbr title="netrc machine"><em>atresplayer</em></abbr>]
+ - **AtScaleConfEvent**
  - **ATTTechChannel**
  - **ATVAt**
  - **AudiMedia**
@@ -276,6 +277,8 @@
  - **dailymotion**: [<abbr title="netrc machine"><em>dailymotion</em></abbr>]
  - **dailymotion:playlist**: [<abbr title="netrc machine"><em>dailymotion</em></abbr>]
  - **dailymotion:user**: [<abbr title="netrc machine"><em>dailymotion</em></abbr>]
+ - **DailyWire**
+ - **DailyWirePodcast**
  - **damtomo:record**
  - **damtomo:video**
  - **daum.net**
@@ -322,8 +325,8 @@
  - **drtv**
  - **drtv:live**
  - **DTube**
- - **duboku**: www.duboku.co
- - **duboku:list**: www.duboku.co entire series
+ - **duboku**: www.duboku.io
+ - **duboku:list**: www.duboku.io entire series
  - **Dumpert**
  - **dvtv**: http://video.aktualne.cz/
  - **dw**
@@ -403,6 +406,8 @@
  - **FranceTVSite**
  - **Freesound**
  - **freespeech.org**
+ - **freetv:series**
+ - **FreeTvMovies**
  - **FrontendMasters**: [<abbr title="netrc machine"><em>frontendmasters</em></abbr>]
  - **FrontendMastersCourse**: [<abbr title="netrc machine"><em>frontendmasters</em></abbr>]
  - **FrontendMastersLesson**: [<abbr title="netrc machine"><em>frontendmasters</em></abbr>]
@@ -413,6 +418,7 @@
  - **Funk**
  - **Fusion**
  - **Fux**
+ - **FuyinTV**
  - **Gab**
  - **GabTV**
  - **Gaia**: [<abbr title="netrc machine"><em>gaia</em></abbr>]
@@ -452,6 +458,7 @@
  - **google:podcasts**
  - **google:podcasts:feed**
  - **GoogleDrive**
+ - **GoogleDrive:Folder**
  - **GoPro**
  - **Goshgay**
  - **GoToStage**
@@ -535,6 +542,7 @@
  - **Iwara**
  - **iwara:playlist**
  - **iwara:user**
+ - **Ixigua**
  - **Izlesene**
  - **Jable**
  - **JablePlaylist**
@@ -554,12 +562,14 @@
  - **Ketnet**
  - **khanacademy**
  - **khanacademy:unit**
+ - **Kicker**
  - **KickStarter**
  - **KinjaEmbed**
  - **KinoPoisk**
  - **KonserthusetPlay**
  - **Koo**
  - **KrasView**: Красвью
+ - **KTH**
  - **Ku6**
  - **KUSI**
  - **kuwo:album**: 酷我音乐 - 专辑
@@ -609,6 +619,7 @@
  - **LiveJournal**
  - **livestream**
  - **livestream:original**
+ - **Livestreamfails**
  - **Lnk**
  - **LnkGo**
  - **loc**: Library of Congress
@@ -675,6 +686,7 @@
  - **miomio.tv**
  - **mirrativ**
  - **mirrativ:user**
+ - **MirrorCoUK**
  - **MiTele**: mitele.es
  - **mixch**
  - **mixch:archive**
@@ -740,6 +752,7 @@
  - **NationalGeographicTV**
  - **Naver**
  - **Naver:live**
+ - **navernow**
  - **NBA**
  - **nba:watch**
  - **nba:watch:collection**
@@ -769,6 +782,8 @@
  - **netease:singer**: 网易云音乐 - 歌手
  - **netease:song**: 网易云音乐
  - **NetPlus**: [<abbr title="netrc machine"><em>netplus</em></abbr>]
+ - **Netverse**
+ - **NetversePlaylist**
  - **Netzkino**
  - **Newgrounds**
  - **Newgrounds:playlist**
@@ -932,6 +947,7 @@
  - **PlayPlusTV**: [<abbr title="netrc machine"><em>playplustv</em></abbr>]
  - **PlayStuff**
  - **PlaysTV**
+ - **PlaySuisse**
  - **Playtvak**: Playtvak.cz, iDNES.cz and Lidovky.cz
  - **Playvid**
  - **PlayVids**
@@ -942,7 +958,6 @@
  - **Podchaser**
  - **podomatic**
  - **Pokemon**
- - **PokemonSoundLibrary**
  - **PokemonWatch**
  - **PokerGo**: [<abbr title="netrc machine"><em>pokergo</em></abbr>]
  - **PokerGoCollection**: [<abbr title="netrc machine"><em>pokergo</em></abbr>]
@@ -969,6 +984,7 @@
  - **PornoVoisines**
  - **PornoXO**
  - **PornTube**
+ - **PremiershipRugby**
  - **PressTV**
  - **ProjectVeritas**
  - **prosiebensat1**: ProSiebenSat.1 Digital
@@ -1100,6 +1116,7 @@
  - **ScreencastOMatic**
  - **ScrippsNetworks**
  - **scrippsnetworks:watch**
+ - **Scrolller**
  - **SCTE**: [<abbr title="netrc machine"><em>scte</em></abbr>]
  - **SCTECourse**: [<abbr title="netrc machine"><em>scte</em></abbr>]
  - **Seeker**
@@ -1150,6 +1167,7 @@
  - **southpark.cc.com**
  - **southpark.cc.com:español**
  - **southpark.de**
+ - **southpark.lat**
  - **southpark.nl**
  - **southparkstudios.dk**
  - **SovietsCloset**
@@ -1175,6 +1193,7 @@
  - **stanfordoc**: Stanford Open ClassRoom
  - **startv**
  - **Steam**
+ - **SteamCommunityBroadcast**
  - **Stitcher**
  - **StitcherShow**
  - **StoryFire**
@@ -1189,6 +1208,7 @@
  - **StretchInternet**
  - **Stripchat**
  - **stv:player**
+ - **Substack**
  - **SunPorno**
  - **sverigesradio:episode**
  - **sverigesradio:publication**
@@ -1412,7 +1432,8 @@
  - **vimeo:watchlater**: [<abbr title="netrc machine"><em>vimeo</em></abbr>] Vimeo watch later list, ":vimeowatchlater" keyword (requires authentication)
  - **Vimm:recording**
  - **Vimm:stream**
- - **Vimp**
+ - **ViMP**
+ - **ViMP:Playlist**
  - **Vimple**: Vimple - one-click video hosting
  - **Vine**
  - **vine:user**
@@ -1463,6 +1484,7 @@
  - **washingtonpost:article**
  - **wat.tv**
  - **WatchBox**
+ - **WatchESPN**
  - **WatchIndianPorn**: Watch Indian Porn
  - **WDR**
  - **wdr:mobile**: (**Currently broken**)
@@ -1535,6 +1557,7 @@
  - **YourPorn**
  - **YourUpload**
  - **youtube**: YouTube
+ - **youtube:clip**
  - **youtube:favorites**: YouTube liked videos; ":ytfav" keyword (requires cookies)
  - **youtube:history**: Youtube watch history; ":ythis" keyword (requires cookies)
  - **youtube:music:search_url**: YouTube music search URLs with selectable sections (Eg: #songs)

test/helper.py

@@ -9,7 +9,7 @@ import types
 
 import yt_dlp.extractor
 from yt_dlp import YoutubeDL
-from yt_dlp.compat import compat_os_name, compat_str
+from yt_dlp.compat import compat_os_name
 from yt_dlp.utils import preferredencoding, write_string
 
 if 'pytest' in sys.modules:
@@ -44,7 +44,7 @@ def try_rm(filename):
             raise
 
 
-def report_warning(message):
+def report_warning(message, *args, **kwargs):
     '''
     Print the message to stderr, it will be prefixed with 'WARNING:'
     If stderr is a tty file the 'WARNING:' will be colored
@@ -67,10 +67,10 @@ class FakeYDL(YoutubeDL):
         super().__init__(params, auto_init=False)
         self.result = []
 
-    def to_screen(self, s, skip_eol=None):
+    def to_screen(self, s, *args, **kwargs):
         print(s)
 
-    def trouble(self, s, tb=None):
+    def trouble(self, s, *args, **kwargs):
         raise Exception(s)
 
     def download(self, x):
@@ -80,10 +80,10 @@ class FakeYDL(YoutubeDL):
         # Silence an expected warning matching a regex
         old_report_warning = self.report_warning
 
-        def report_warning(self, message):
+        def report_warning(self, message, *args, **kwargs):
             if re.match(regex, message):
                 return
-            old_report_warning(message)
+            old_report_warning(message, *args, **kwargs)
         self.report_warning = types.MethodType(report_warning, self)
 
 
@@ -96,29 +96,29 @@ md5 = lambda s: hashlib.md5(s.encode()).hexdigest()
 
 
 def expect_value(self, got, expected, field):
-    if isinstance(expected, compat_str) and expected.startswith('re:'):
+    if isinstance(expected, str) and expected.startswith('re:'):
         match_str = expected[len('re:'):]
         match_rex = re.compile(match_str)
 
         self.assertTrue(
-            isinstance(got, compat_str),
-            f'Expected a {compat_str.__name__} object, but got {type(got).__name__} for field {field}')
+            isinstance(got, str),
+            f'Expected a {str.__name__} object, but got {type(got).__name__} for field {field}')
         self.assertTrue(
             match_rex.match(got),
             f'field {field} (value: {got!r}) should match {match_str!r}')
-    elif isinstance(expected, compat_str) and expected.startswith('startswith:'):
+    elif isinstance(expected, str) and expected.startswith('startswith:'):
         start_str = expected[len('startswith:'):]
         self.assertTrue(
-            isinstance(got, compat_str),
-            f'Expected a {compat_str.__name__} object, but got {type(got).__name__} for field {field}')
+            isinstance(got, str),
+            f'Expected a {str.__name__} object, but got {type(got).__name__} for field {field}')
         self.assertTrue(
             got.startswith(start_str),
             f'field {field} (value: {got!r}) should start with {start_str!r}')
-    elif isinstance(expected, compat_str) and expected.startswith('contains:'):
+    elif isinstance(expected, str) and expected.startswith('contains:'):
         contains_str = expected[len('contains:'):]
         self.assertTrue(
-            isinstance(got, compat_str),
-            f'Expected a {compat_str.__name__} object, but got {type(got).__name__} for field {field}')
+            isinstance(got, str),
+            f'Expected a {str.__name__} object, but got {type(got).__name__} for field {field}')
         self.assertTrue(
             contains_str in got,
             f'field {field} (value: {got!r}) should contain {contains_str!r}')
@@ -142,12 +142,12 @@ def expect_value(self, got, expected, field):
                     index, field, type_expected, type_got))
             expect_value(self, item_got, item_expected, field)
     else:
-        if isinstance(expected, compat_str) and expected.startswith('md5:'):
+        if isinstance(expected, str) and expected.startswith('md5:'):
             self.assertTrue(
-                isinstance(got, compat_str),
+                isinstance(got, str),
                 f'Expected field {field} to be a unicode object, but got value {got!r} of type {type(got)!r}')
             got = 'md5:' + md5(got)
-        elif isinstance(expected, compat_str) and re.match(r'^(?:min|max)?count:\d+', expected):
+        elif isinstance(expected, str) and re.match(r'^(?:min|max)?count:\d+', expected):
             self.assertTrue(
                 isinstance(got, (list, dict)),
                 f'Expected field {field} to be a list or a dict, but it is of type {type(got).__name__}')
@@ -236,7 +236,7 @@ def expect_info_dict(self, got_dict, expected_dict):
     missing_keys = set(test_info_dict.keys()) - set(expected_dict.keys())
     if missing_keys:
         def _repr(v):
-            if isinstance(v, compat_str):
+            if isinstance(v, str):
                 return "'%s'" % v.replace('\\', '\\\\').replace("'", "\\'").replace('\n', '\\n')
             elif isinstance(v, type):
                 return v.__name__
@@ -301,9 +301,9 @@ def assertEqual(self, got, expected, msg=None):
 def expect_warnings(ydl, warnings_re):
     real_warning = ydl.report_warning
 
-    def _report_warning(w):
+    def _report_warning(w, *args, **kwargs):
         if not any(re.search(w_re, w) for w_re in warnings_re):
-            real_warning(w)
+            real_warning(w, *args, **kwargs)
 
     ydl.report_warning = _report_warning
 

test/test_InfoExtractor.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,10 +7,12 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-import threading
-from test.helper import FakeYDL, expect_dict, expect_value, http_server_port
 
-from yt_dlp.compat import compat_etree_fromstring, compat_http_server
+import http.server
+import threading
+
+from test.helper import FakeYDL, expect_dict, expect_value, http_server_port
+from yt_dlp.compat import compat_etree_fromstring
 from yt_dlp.extractor import YoutubeIE, get_info_extractor
 from yt_dlp.extractor.common import InfoExtractor
 from yt_dlp.utils import (
@@ -23,7 +26,7 @@ TEAPOT_RESPONSE_STATUS = 418
 TEAPOT_RESPONSE_BODY = "<h1>418 I'm a teapot</h1>"
 
 
-class InfoExtractorTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
+class InfoExtractorTestRequestHandler(http.server.BaseHTTPRequestHandler):
     def log_message(self, format, *args):
         pass
 
@@ -502,6 +505,24 @@ class TestInfoExtractor(unittest.TestCase):
                 }],
             })
 
+        # from https://0000.studio/
+        # with type attribute but without extension in URL
+        expect_dict(
+            self,
+            self.ie._parse_html5_media_entries(
+                'https://0000.studio',
+                r'''
+                <video src="https://d1ggyt9m8pwf3g.cloudfront.net/protected/ap-northeast-1:1864af40-28d5-492b-b739-b32314b1a527/archive/clip/838db6a7-8973-4cd6-840d-8517e4093c92"
+                    controls="controls" type="video/mp4" preload="metadata" autoplay="autoplay" playsinline class="object-contain">
+                </video>
+                ''', None)[0],
+            {
+                'formats': [{
+                    'url': 'https://d1ggyt9m8pwf3g.cloudfront.net/protected/ap-northeast-1:1864af40-28d5-492b-b739-b32314b1a527/archive/clip/838db6a7-8973-4cd6-840d-8517e4093c92',
+                    'ext': 'mp4',
+                }],
+            })
+
     def test_extract_jwplayer_data_realworld(self):
         # from http://www.suffolk.edu/sjc/
         expect_dict(
@@ -1637,7 +1658,7 @@ jwplayer("mediaplayer").setup({"abouttext":"Visit Indie DB","aboutlink":"http:\/
         # or the underlying `_download_webpage_handle` returning no content
         # when a response matches `expected_status`.
 
-        httpd = compat_http_server.HTTPServer(
+        httpd = http.server.HTTPServer(
             ('127.0.0.1', 0), InfoExtractorTestRequestHandler)
         port = http_server_port(httpd)
         server_thread = threading.Thread(target=httpd.serve_forever)

test/test_YoutubeDL.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,23 +7,21 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
 import copy
 import json
-from test.helper import FakeYDL, assertRegexpMatches
+import urllib.error
 
+from test.helper import FakeYDL, assertRegexpMatches
 from yt_dlp import YoutubeDL
-from yt_dlp.compat import (
-    compat_os_name,
-    compat_setenv,
-    compat_str,
-    compat_urllib_error,
-)
+from yt_dlp.compat import compat_os_name
 from yt_dlp.extractor import YoutubeIE
 from yt_dlp.extractor.common import InfoExtractor
 from yt_dlp.postprocessor.common import PostProcessor
 from yt_dlp.utils import (
     ExtractorError,
     LazyList,
+    OnDemandPagedList,
     int_or_none,
     match_filter_func,
 )
@@ -39,7 +38,7 @@ class YDL(FakeYDL):
     def process_info(self, info_dict):
         self.downloaded_info_dicts.append(info_dict.copy())
 
-    def to_screen(self, msg):
+    def to_screen(self, msg, *args, **kwargs):
         self.msgs.append(msg)
 
     def dl(self, *args, **kwargs):
@@ -840,14 +839,14 @@ class TestYoutubeDL(unittest.TestCase):
         # test('%(foo|)s', ('', '_'))  # fixme
 
         # Environment variable expansion for prepare_filename
-        compat_setenv('__yt_dlp_var', 'expanded')
+        os.environ['__yt_dlp_var'] = 'expanded'
         envvar = '%__yt_dlp_var%' if compat_os_name == 'nt' else '$__yt_dlp_var'
         test(envvar, (envvar, 'expanded'))
         if compat_os_name == 'nt':
             test('%s%', ('%s%', '%s%'))
-            compat_setenv('s', 'expanded')
+            os.environ['s'] = 'expanded'
             test('%s%', ('%s%', 'expanded'))  # %s% should be expanded before escaping %s
-            compat_setenv('(test)s', 'expanded')
+            os.environ['(test)s'] = 'expanded'
             test('%(test)s%', ('NA%', 'expanded'))  # Environment should take priority over template
 
         # Path expansion and escaping
@@ -989,41 +988,79 @@ class TestYoutubeDL(unittest.TestCase):
         self.assertEqual(res, [])
 
     def test_playlist_items_selection(self):
-        entries = [{
-            'id': compat_str(i),
-            'title': compat_str(i),
-            'url': TEST_URL,
-        } for i in range(1, 5)]
-        playlist = {
-            '_type': 'playlist',
-            'id': 'test',
-            'entries': entries,
-            'extractor': 'test:playlist',
-            'extractor_key': 'test:playlist',
-            'webpage_url': 'http://example.com',
-        }
+        INDICES, PAGE_SIZE = list(range(1, 11)), 3
 
-        def get_downloaded_info_dicts(params):
+        def entry(i, evaluated):
+            evaluated.append(i)
+            return {
+                'id': str(i),
+                'title': str(i),
+                'url': TEST_URL,
+            }
+
+        def pagedlist_entries(evaluated):
+            def page_func(n):
+                start = PAGE_SIZE * n
+                for i in INDICES[start: start + PAGE_SIZE]:
+                    yield entry(i, evaluated)
+            return OnDemandPagedList(page_func, PAGE_SIZE)
+
+        def page_num(i):
+            return (i + PAGE_SIZE - 1) // PAGE_SIZE
+
+        def generator_entries(evaluated):
+            for i in INDICES:
+                yield entry(i, evaluated)
+
+        def list_entries(evaluated):
+            return list(generator_entries(evaluated))
+
+        def lazylist_entries(evaluated):
+            return LazyList(generator_entries(evaluated))
+
+        def get_downloaded_info_dicts(params, entries):
             ydl = YDL(params)
-            # make a deep copy because the dictionary and nested entries
-            # can be modified
-            ydl.process_ie_result(copy.deepcopy(playlist))
+            ydl.process_ie_result({
+                '_type': 'playlist',
+                'id': 'test',
+                'extractor': 'test:playlist',
+                'extractor_key': 'test:playlist',
+                'webpage_url': 'http://example.com',
+                'entries': entries,
+            })
             return ydl.downloaded_info_dicts
 
-        def test_selection(params, expected_ids):
-            results = [
-                (v['playlist_autonumber'] - 1, (int(v['id']), v['playlist_index']))
-                for v in get_downloaded_info_dicts(params)]
-            self.assertEqual(results, list(enumerate(zip(expected_ids, expected_ids))))
+        def test_selection(params, expected_ids, evaluate_all=False):
+            expected_ids = list(expected_ids)
+            if evaluate_all:
+                generator_eval = pagedlist_eval = INDICES
+            elif not expected_ids:
+                generator_eval = pagedlist_eval = []
+            else:
+                generator_eval = INDICES[0: max(expected_ids)]
+                pagedlist_eval = INDICES[PAGE_SIZE * page_num(min(expected_ids)) - PAGE_SIZE:
+                                         PAGE_SIZE * page_num(max(expected_ids))]
 
-        test_selection({}, [1, 2, 3, 4])
-        test_selection({'playlistend': 10}, [1, 2, 3, 4])
-        test_selection({'playlistend': 2}, [1, 2])
-        test_selection({'playliststart': 10}, [])
-        test_selection({'playliststart': 2}, [2, 3, 4])
-        test_selection({'playlist_items': '2-4'}, [2, 3, 4])
+            for name, func, expected_eval in (
+                ('list', list_entries, INDICES),
+                ('Generator', generator_entries, generator_eval),
+                # ('LazyList', lazylist_entries, generator_eval),  # Generator and LazyList follow the exact same code path
+                ('PagedList', pagedlist_entries, pagedlist_eval),
+            ):
+                evaluated = []
+                entries = func(evaluated)
+                results = [(v['playlist_autonumber'] - 1, (int(v['id']), v['playlist_index']))
+                           for v in get_downloaded_info_dicts(params, entries)]
+                self.assertEqual(results, list(enumerate(zip(expected_ids, expected_ids))), f'Entries of {name} for {params}')
+                self.assertEqual(sorted(evaluated), expected_eval, f'Evaluation of {name} for {params}')
+        test_selection({}, INDICES)
+        test_selection({'playlistend': 20}, INDICES, True)
+        test_selection({'playlistend': 2}, INDICES[:2])
+        test_selection({'playliststart': 11}, [], True)
+        test_selection({'playliststart': 2}, INDICES[1:])
+        test_selection({'playlist_items': '2-4'}, INDICES[1:4])
         test_selection({'playlist_items': '2,4'}, [2, 4])
-        test_selection({'playlist_items': '10'}, [])
+        test_selection({'playlist_items': '20'}, [], True)
         test_selection({'playlist_items': '0'}, [])
 
         # Tests for https://github.com/ytdl-org/youtube-dl/issues/10591
@@ -1032,15 +1069,37 @@ class TestYoutubeDL(unittest.TestCase):
 
         # Tests for https://github.com/yt-dlp/yt-dlp/issues/720
         # https://github.com/yt-dlp/yt-dlp/issues/302
-        test_selection({'playlistreverse': True}, [4, 3, 2, 1])
-        test_selection({'playliststart': 2, 'playlistreverse': True}, [4, 3, 2])
+        test_selection({'playlistreverse': True}, INDICES[::-1])
+        test_selection({'playliststart': 2, 'playlistreverse': True}, INDICES[:0:-1])
         test_selection({'playlist_items': '2,4', 'playlistreverse': True}, [4, 2])
         test_selection({'playlist_items': '4,2'}, [4, 2])
 
+        # Tests for --playlist-items start:end:step
+        test_selection({'playlist_items': ':'}, INDICES, True)
+        test_selection({'playlist_items': '::1'}, INDICES, True)
+        test_selection({'playlist_items': '::-1'}, INDICES[::-1], True)
+        test_selection({'playlist_items': ':6'}, INDICES[:6])
+        test_selection({'playlist_items': ':-6'}, INDICES[:-5], True)
+        test_selection({'playlist_items': '-1:6:-2'}, INDICES[:4:-2], True)
+        test_selection({'playlist_items': '9:-6:-2'}, INDICES[8:3:-2], True)
+
+        test_selection({'playlist_items': '1:inf:2'}, INDICES[::2], True)
+        test_selection({'playlist_items': '-2:inf'}, INDICES[-2:], True)
+        test_selection({'playlist_items': ':inf:-1'}, [], True)
+        test_selection({'playlist_items': '0-2:2'}, [2])
+        test_selection({'playlist_items': '1-:2'}, INDICES[::2], True)
+        test_selection({'playlist_items': '0--2:2'}, INDICES[1:-1:2], True)
+
+        test_selection({'playlist_items': '10::3'}, [10], True)
+        test_selection({'playlist_items': '-1::3'}, [10], True)
+        test_selection({'playlist_items': '11::3'}, [], True)
+        test_selection({'playlist_items': '-15::2'}, INDICES[1::2], True)
+        test_selection({'playlist_items': '-15::15'}, [], True)
+
     def test_urlopen_no_file_protocol(self):
         # see https://github.com/ytdl-org/youtube-dl/issues/8227
         ydl = YDL()
-        self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd')
+        self.assertRaises(urllib.error.URLError, ydl.urlopen, 'file:///etc/passwd')
 
     def test_do_not_override_ie_key_in_url_transparent(self):
         ydl = YDL()
@@ -1126,7 +1185,7 @@ class TestYoutubeDL(unittest.TestCase):
 
             def _entries(self):
                 for n in range(3):
-                    video_id = compat_str(n)
+                    video_id = str(n)
                     yield {
                         '_type': 'url_transparent',
                         'ie_key': VideoIE.ie_key(),

test/test_YoutubeDLCookieJar.py

@@ -1,12 +1,16 @@
 #!/usr/bin/env python3
+
+# Allow direct execution
 import os
-import re
 import sys
-import tempfile
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import re
+import tempfile
+
 from yt_dlp.utils import YoutubeDLCookieJar
 
 

test/test_aes.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,6 +7,7 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
 import base64
 
 from yt_dlp.aes import (

test/test_age_restriction.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,8 +7,8 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import is_download_test, try_rm
 
+from test.helper import is_download_test, try_rm
 from yt_dlp import YoutubeDL
 
 

test/test_all_urls.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
-import collections
 import os
 import sys
 import unittest
@@ -8,8 +8,9 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
-from test.helper import gettestcases
+import collections
 
+from test.helper import gettestcases
 from yt_dlp.extractor import FacebookIE, YoutubeIE, gen_extractors
 
 

test/test_cache.py

@@ -1,15 +1,16 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
-import shutil
 import sys
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
-from test.helper import FakeYDL
+import shutil
 
+from test.helper import FakeYDL
 from yt_dlp.cache import Cache
 
 

test/test_compat.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -7,16 +8,14 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
+import struct
+import urllib.parse
+
 from yt_dlp import compat
 from yt_dlp.compat import (
     compat_etree_fromstring,
     compat_expanduser,
-    compat_getenv,
-    compat_setenv,
-    compat_str,
-    compat_struct_unpack,
     compat_urllib_parse_unquote,
-    compat_urllib_parse_unquote_plus,
     compat_urllib_parse_urlencode,
 )
 
@@ -26,28 +25,19 @@ class TestCompat(unittest.TestCase):
         with self.assertWarns(DeprecationWarning):
             compat.compat_basestring
 
+        with self.assertWarns(DeprecationWarning):
+            compat.WINDOWS_VT_MODE
+
         compat.asyncio.events  # Must not raise error
 
-    def test_compat_getenv(self):
-        test_str = 'тест'
-        compat_setenv('yt_dlp_COMPAT_GETENV', test_str)
-        self.assertEqual(compat_getenv('yt_dlp_COMPAT_GETENV'), test_str)
-
-    def test_compat_setenv(self):
-        test_var = 'yt_dlp_COMPAT_SETENV'
-        test_str = 'тест'
-        compat_setenv(test_var, test_str)
-        compat_getenv(test_var)
-        self.assertEqual(compat_getenv(test_var), test_str)
-
     def test_compat_expanduser(self):
         old_home = os.environ.get('HOME')
         test_str = R'C:\Documents and Settings\тест\Application Data'
         try:
-            compat_setenv('HOME', test_str)
+            os.environ['HOME'] = test_str
             self.assertEqual(compat_expanduser('~'), test_str)
         finally:
-            compat_setenv('HOME', old_home or '')
+            os.environ['HOME'] = old_home or ''
 
     def test_compat_urllib_parse_unquote(self):
         self.assertEqual(compat_urllib_parse_unquote('abc%20def'), 'abc def')
@@ -69,8 +59,8 @@ class TestCompat(unittest.TestCase):
             '''(^◣_◢^)っ︻デ═一    ⇀    ⇀    ⇀    ⇀    ⇀    ↶%I%Break%Things%''')
 
     def test_compat_urllib_parse_unquote_plus(self):
-        self.assertEqual(compat_urllib_parse_unquote_plus('abc%20def'), 'abc def')
-        self.assertEqual(compat_urllib_parse_unquote_plus('%7e/abc+def'), '~/abc def')
+        self.assertEqual(urllib.parse.unquote_plus('abc%20def'), 'abc def')
+        self.assertEqual(urllib.parse.unquote_plus('%7e/abc+def'), '~/abc def')
 
     def test_compat_urllib_parse_urlencode(self):
         self.assertEqual(compat_urllib_parse_urlencode({'abc': 'def'}), 'abc=def')
@@ -91,11 +81,11 @@ class TestCompat(unittest.TestCase):
             </root>
         '''
         doc = compat_etree_fromstring(xml.encode())
-        self.assertTrue(isinstance(doc.attrib['foo'], compat_str))
-        self.assertTrue(isinstance(doc.attrib['spam'], compat_str))
-        self.assertTrue(isinstance(doc.find('normal').text, compat_str))
-        self.assertTrue(isinstance(doc.find('chinese').text, compat_str))
-        self.assertTrue(isinstance(doc.find('foo/bar').text, compat_str))
+        self.assertTrue(isinstance(doc.attrib['foo'], str))
+        self.assertTrue(isinstance(doc.attrib['spam'], str))
+        self.assertTrue(isinstance(doc.find('normal').text, str))
+        self.assertTrue(isinstance(doc.find('chinese').text, str))
+        self.assertTrue(isinstance(doc.find('foo/bar').text, str))
 
     def test_compat_etree_fromstring_doctype(self):
         xml = '''<?xml version="1.0"?>
@@ -104,7 +94,7 @@ class TestCompat(unittest.TestCase):
         compat_etree_fromstring(xml)
 
     def test_struct_unpack(self):
-        self.assertEqual(compat_struct_unpack('!B', b'\x00'), (0,))
+        self.assertEqual(struct.unpack('!B', b'\x00'), (0,))
 
 
 if __name__ == '__main__':

test/test_cookies.py

@@ -14,16 +14,16 @@ from yt_dlp.cookies import (
 
 
 class Logger:
-    def debug(self, message):
+    def debug(self, message, *args, **kwargs):
         print(f'[verbose] {message}')
 
-    def info(self, message):
+    def info(self, message, *args, **kwargs):
         print(message)
 
-    def warning(self, message, only_once=False):
+    def warning(self, message, *args, **kwargs):
         self.error(message)
 
-    def error(self, message):
+    def error(self, message, *args, **kwargs):
         raise Exception(message)
 
 

test/test_download.py

@@ -1,14 +1,19 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
-import hashlib
-import json
 import os
-import socket
 import sys
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import hashlib
+import http.client
+import json
+import socket
+import urllib.error
+
 from test.helper import (
     assertGreaterEqual,
     expect_info_dict,
@@ -20,12 +25,7 @@ from test.helper import (
     try_rm,
 )
 
-import yt_dlp.YoutubeDL
-from yt_dlp.compat import (
-    compat_http_client,
-    compat_HTTPError,
-    compat_urllib_error,
-)
+import yt_dlp.YoutubeDL  # isort: split
 from yt_dlp.extractor import get_info_extractor
 from yt_dlp.utils import (
     DownloadError,
@@ -43,7 +43,7 @@ class YoutubeDL(yt_dlp.YoutubeDL):
         self.processed_info_dicts = []
         super().__init__(*args, **kwargs)
 
-    def report_warning(self, message):
+    def report_warning(self, message, *args, **kwargs):
         # Don't accept warnings during tests
         raise ExtractorError(message)
 
@@ -102,9 +102,10 @@ def generator(test_case, tname):
 
         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))
+            self.skipTest(reason)
+
         if not ie.working():
             print_skipping('IE marked as not _WORKING')
-            return
 
         for tc in test_cases:
             info_dict = tc.get('info_dict', {})
@@ -118,11 +119,10 @@ def generator(test_case, tname):
 
         if 'skip' in test_case:
             print_skipping(test_case['skip'])
-            return
+
         for other_ie in other_ies:
             if not other_ie.working():
                 print_skipping('test depends on %sIE, marked as not WORKING' % other_ie.ie_key())
-                return
 
         params = get_params(test_case.get('params', {}))
         params['outtmpl'] = tname + '_' + params['outtmpl']
@@ -167,7 +167,7 @@ def generator(test_case, tname):
                         force_generic_extractor=params.get('force_generic_extractor', False))
                 except (DownloadError, ExtractorError) as err:
                     # Check if the exception is not a network related one
-                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
+                    if not err.exc_info[0] in (urllib.error.URLError, socket.timeout, UnavailableVideoError, http.client.BadStatusLine) or (err.exc_info[0] == urllib.error.HTTPError and err.exc_info[1].code == 503):
                         raise
 
                     if try_num == RETRIES:
@@ -273,7 +273,11 @@ def batch_generator(name, num_tests):
 
     def test_template(self):
         for i in range(num_tests):
-            getattr(self, f'test_{name}_{i}' if i else f'test_{name}')()
+            test_name = f'test_{name}_{i}' if i else f'test_{name}'
+            try:
+                getattr(self, test_name)()
+            except unittest.SkipTest:
+                print(f'Skipped {test_name}')
 
     return test_template
 

test/test_downloader_http.py

@@ -1,17 +1,19 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
-import re
 import sys
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-import threading
-from test.helper import http_server_port, try_rm
 
+import http.server
+import re
+import threading
+
+from test.helper import http_server_port, try_rm
 from yt_dlp import YoutubeDL
-from yt_dlp.compat import compat_http_server
 from yt_dlp.downloader.http import HttpFD
 from yt_dlp.utils import encodeFilename
 
@@ -21,7 +23,7 @@ TEST_DIR = os.path.dirname(os.path.abspath(__file__))
 TEST_SIZE = 10 * 1024
 
 
-class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
+class HTTPTestRequestHandler(http.server.BaseHTTPRequestHandler):
     def log_message(self, format, *args):
         pass
 
@@ -78,7 +80,7 @@ class FakeLogger:
 
 class TestHttpFD(unittest.TestCase):
     def setUp(self):
-        self.httpd = compat_http_server.HTTPServer(
+        self.httpd = http.server.HTTPServer(
             ('127.0.0.1', 0), HTTPTestRequestHandler)
         self.port = http_server_port(self.httpd)
         self.server_thread = threading.Thread(target=self.httpd.serve_forever)

test/test_execution.py

@@ -1,12 +1,16 @@
 #!/usr/bin/env python3
-import contextlib
+
+# Allow direct execution
 import os
-import subprocess
 import sys
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import contextlib
+import subprocess
+
 from yt_dlp.utils import encodeArgument
 
 rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))

test/test_http.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,17 +7,19 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import http.server
 import ssl
 import threading
-from test.helper import http_server_port
+import urllib.request
 
+from test.helper import http_server_port
 from yt_dlp import YoutubeDL
-from yt_dlp.compat import compat_http_server, compat_urllib_request
 
 TEST_DIR = os.path.dirname(os.path.abspath(__file__))
 
 
-class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
+class HTTPTestRequestHandler(http.server.BaseHTTPRequestHandler):
     def log_message(self, format, *args):
         pass
 
@@ -53,7 +56,7 @@ class FakeLogger:
 
 class TestHTTP(unittest.TestCase):
     def setUp(self):
-        self.httpd = compat_http_server.HTTPServer(
+        self.httpd = http.server.HTTPServer(
             ('127.0.0.1', 0), HTTPTestRequestHandler)
         self.port = http_server_port(self.httpd)
         self.server_thread = threading.Thread(target=self.httpd.serve_forever)
@@ -64,7 +67,7 @@ class TestHTTP(unittest.TestCase):
 class TestHTTPS(unittest.TestCase):
     def setUp(self):
         certfn = os.path.join(TEST_DIR, 'testcert.pem')
-        self.httpd = compat_http_server.HTTPServer(
+        self.httpd = http.server.HTTPServer(
             ('127.0.0.1', 0), HTTPTestRequestHandler)
         sslctx = ssl.SSLContext(ssl.PROTOCOL_TLS_SERVER)
         sslctx.load_cert_chain(certfn, None)
@@ -90,7 +93,7 @@ class TestClientCert(unittest.TestCase):
         certfn = os.path.join(TEST_DIR, 'testcert.pem')
         self.certdir = os.path.join(TEST_DIR, 'testdata', 'certificate')
         cacertfn = os.path.join(self.certdir, 'ca.crt')
-        self.httpd = compat_http_server.HTTPServer(('127.0.0.1', 0), HTTPTestRequestHandler)
+        self.httpd = http.server.HTTPServer(('127.0.0.1', 0), HTTPTestRequestHandler)
         sslctx = ssl.SSLContext(ssl.PROTOCOL_TLS_SERVER)
         sslctx.verify_mode = ssl.CERT_REQUIRED
         sslctx.load_verify_locations(cafile=cacertfn)
@@ -130,7 +133,7 @@ class TestClientCert(unittest.TestCase):
 
 
 def _build_proxy_handler(name):
-    class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
+    class HTTPTestRequestHandler(http.server.BaseHTTPRequestHandler):
         proxy_name = name
 
         def log_message(self, format, *args):
@@ -146,14 +149,14 @@ def _build_proxy_handler(name):
 
 class TestProxy(unittest.TestCase):
     def setUp(self):
-        self.proxy = compat_http_server.HTTPServer(
+        self.proxy = http.server.HTTPServer(
             ('127.0.0.1', 0), _build_proxy_handler('normal'))
         self.port = http_server_port(self.proxy)
         self.proxy_thread = threading.Thread(target=self.proxy.serve_forever)
         self.proxy_thread.daemon = True
         self.proxy_thread.start()
 
-        self.geo_proxy = compat_http_server.HTTPServer(
+        self.geo_proxy = http.server.HTTPServer(
             ('127.0.0.1', 0), _build_proxy_handler('geo'))
         self.geo_port = http_server_port(self.geo_proxy)
         self.geo_proxy_thread = threading.Thread(target=self.geo_proxy.serve_forever)
@@ -170,7 +173,7 @@ class TestProxy(unittest.TestCase):
         response = ydl.urlopen(url).read().decode()
         self.assertEqual(response, f'normal: {url}')
 
-        req = compat_urllib_request.Request(url)
+        req = urllib.request.Request(url)
         req.add_header('Ytdl-request-proxy', geo_proxy)
         response = ydl.urlopen(req).read().decode()
         self.assertEqual(response, f'geo: {url}')

test/test_iqiyi_sdk_interpreter.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,8 +7,8 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL, is_download_test
 
+from test.helper import FakeYDL, is_download_test
 from yt_dlp.extractor import IqiyiIE
 
 

test/test_jsinterp.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,6 +7,7 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
 from yt_dlp.jsinterp import JSInterpreter
 
 

test/test_netrc.py

@@ -1,3 +1,6 @@
+#!/usr/bin/env python3
+
+# Allow direct execution
 import os
 import sys
 import unittest

test/test_overwrites.py

@@ -1,11 +1,15 @@
 #!/usr/bin/env python3
+
+# Allow direct execution
 import os
-import subprocess
 import sys
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import subprocess
+
 from test.helper import is_download_test, try_rm
 
 root_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))

test/test_post_hooks.py

@@ -1,13 +1,15 @@
 #!/usr/bin/env python3
+
+# Allow direct execution
 import os
 import sys
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import get_params, is_download_test, try_rm
 
-import yt_dlp.YoutubeDL
+from test.helper import get_params, is_download_test, try_rm
+import yt_dlp.YoutubeDL  # isort: split
 from yt_dlp.utils import DownloadError
 
 

test/test_postprocessors.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,6 +7,7 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
 from yt_dlp import YoutubeDL
 from yt_dlp.compat import compat_shlex_quote
 from yt_dlp.postprocessor import (

test/test_socks.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,11 +7,12 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
 import random
 import subprocess
-from test.helper import FakeYDL, get_params, is_download_test
+import urllib.request
 
-from yt_dlp.compat import compat_str, compat_urllib_request
+from test.helper import FakeYDL, get_params, is_download_test
 
 
 @is_download_test
@@ -51,7 +53,7 @@ class TestMultipleSocks(unittest.TestCase):
         if params is None:
             return
         ydl = FakeYDL()
-        req = compat_urllib_request.Request('http://yt-dl.org/ip')
+        req = urllib.request.Request('http://yt-dl.org/ip')
         req.add_header('Ytdl-request-proxy', params['secondary_proxy'])
         self.assertEqual(
             ydl.urlopen(req).read().decode(),
@@ -62,7 +64,7 @@ class TestMultipleSocks(unittest.TestCase):
         if params is None:
             return
         ydl = FakeYDL()
-        req = compat_urllib_request.Request('https://yt-dl.org/ip')
+        req = urllib.request.Request('https://yt-dl.org/ip')
         req.add_header('Ytdl-request-proxy', params['secondary_proxy'])
         self.assertEqual(
             ydl.urlopen(req).read().decode(),
@@ -99,13 +101,13 @@ class TestSocks(unittest.TestCase):
         return ydl.urlopen('http://yt-dl.org/ip').read().decode()
 
     def test_socks4(self):
-        self.assertTrue(isinstance(self._get_ip('socks4'), compat_str))
+        self.assertTrue(isinstance(self._get_ip('socks4'), str))
 
     def test_socks4a(self):
-        self.assertTrue(isinstance(self._get_ip('socks4a'), compat_str))
+        self.assertTrue(isinstance(self._get_ip('socks4a'), str))
 
     def test_socks5(self):
-        self.assertTrue(isinstance(self._get_ip('socks5'), compat_str))
+        self.assertTrue(isinstance(self._get_ip('socks5'), str))
 
 
 if __name__ == '__main__':

test/test_subtitles.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,8 +7,8 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL, is_download_test, md5
 
+from test.helper import FakeYDL, is_download_test, md5
 from yt_dlp.extractor import (
     NPOIE,
     NRKTVIE,
@@ -38,6 +39,9 @@ class BaseTestSubtitles(unittest.TestCase):
         self.DL = FakeYDL()
         self.ie = self.IE()
         self.DL.add_info_extractor(self.ie)
+        if not self.IE.working():
+            print('Skipping: %s marked as not _WORKING' % self.IE.ie_key())
+            self.skipTest('IE marked as not _WORKING')
 
     def getInfoDict(self):
         info_dict = self.DL.extract_info(self.url, download=False)
@@ -57,6 +61,21 @@ class BaseTestSubtitles(unittest.TestCase):
 
 @is_download_test
 class TestYoutubeSubtitles(BaseTestSubtitles):
+    # Available subtitles for QRS8MkLhQmM:
+    # Language formats
+    # ru       vtt, ttml, srv3, srv2, srv1, json3
+    # fr       vtt, ttml, srv3, srv2, srv1, json3
+    # en       vtt, ttml, srv3, srv2, srv1, json3
+    # nl       vtt, ttml, srv3, srv2, srv1, json3
+    # de       vtt, ttml, srv3, srv2, srv1, json3
+    # ko       vtt, ttml, srv3, srv2, srv1, json3
+    # it       vtt, ttml, srv3, srv2, srv1, json3
+    # zh-Hant  vtt, ttml, srv3, srv2, srv1, json3
+    # hi       vtt, ttml, srv3, srv2, srv1, json3
+    # pt-BR    vtt, ttml, srv3, srv2, srv1, json3
+    # es-MX    vtt, ttml, srv3, srv2, srv1, json3
+    # ja       vtt, ttml, srv3, srv2, srv1, json3
+    # pl       vtt, ttml, srv3, srv2, srv1, json3
     url = 'QRS8MkLhQmM'
     IE = YoutubeIE
 
@@ -65,47 +84,60 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(len(subtitles.keys()), 13)
-        self.assertEqual(md5(subtitles['en']), '688dd1ce0981683867e7fe6fde2a224b')
-        self.assertEqual(md5(subtitles['it']), '31324d30b8430b309f7f5979a504a769')
+        self.assertEqual(md5(subtitles['en']), 'ae1bd34126571a77aabd4d276b28044d')
+        self.assertEqual(md5(subtitles['it']), '0e0b667ba68411d88fd1c5f4f4eab2f9')
         for lang in ['fr', 'de']:
             self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
-    def test_youtube_subtitles_ttml_format(self):
+    def _test_subtitles_format(self, fmt, md5_hash, lang='en'):
         self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitlesformat'] = 'ttml'
+        self.DL.params['subtitlesformat'] = fmt
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), 'c97ddf1217390906fa9fbd34901f3da2')
+        self.assertEqual(md5(subtitles[lang]), md5_hash)
+
+    def test_youtube_subtitles_ttml_format(self):
+        self._test_subtitles_format('ttml', 'c97ddf1217390906fa9fbd34901f3da2')
 
     def test_youtube_subtitles_vtt_format(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitlesformat'] = 'vtt'
+        self._test_subtitles_format('vtt', 'ae1bd34126571a77aabd4d276b28044d')
+
+    def test_youtube_subtitles_json3_format(self):
+        self._test_subtitles_format('json3', '688dd1ce0981683867e7fe6fde2a224b')
+
+    def _test_automatic_captions(self, url, lang):
+        self.url = url
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslangs'] = [lang]
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), 'ae1bd34126571a77aabd4d276b28044d')
+        self.assertTrue(subtitles[lang] is not None)
 
     def test_youtube_automatic_captions(self):
-        self.url = '8YoUxe5ncPo'
-        self.DL.params['writeautomaticsub'] = True
-        self.DL.params['subtitleslangs'] = ['it']
-        subtitles = self.getSubtitles()
-        self.assertTrue(subtitles['it'] is not None)
-
-    def test_youtube_no_automatic_captions(self):
-        self.url = 'QRS8MkLhQmM'
-        self.DL.params['writeautomaticsub'] = True
-        subtitles = self.getSubtitles()
-        self.assertTrue(not subtitles)
+        # Available automatic captions for 8YoUxe5ncPo:
+        # Language formats (all in vtt, ttml, srv3, srv2, srv1, json3)
+        # gu, zh-Hans, zh-Hant, gd, ga, gl, lb, la, lo, tt, tr,
+        # lv, lt, tk, th, tg, te, fil, haw, yi, ceb, yo, de, da,
+        # el, eo, en, eu, et, es, ru, rw, ro, bn, be, bg, uk, jv,
+        # bs, ja, or, xh, co, ca, cy, cs, ps, pt, pa, vi, pl, hy,
+        # hr, ht, hu, hmn, hi, ha, mg, uz, ml, mn, mi, mk, ur,
+        # mt, ms, mr, ug, ta, my, af, sw, is, am,
+        #                                         *it*, iw, sv, ar,
+        # su, zu, az, id, ig, nl, no, ne, ny, fr, ku, fy, fa, fi,
+        # ka, kk, sr, sq, ko, kn, km, st, sk, si, so, sn, sm, sl,
+        # ky, sd
+        # ...
+        self._test_automatic_captions('8YoUxe5ncPo', 'it')
 
+    @unittest.skip('Video unavailable')
     def test_youtube_translated_subtitles(self):
-        # This video has a subtitles track, which can be translated
-        self.url = 'i0ZabxXmH4Y'
-        self.DL.params['writeautomaticsub'] = True
-        self.DL.params['subtitleslangs'] = ['it']
-        subtitles = self.getSubtitles()
-        self.assertTrue(subtitles['it'] is not None)
+        # This video has a subtitles track, which can be translated (#4555)
+        self._test_automatic_captions('Ky9eprVWzlI', 'it')
 
     def test_youtube_nosubtitles(self):
         self.DL.expect_warning('video doesn\'t have subtitles')
-        self.url = 'n5BB19UTcdA'
+        # Available automatic captions for 8YoUxe5ncPo:
+        # ...
+        # 8YoUxe5ncPo has no subtitles
+        self.url = '8YoUxe5ncPo'
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
@@ -137,6 +169,7 @@ class TestDailymotionSubtitles(BaseTestSubtitles):
 
 
 @is_download_test
+@unittest.skip('IE broken')
 class TestTedSubtitles(BaseTestSubtitles):
     url = 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html'
     IE = TedTalkIE
@@ -162,12 +195,12 @@ class TestVimeoSubtitles(BaseTestSubtitles):
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(set(subtitles.keys()), {'de', 'en', 'es', 'fr'})
-        self.assertEqual(md5(subtitles['en']), '8062383cf4dec168fc40a088aa6d5888')
-        self.assertEqual(md5(subtitles['fr']), 'b6191146a6c5d3a452244d853fde6dc8')
+        self.assertEqual(md5(subtitles['en']), '386cbc9320b94e25cb364b97935e5dd1')
+        self.assertEqual(md5(subtitles['fr']), 'c9b69eef35bc6641c0d4da8a04f9dfac')
 
     def test_nosubtitles(self):
         self.DL.expect_warning('video doesn\'t have subtitles')
-        self.url = 'http://vimeo.com/56015672'
+        self.url = 'http://vimeo.com/68093876'
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
@@ -175,6 +208,7 @@ class TestVimeoSubtitles(BaseTestSubtitles):
 
 
 @is_download_test
+@unittest.skip('IE broken')
 class TestWallaSubtitles(BaseTestSubtitles):
     url = 'http://vod.walla.co.il/movie/2705958/the-yes-men'
     IE = WallaIE
@@ -197,6 +231,7 @@ class TestWallaSubtitles(BaseTestSubtitles):
 
 
 @is_download_test
+@unittest.skip('IE broken')
 class TestCeskaTelevizeSubtitles(BaseTestSubtitles):
     url = 'http://www.ceskatelevize.cz/ivysilani/10600540290-u6-uzasny-svet-techniky'
     IE = CeskaTelevizeIE
@@ -219,6 +254,7 @@ class TestCeskaTelevizeSubtitles(BaseTestSubtitles):
 
 
 @is_download_test
+@unittest.skip('IE broken')
 class TestLyndaSubtitles(BaseTestSubtitles):
     url = 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html'
     IE = LyndaIE
@@ -232,6 +268,7 @@ class TestLyndaSubtitles(BaseTestSubtitles):
 
 
 @is_download_test
+@unittest.skip('IE broken')
 class TestNPOSubtitles(BaseTestSubtitles):
     url = 'http://www.npo.nl/nos-journaal/28-08-2014/POW_00722860'
     IE = NPOIE
@@ -245,6 +282,7 @@ class TestNPOSubtitles(BaseTestSubtitles):
 
 
 @is_download_test
+@unittest.skip('IE broken')
 class TestMTVSubtitles(BaseTestSubtitles):
     url = 'http://www.cc.com/video-clips/p63lk0/adam-devine-s-house-party-chasing-white-swans'
     IE = ComedyCentralIE
@@ -269,8 +307,8 @@ class TestNRKSubtitles(BaseTestSubtitles):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(set(subtitles.keys()), {'no'})
-        self.assertEqual(md5(subtitles['no']), '544fa917d3197fcbee64634559221cc2')
+        self.assertEqual(set(subtitles.keys()), {'nb-ttv'})
+        self.assertEqual(md5(subtitles['nb-ttv']), '67e06ff02d0deaf975e68f6cb8f6a149')
 
 
 @is_download_test
@@ -295,6 +333,7 @@ class TestRaiPlaySubtitles(BaseTestSubtitles):
 
 
 @is_download_test
+@unittest.skip('IE broken - DRM only')
 class TestVikiSubtitles(BaseTestSubtitles):
     url = 'http://www.viki.com/videos/1060846v-punch-episode-18'
     IE = VikiIE
@@ -323,6 +362,7 @@ class TestThePlatformSubtitles(BaseTestSubtitles):
 
 
 @is_download_test
+@unittest.skip('IE broken')
 class TestThePlatformFeedSubtitles(BaseTestSubtitles):
     url = 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207'
     IE = ThePlatformFeedIE
@@ -360,7 +400,7 @@ class TestDemocracynowSubtitles(BaseTestSubtitles):
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(set(subtitles.keys()), {'en'})
-        self.assertEqual(md5(subtitles['en']), 'acaca989e24a9e45a6719c9b3d60815c')
+        self.assertEqual(md5(subtitles['en']), 'a3cc4c0b5eadd74d9974f1c1f5101045')
 
     def test_subtitles_in_page(self):
         self.url = 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree'
@@ -368,7 +408,7 @@ class TestDemocracynowSubtitles(BaseTestSubtitles):
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(set(subtitles.keys()), {'en'})
-        self.assertEqual(md5(subtitles['en']), 'acaca989e24a9e45a6719c9b3d60815c')
+        self.assertEqual(md5(subtitles['en']), 'a3cc4c0b5eadd74d9974f1c1f5101045')
 
 
 @is_download_test

test/test_update.py.disabled

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys

test/test_utils.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
-import contextlib
 import os
 import sys
 import unittest
@@ -8,19 +8,16 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
-# Various small unit tests
+import contextlib
 import io
 import itertools
 import json
 import xml.etree.ElementTree
 
 from yt_dlp.compat import (
-    compat_chr,
     compat_etree_fromstring,
-    compat_getenv,
     compat_HTMLParseError,
     compat_os_name,
-    compat_setenv,
 )
 from yt_dlp.utils import (
     Config,
@@ -266,20 +263,20 @@ class TestUtil(unittest.TestCase):
         def env(var):
             return f'%{var}%' if sys.platform == 'win32' else f'${var}'
 
-        compat_setenv('yt_dlp_EXPATH_PATH', 'expanded')
+        os.environ['yt_dlp_EXPATH_PATH'] = 'expanded'
         self.assertEqual(expand_path(env('yt_dlp_EXPATH_PATH')), 'expanded')
 
         old_home = os.environ.get('HOME')
         test_str = R'C:\Documents and Settings\тест\Application Data'
         try:
-            compat_setenv('HOME', test_str)
-            self.assertEqual(expand_path(env('HOME')), compat_getenv('HOME'))
-            self.assertEqual(expand_path('~'), compat_getenv('HOME'))
+            os.environ['HOME'] = test_str
+            self.assertEqual(expand_path(env('HOME')), os.getenv('HOME'))
+            self.assertEqual(expand_path('~'), os.getenv('HOME'))
             self.assertEqual(
                 expand_path('~/%s' % env('yt_dlp_EXPATH_PATH')),
-                '%s/expanded' % compat_getenv('HOME'))
+                '%s/expanded' % os.getenv('HOME'))
         finally:
-            compat_setenv('HOME', old_home or '')
+            os.environ['HOME'] = old_home or ''
 
     def test_prepend_extension(self):
         self.assertEqual(prepend_extension('abc.ext', 'temp'), 'abc.temp.ext')
@@ -1128,7 +1125,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(extract_attributes('<e x="décompose&#769;">'), {'x': 'décompose\u0301'})
         # "Narrow" Python builds don't support unicode code points outside BMP.
         try:
-            compat_chr(0x10000)
+            chr(0x10000)
             supports_outside_bmp = True
         except ValueError:
             supports_outside_bmp = False

test/test_verbose_output.py

@@ -1,11 +1,15 @@
 #!/usr/bin/env python3
+
+# Allow direct execution
 import os
-import subprocess
 import sys
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import subprocess
+
 rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 
 

test/test_write_annotations.py.disabled

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,11 +7,12 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
 import xml.etree.ElementTree
-from test.helper import get_params, is_download_test, try_rm
 
 import yt_dlp.extractor
 import yt_dlp.YoutubeDL
+from test.helper import get_params, is_download_test, try_rm
 
 
 class YoutubeDL(yt_dlp.YoutubeDL):

test/test_youtube_lists.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys
@@ -6,8 +7,8 @@ import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL, is_download_test
 
+from test.helper import FakeYDL, is_download_test
 from yt_dlp.extractor import YoutubeIE, YoutubeTabIE
 
 

test/test_youtube_misc.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
 import os
 import sys

test/test_youtube_signature.py

@@ -1,18 +1,19 @@
 #!/usr/bin/env python3
+
 # Allow direct execution
-import contextlib
 import os
 import sys
 import unittest
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+
+import contextlib
 import re
 import string
 import urllib.request
-from test.helper import FakeYDL, is_download_test
 
-from yt_dlp.compat import compat_str
+from test.helper import FakeYDL, is_download_test
 from yt_dlp.extractor import YoutubeIE
 from yt_dlp.jsinterp import JSInterpreter
 
@@ -157,7 +158,7 @@ def t_factory(name, sig_func, url_pattern):
 def signature(jscode, sig_input):
     func = YoutubeIE(FakeYDL())._parse_sig_js(jscode)
     src_sig = (
-        compat_str(string.printable[:sig_input])
+        str(string.printable[:sig_input])
         if isinstance(sig_input, int) else sig_input)
     return func(src_sig)
 

tox.ini

@@ -1,16 +0,0 @@
-[tox]
-envlist = py26,py27,py33,py34,py35
-
-# Needed?
-[testenv]
-deps =
-   nose
-   coverage
-# We need a valid $HOME for test_compat_expanduser
-passenv = HOME
-defaultargs = test --exclude test_download.py --exclude test_age_restriction.py
-    --exclude test_subtitles.py --exclude test_write_annotations.py
-    --exclude test_youtube_lists.py --exclude test_iqiyi_sdk_interpreter.py
-    --exclude test_socks.py
-commands = nosetests --verbose {posargs:{[testenv]defaultargs}}  # --with-coverage --cover-package=yt_dlp --cover-html
-                                               # test.test_download:TestDownload.test_NowVideo

yt-dlp.sh

@@ -1,2 +1,2 @@
-#!/bin/sh
+#!/usr/bin/env sh
 exec "${PYTHON:-python3}" -bb -Werror -Xdev "$(dirname "$(realpath "$0")")/yt_dlp/__main__.py" "$@"

yt_dlp/__init__.py

@@ -1,17 +1,19 @@
-#!/usr/bin/env python3
 f'You are using an unsupported version of Python. Only Python versions 3.6 and above are supported by yt-dlp'  # noqa: F541
 
 __license__ = 'Public Domain'
 
+import getpass
 import itertools
+import optparse
 import os
 import re
 import sys
 
-from .compat import compat_getpass, compat_os_name, compat_shlex_quote
+from .compat import compat_shlex_quote
 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
 from .downloader import FileDownloader
-from .extractor import GenericIE, list_extractor_classes
+from .downloader.external import get_external_downloader
+from .extractor import list_extractor_classes
 from .extractor.adobepass import MSO_INFO
 from .extractor.common import InfoExtractor
 from .options import parseOpts
@@ -24,7 +26,7 @@ from .postprocessor import (
     MetadataFromFieldPP,
     MetadataParserPP,
 )
-from .update import run_update
+from .update import Updater
 from .utils import (
     NO_DEFAULT,
     POSTPROCESS_WHEN,
@@ -32,41 +34,47 @@ from .utils import (
     DownloadCancelled,
     DownloadError,
     GeoUtils,
+    PlaylistEntries,
     SameFileError,
     decodeOption,
+    download_range_func,
     expand_path,
     float_or_none,
+    format_field,
     int_or_none,
     match_filter_func,
     parse_duration,
     preferredencoding,
     read_batch_urls,
+    read_stdin,
     render_table,
     setproctitle,
     std_headers,
     traverse_obj,
+    variadic,
     write_string,
 )
 from .YoutubeDL import YoutubeDL
 
 
+def _exit(status=0, *args):
+    for msg in args:
+        sys.stderr.write(msg)
+    raise SystemExit(status)
+
+
 def get_urls(urls, batchfile, verbose):
     # Batch file verification
     batch_urls = []
     if batchfile is not None:
         try:
-            if batchfile == '-':
-                write_string('Reading URLs from stdin - EOF (%s) to end:\n' % (
-                    'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D'))
-                batchfd = sys.stdin
-            else:
-                batchfd = open(
-                    expand_path(batchfile), encoding='utf-8', errors='ignore')
-            batch_urls = read_batch_urls(batchfd)
+            batch_urls = read_batch_urls(
+                read_stdin('URLs') if batchfile == '-'
+                else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
             if verbose:
                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
         except OSError:
-            sys.exit('ERROR: batch file %s could not be read' % batchfile)
+            _exit(f'ERROR: batch file {batchfile} could not be read')
     _enc = preferredencoding()
     return [
         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
@@ -74,6 +82,10 @@ def get_urls(urls, batchfile, verbose):
 
 
 def print_extractor_information(opts, urls):
+    # Importing GenericIE is currently slow since it imports other extractors
+    # TODO: Move this back to module level after generalization of embed detection
+    from .extractor.generic import GenericIE
+
     out = ''
     if opts.list_extractors:
         urls = dict.fromkeys(urls, False)
@@ -209,15 +221,11 @@ def validate_options(opts):
         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 
     # Postprocessor formats
-    validate_in('audio format', opts.audioformat, ['best'] + list(FFmpegExtractAudioPP.SUPPORTED_EXTS))
+    validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
-    validate_in('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS)
-    if opts.recodevideo is not None:
-        opts.recodevideo = opts.recodevideo.replace(' ', '')
-        validate_regex('video recode format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
-    if opts.remuxvideo is not None:
-        opts.remuxvideo = opts.remuxvideo.replace(' ', '')
-        validate_regex('video remux format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
+    validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
+    validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
+    validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
     if opts.audioquality:
         opts.audioquality = opts.audioquality.strip('k').strip('K')
         # int_or_none prevents inf, nan
@@ -239,6 +247,28 @@ def validate_options(opts):
     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 
+    # Retry sleep function
+    def parse_sleep_func(expr):
+        NUMBER_RE = r'\d+(?:\.\d+)?'
+        op, start, limit, step, *_ = tuple(re.fullmatch(
+            rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
+            expr.strip()).groups()) + (None, None)
+
+        if op == 'exp':
+            return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
+        else:
+            default_step = start if op or limit else 0
+            return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
+
+    for key, expr in opts.retry_sleep.items():
+        if not expr:
+            del opts.retry_sleep[key]
+            continue
+        try:
+            opts.retry_sleep[key] = parse_sleep_func(expr)
+        except AttributeError:
+            raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
+
     # Bytes
     def parse_bytes(name, value):
         if value is None:
@@ -283,20 +313,25 @@ def validate_options(opts):
             'Cannot download a video and extract audio into the same file! '
             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 
-    # Remove chapters
-    remove_chapters_patterns, opts.remove_ranges = [], []
-    for regex in opts.remove_chapters or []:
-        if regex.startswith('*'):
-            dur = list(map(parse_duration, regex[1:].split('-')))
-            if len(dur) == 2 and all(t is not None for t in dur):
-                opts.remove_ranges.append(tuple(dur))
+    def parse_chapters(name, value):
+        chapters, ranges = [], []
+        for regex in value or []:
+            if regex.startswith('*'):
+                for range in regex[1:].split(','):
+                    dur = tuple(map(parse_duration, range.strip().split('-')))
+                    if len(dur) == 2 and all(t is not None for t in dur):
+                        ranges.append(dur)
+                    else:
+                        raise ValueError(f'invalid {name} time range "{regex}". Must be of the form *start-end')
                 continue
-            raise ValueError(f'invalid --remove-chapters time range "{regex}". Must be of the form *start-end')
-        try:
-            remove_chapters_patterns.append(re.compile(regex))
-        except re.error as err:
-            raise ValueError(f'invalid --remove-chapters regex "{regex}" - {err}')
-    opts.remove_chapters = remove_chapters_patterns
+            try:
+                chapters.append(re.compile(regex))
+            except re.error as err:
+                raise ValueError(f'invalid {name} regex "{regex}" - {err}')
+        return chapters, ranges
+
+    opts.remove_chapters, opts.remove_ranges = parse_chapters('--remove-chapters', opts.remove_chapters)
+    opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges))
 
     # Cookies from browser
     if opts.cookiesfrombrowser:
@@ -340,6 +375,12 @@ def validate_options(opts):
     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 
     # Other options
+    if opts.playlist_items is not None:
+        try:
+            tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
+        except Exception as err:
+            raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
+
     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
     if geo_bypass_code is not None:
         try:
@@ -360,6 +401,17 @@ def validate_options(opts):
     if opts.no_sponsorblock:
         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 
+    default_downloader = None
+    for proto, path in opts.external_downloader.items():
+        if path == 'native':
+            continue
+        ed = get_external_downloader(path)
+        if ed is None:
+            raise ValueError(
+                f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
+        elif ed and proto == 'default':
+            default_downloader = ed.get_basename()
+
     warnings, deprecation_warnings = [], []
 
     # Common mistake: -f best
@@ -370,13 +422,18 @@ def validate_options(opts):
             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 
     # --(postprocessor/downloader)-args without name
-    def report_args_compat(name, value, key1, key2=None):
+    def report_args_compat(name, value, key1, key2=None, where=None):
         if key1 in value and key2 not in value:
-            warnings.append(f'{name} arguments given without specifying name. The arguments will be given to all {name}s')
+            warnings.append(f'{name.title()} arguments given without specifying name. '
+                            f'The arguments will be given to {where or f"all {name}s"}')
             return True
         return False
 
-    report_args_compat('external downloader', opts.external_downloader_args, 'default')
+    if report_args_compat('external downloader', opts.external_downloader_args,
+                          'default', where=default_downloader) and default_downloader:
+        # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
+        opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
+
     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
         opts.postprocessor_args.setdefault('sponskrub', [])
@@ -395,6 +452,9 @@ def validate_options(opts):
         setattr(opts, opt1, default)
 
     # Conflicting options
+    report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
+    report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
+    report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
@@ -471,9 +531,9 @@ def validate_options(opts):
 
     # Ask for passwords
     if opts.username is not None and opts.password is None:
-        opts.password = compat_getpass('Type account password and press [Return]: ')
+        opts.password = getpass.getpass('Type account password and press [Return]: ')
     if opts.ap_username is not None and opts.ap_password is None:
-        opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
+        opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
 
     return warnings, deprecation_warnings
 
@@ -627,7 +687,7 @@ def parse_options(argv=None):
     final_ext = (
         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
-        else opts.audioformat if (opts.extractaudio and opts.audioformat != 'best')
+        else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
         else None)
 
     return parser, opts, urls, {
@@ -686,6 +746,7 @@ def parse_options(argv=None):
         'file_access_retries': opts.file_access_retries,
         'fragment_retries': opts.fragment_retries,
         'extractor_retries': opts.extractor_retries,
+        'retry_sleep_functions': opts.retry_sleep,
         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
         'keep_fragments': opts.keep_fragments,
         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
@@ -700,6 +761,7 @@ def parse_options(argv=None):
         'playlistend': opts.playlistend,
         'playlistreverse': opts.playlist_reverse,
         'playlistrandom': opts.playlist_random,
+        'lazy_playlist': opts.lazy_playlist,
         'noplaylist': opts.noplaylist,
         'logtostderr': opts.outtmpl.get('default') == '-',
         'consoletitle': opts.consoletitle,
@@ -731,6 +793,7 @@ def parse_options(argv=None):
         'verbose': opts.verbose,
         'dump_intermediate_pages': opts.dump_intermediate_pages,
         'write_pages': opts.write_pages,
+        'load_pages': opts.load_pages,
         'test': opts.test,
         'keepvideo': opts.keepvideo,
         'min_filesize': opts.min_filesize,
@@ -779,6 +842,8 @@ def parse_options(argv=None):
         'max_sleep_interval': opts.max_sleep_interval,
         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
         'external_downloader': opts.external_downloader,
+        'download_ranges': opts.download_ranges,
+        'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
         'list_thumbnails': opts.list_thumbnails,
         'playlist_items': opts.playlist_items,
         'xattr_set_filesize': opts.xattr_set_filesize,
@@ -810,62 +875,63 @@ def _real_main(argv=None):
     if opts.dump_user_agent:
         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
         write_string(f'{ua}\n', out=sys.stdout)
-        sys.exit(0)
+        return
 
     if print_extractor_information(opts, all_urls):
-        sys.exit(0)
+        return
 
     with YoutubeDL(ydl_opts) as ydl:
+        pre_process = opts.update_self or opts.rm_cachedir
         actual_use = all_urls or opts.load_info_filename
 
-        # Remove cache dir
         if opts.rm_cachedir:
             ydl.cache.remove()
 
-        # Update version
-        if opts.update_self:
-            # If updater returns True, exit. Required for windows
-            if run_update(ydl):
-                if actual_use:
-                    sys.exit('ERROR: The program must exit for the update to complete')
-                sys.exit()
+        updater = Updater(ydl)
+        if opts.update_self and updater.update() and actual_use:
+            if updater.cmd:
+                return updater.restart()
+            # This code is reachable only for zip variant in py < 3.10
+            # It makes sense to exit here, but the old behavior is to continue
+            ydl.report_warning('Restart yt-dlp to use the updated version')
+            # return 100, 'ERROR: The program must exit for the update to complete'
 
-        # Maybe do nothing
         if not actual_use:
-            if opts.update_self or opts.rm_cachedir:
-                sys.exit()
+            if pre_process:
+                return ydl._download_retcode
 
             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
             parser.error(
                 'You must provide at least one URL.\n'
                 'Type yt-dlp --help to see a list of all options.')
 
+        parser.destroy()
         try:
             if opts.load_info_filename is not None:
-                retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
+                return ydl.download_with_info_file(expand_path(opts.load_info_filename))
             else:
-                retcode = ydl.download(all_urls)
+                return ydl.download(all_urls)
         except DownloadCancelled:
             ydl.to_screen('Aborting remaining downloads')
-            retcode = 101
-
-    sys.exit(retcode)
+            return 101
 
 
 def main(argv=None):
     try:
-        _real_main(argv)
+        _exit(*variadic(_real_main(argv)))
     except DownloadError:
-        sys.exit(1)
+        _exit(1)
     except SameFileError as e:
-        sys.exit(f'ERROR: {e}')
+        _exit(f'ERROR: {e}')
     except KeyboardInterrupt:
-        sys.exit('\nERROR: Interrupted by user')
+        _exit('\nERROR: Interrupted by user')
     except BrokenPipeError as e:
         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
         devnull = os.open(os.devnull, os.O_WRONLY)
         os.dup2(devnull, sys.stdout.fileno())
-        sys.exit(f'\nERROR: {e}')
+        _exit(f'\nERROR: {e}')
+    except optparse.OptParseError as e:
+        _exit(2, f'\n{e}')
 
 
 from .extractor import gen_extractors, list_extractors

yt_dlp/__main__.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+
 # Execute with
 # $ python -m yt_dlp
 

yt_dlp/aes.py

@@ -1,6 +1,7 @@
+import base64
 from math import ceil
 
-from .compat import compat_b64decode, compat_ord
+from .compat import compat_ord
 from .dependencies import Cryptodome_AES
 from .utils import bytes_to_intlist, intlist_to_bytes
 
@@ -264,7 +265,7 @@ def aes_decrypt_text(data, password, key_size_bytes):
     """
     NONCE_LENGTH_BYTES = 8
 
-    data = bytes_to_intlist(compat_b64decode(data))
+    data = bytes_to_intlist(base64.b64decode(data))
     password = bytes_to_intlist(password.encode())
 
     key = password[:key_size_bytes] + [0] * (key_size_bytes - len(password))

yt_dlp/cache.py

@@ -6,7 +6,6 @@ import re
 import shutil
 import traceback
 
-from .compat import compat_getenv
 from .utils import expand_path, write_json_file
 
 
@@ -17,7 +16,7 @@ class Cache:
     def _get_root_dir(self):
         res = self._ydl.params.get('cachedir')
         if res is None:
-            cache_root = compat_getenv('XDG_CACHE_HOME', '~/.cache')
+            cache_root = os.getenv('XDG_CACHE_HOME', '~/.cache')
             res = os.path.join(cache_root, 'yt-dlp')
         return expand_path(res)
 

yt_dlp/compat/__init__.py

@@ -1,6 +1,4 @@
-import contextlib
 import os
-import subprocess
 import sys
 import warnings
 import xml.etree.ElementTree as etree
@@ -9,10 +7,14 @@ from . import re
 from ._deprecated import *  # noqa: F401, F403
 from .compat_utils import passthrough_module
 
-
 # XXX: Implement this the same way as other DeprecationWarnings without circular import
-passthrough_module(__name__, '._legacy', callback=lambda attr: warnings.warn(
-    DeprecationWarning(f'{__name__}.{attr} is deprecated'), stacklevel=2))
+try:
+    passthrough_module(__name__, '._legacy', callback=lambda attr: warnings.warn(
+        DeprecationWarning(f'{__name__}.{attr} is deprecated'), stacklevel=2))
+    HAS_LEGACY = True
+except ModuleNotFoundError:
+    # Keep working even without _legacy module
+    HAS_LEGACY = False
 del passthrough_module
 
 
@@ -52,7 +54,7 @@ if compat_os_name == 'nt' and sys.version_info < (3, 8):
     def compat_realpath(path):
         while os.path.islink(path):
             path = os.path.abspath(os.readlink(path))
-        return path
+        return os.path.realpath(path)
 else:
     compat_realpath = os.path.realpath
 
@@ -74,17 +76,3 @@ if compat_os_name in ('nt', 'ce'):
         return userhome + path[i:]
 else:
     compat_expanduser = os.path.expanduser
-
-
-WINDOWS_VT_MODE = False if compat_os_name == 'nt' else None
-
-
-def windows_enable_vt_mode():  # TODO: Do this the proper way https://bugs.python.org/issue30075
-    if compat_os_name != 'nt':
-        return
-    global WINDOWS_VT_MODE
-    startupinfo = subprocess.STARTUPINFO()
-    startupinfo.dwFlags |= subprocess.STARTF_USESHOWWINDOW
-    with contextlib.suppress(Exception):
-        subprocess.Popen('', shell=True, startupinfo=startupinfo).wait()
-        WINDOWS_VT_MODE = True

yt_dlp/compat/_deprecated.py

@@ -1,52 +1,16 @@
 """Deprecated - New code should avoid these"""
 
 import base64
-import getpass
-import html
-import html.parser
-import http
-import http.client
-import http.cookiejar
-import http.cookies
-import http.server
-import itertools
-import os
-import shutil
-import struct
-import tokenize
-import urllib
+import urllib.error
+import urllib.parse
+
+compat_str = str
 
 compat_b64decode = base64.b64decode
-compat_chr = chr
-compat_cookiejar = http.cookiejar
-compat_cookiejar_Cookie = http.cookiejar.Cookie
-compat_cookies_SimpleCookie = http.cookies.SimpleCookie
-compat_get_terminal_size = shutil.get_terminal_size
-compat_getenv = os.getenv
-compat_getpass = getpass.getpass
-compat_html_entities = html.entities
-compat_html_entities_html5 = html.entities.html5
-compat_HTMLParser = html.parser.HTMLParser
-compat_http_client = http.client
-compat_http_server = http.server
+
 compat_HTTPError = urllib.error.HTTPError
-compat_itertools_count = itertools.count
+compat_urlparse = urllib.parse
 compat_parse_qs = urllib.parse.parse_qs
-compat_str = str
-compat_struct_pack = struct.pack
-compat_struct_unpack = struct.unpack
-compat_tokenize_tokenize = tokenize.tokenize
-compat_urllib_error = urllib.error
 compat_urllib_parse_unquote = urllib.parse.unquote
-compat_urllib_parse_unquote_plus = urllib.parse.unquote_plus
 compat_urllib_parse_urlencode = urllib.parse.urlencode
 compat_urllib_parse_urlparse = urllib.parse.urlparse
-compat_urllib_request = urllib.request
-compat_urlparse = compat_urllib_parse = urllib.parse
-
-
-def compat_setenv(key, value, env=os.environ):
-    env[key] = value
-
-
-__all__ = [x for x in globals() if x.startswith('compat_')]

yt_dlp/compat/_legacy.py

@@ -2,18 +2,27 @@
 
 import collections
 import ctypes
-import http
+import getpass
+import html.entities
+import html.parser
 import http.client
 import http.cookiejar
 import http.cookies
 import http.server
+import itertools
+import os
 import shlex
+import shutil
 import socket
 import struct
-import urllib
+import tokenize
+import urllib.error
+import urllib.parse
+import urllib.request
 import xml.etree.ElementTree as etree
 from subprocess import DEVNULL
 
+from .compat_utils import passthrough_module  # isort: split
 from .asyncio import run as compat_asyncio_run  # noqa: F401
 from .re import Pattern as compat_Pattern  # noqa: F401
 from .re import match as compat_Match  # noqa: F401
@@ -21,6 +30,8 @@ from ..dependencies import Cryptodome_AES as compat_pycrypto_AES  # noqa: F401
 from ..dependencies import brotli as compat_brotli  # noqa: F401
 from ..dependencies import websockets as compat_websockets  # noqa: F401
 
+passthrough_module(__name__, '...utils', ('WINDOWS_VT_MODE', 'windows_enable_vt_mode'))
+
 
 # compat_ctypes_WINFUNCTYPE = ctypes.WINFUNCTYPE
 # will not work since ctypes.WINFUNCTYPE does not exist in UNIX machines
@@ -28,14 +39,31 @@ def compat_ctypes_WINFUNCTYPE(*args, **kwargs):
     return ctypes.WINFUNCTYPE(*args, **kwargs)
 
 
+def compat_setenv(key, value, env=os.environ):
+    env[key] = value
+
+
 compat_basestring = str
+compat_chr = chr
 compat_collections_abc = collections.abc
+compat_cookiejar = http.cookiejar
+compat_cookiejar_Cookie = http.cookiejar.Cookie
 compat_cookies = http.cookies
+compat_cookies_SimpleCookie = http.cookies.SimpleCookie
 compat_etree_Element = etree.Element
 compat_etree_register_namespace = etree.register_namespace
 compat_filter = filter
+compat_get_terminal_size = shutil.get_terminal_size
+compat_getenv = os.getenv
+compat_getpass = getpass.getpass
+compat_html_entities = html.entities
+compat_html_entities_html5 = html.entities.html5
+compat_HTMLParser = html.parser.HTMLParser
+compat_http_client = http.client
+compat_http_server = http.server
 compat_input = input
 compat_integer_types = (int, )
+compat_itertools_count = itertools.count
 compat_kwargs = lambda kwargs: kwargs
 compat_map = map
 compat_numeric_types = (int, float, complex)
@@ -43,11 +71,18 @@ compat_print = print
 compat_shlex_split = shlex.split
 compat_socket_create_connection = socket.create_connection
 compat_Struct = struct.Struct
+compat_struct_pack = struct.pack
+compat_struct_unpack = struct.unpack
 compat_subprocess_get_DEVNULL = lambda: DEVNULL
+compat_tokenize_tokenize = tokenize.tokenize
+compat_urllib_error = urllib.error
+compat_urllib_parse = urllib.parse
 compat_urllib_parse_quote = urllib.parse.quote
 compat_urllib_parse_quote_plus = urllib.parse.quote_plus
+compat_urllib_parse_unquote_plus = urllib.parse.unquote_plus
 compat_urllib_parse_unquote_to_bytes = urllib.parse.unquote_to_bytes
 compat_urllib_parse_urlunparse = urllib.parse.urlunparse
+compat_urllib_request = urllib.request
 compat_urllib_request_DataHandler = urllib.request.DataHandler
 compat_urllib_response = urllib.response
 compat_urlretrieve = urllib.request.urlretrieve

yt_dlp/compat/compat_utils.py

@@ -4,7 +4,6 @@ import importlib
 import sys
 import types
 
-
 _NO_ATTRIBUTE = object()
 
 _Package = collections.namedtuple('Package', ('name', 'version'))
@@ -31,9 +30,9 @@ def _is_package(module):
     return True
 
 
-def passthrough_module(parent, child, *, callback=lambda _: None):
+def passthrough_module(parent, child, allowed_attributes=None, *, callback=lambda _: None):
     parent_module = importlib.import_module(parent)
-    child_module = importlib.import_module(child, parent)
+    child_module = None  # Import child module only as needed
 
     class PassthroughModule(types.ModuleType):
         def __getattr__(self, attr):
@@ -41,19 +40,30 @@ def passthrough_module(parent, child, *, callback=lambda _: None):
                 with contextlib.suppress(ImportError):
                     return importlib.import_module(f'.{attr}', parent)
 
-            ret = _NO_ATTRIBUTE
+            ret = self.__from_child(attr)
+            if ret is _NO_ATTRIBUTE:
+                raise AttributeError(f'module {parent} has no attribute {attr}')
+            callback(attr)
+            return ret
+
+        def __from_child(self, attr):
+            if allowed_attributes is None:
+                if attr.startswith('__') and attr.endswith('__'):
+                    return _NO_ATTRIBUTE
+            elif attr not in allowed_attributes:
+                return _NO_ATTRIBUTE
+
+            nonlocal child_module
+            child_module = child_module or importlib.import_module(child, parent)
+
             with contextlib.suppress(AttributeError):
-                ret = getattr(child_module, attr)
+                return getattr(child_module, attr)
 
             if _is_package(child_module):
                 with contextlib.suppress(ImportError):
-                    ret = importlib.import_module(f'.{attr}', child)
+                    return importlib.import_module(f'.{attr}', child)
 
-            if ret is _NO_ATTRIBUTE:
-                raise AttributeError(f'module {parent} has no attribute {attr}')
-
-            callback(attr)
-            return ret
+            return _NO_ATTRIBUTE
 
     # Python 3.6 does not have module level __getattr__
     # https://peps.python.org/pep-0562/

yt_dlp/compat/functools.py

@@ -0,0 +1,26 @@
+# flake8: noqa: F405
+from functools import *  # noqa: F403
+
+from .compat_utils import passthrough_module
+
+passthrough_module(__name__, 'functools')
+del passthrough_module
+
+try:
+    cache  # >= 3.9
+except NameError:
+    cache = lru_cache(maxsize=None)
+
+try:
+    cached_property  # >= 3.8
+except NameError:
+    class cached_property:
+        def __init__(self, func):
+            update_wrapper(self, func)
+            self.func = func
+
+        def __get__(self, instance, _):
+            if instance is None:
+                return self
+            setattr(instance, self.func.__name__, self.func(instance))
+            return getattr(instance, self.func.__name__)

yt_dlp/cookies.py

@@ -1,5 +1,7 @@
+import base64
 import contextlib
 import ctypes
+import http.cookiejar
 import json
 import os
 import shutil
@@ -17,7 +19,6 @@ from .aes import (
     aes_gcm_decrypt_and_verify_bytes,
     unpad_pkcs7,
 )
-from .compat import compat_b64decode, compat_cookiejar_Cookie
 from .dependencies import (
     _SECRETSTORAGE_UNAVAILABLE_REASON,
     secretstorage,
@@ -142,7 +143,7 @@ def _extract_firefox_cookies(profile, logger):
                 total_cookie_count = len(table)
                 for i, (host, name, value, path, expiry, is_secure) in enumerate(table):
                     progress_bar.print(f'Loading cookie {i: 6d}/{total_cookie_count: 6d}')
-                    cookie = compat_cookiejar_Cookie(
+                    cookie = http.cookiejar.Cookie(
                         version=0, name=name, value=value, port=None, port_specified=False,
                         domain=host, domain_specified=bool(host), domain_initial_dot=host.startswith('.'),
                         path=path, path_specified=bool(path), secure=is_secure, expires=expiry, discard=False,
@@ -156,30 +157,16 @@ def _extract_firefox_cookies(profile, logger):
 
 
 def _firefox_browser_dir():
-    if sys.platform in ('linux', 'linux2'):
-        return os.path.expanduser('~/.mozilla/firefox')
-    elif sys.platform == 'win32':
+    if sys.platform in ('cygwin', 'win32'):
         return os.path.expandvars(R'%APPDATA%\Mozilla\Firefox\Profiles')
     elif sys.platform == 'darwin':
         return os.path.expanduser('~/Library/Application Support/Firefox')
-    else:
-        raise ValueError(f'unsupported platform: {sys.platform}')
+    return os.path.expanduser('~/.mozilla/firefox')
 
 
 def _get_chromium_based_browser_settings(browser_name):
     # https://chromium.googlesource.com/chromium/src/+/HEAD/docs/user_data_dir.md
-    if sys.platform in ('linux', 'linux2'):
-        config = _config_home()
-        browser_dir = {
-            'brave': os.path.join(config, 'BraveSoftware/Brave-Browser'),
-            'chrome': os.path.join(config, 'google-chrome'),
-            'chromium': os.path.join(config, 'chromium'),
-            'edge': os.path.join(config, 'microsoft-edge'),
-            'opera': os.path.join(config, 'opera'),
-            'vivaldi': os.path.join(config, 'vivaldi'),
-        }[browser_name]
-
-    elif sys.platform == 'win32':
+    if sys.platform in ('cygwin', 'win32'):
         appdata_local = os.path.expandvars('%LOCALAPPDATA%')
         appdata_roaming = os.path.expandvars('%APPDATA%')
         browser_dir = {
@@ -203,7 +190,15 @@ def _get_chromium_based_browser_settings(browser_name):
         }[browser_name]
 
     else:
-        raise ValueError(f'unsupported platform: {sys.platform}')
+        config = _config_home()
+        browser_dir = {
+            'brave': os.path.join(config, 'BraveSoftware/Brave-Browser'),
+            'chrome': os.path.join(config, 'google-chrome'),
+            'chromium': os.path.join(config, 'chromium'),
+            'edge': os.path.join(config, 'microsoft-edge'),
+            'opera': os.path.join(config, 'opera'),
+            'vivaldi': os.path.join(config, 'vivaldi'),
+        }[browser_name]
 
     # Linux keyring names can be determined by snooping on dbus while opening the browser in KDE:
     # dbus-monitor "interface='org.kde.KWallet'" "type=method_return"
@@ -303,7 +298,7 @@ def _process_chrome_cookie(decryptor, host_key, name, value, encrypted_value, pa
         if value is None:
             return is_encrypted, None
 
-    return is_encrypted, compat_cookiejar_Cookie(
+    return is_encrypted, http.cookiejar.Cookie(
         version=0, name=name, value=value, port=None, port_specified=False,
         domain=host_key, domain_specified=bool(host_key), domain_initial_dot=host_key.startswith('.'),
         path=path, path_specified=bool(path), secure=is_secure, expires=expires_utc, discard=False,
@@ -343,14 +338,11 @@ class ChromeCookieDecryptor:
 
 
 def get_cookie_decryptor(browser_root, browser_keyring_name, logger, *, keyring=None):
-    if sys.platform in ('linux', 'linux2'):
-        return LinuxChromeCookieDecryptor(browser_keyring_name, logger, keyring=keyring)
-    elif sys.platform == 'darwin':
+    if sys.platform == 'darwin':
         return MacChromeCookieDecryptor(browser_keyring_name, logger)
-    elif sys.platform == 'win32':
+    elif sys.platform in ('win32', 'cygwin'):
         return WindowsChromeCookieDecryptor(browser_root, logger)
-    else:
-        raise NotImplementedError(f'Chrome cookie decryption is not supported on this platform: {sys.platform}')
+    return LinuxChromeCookieDecryptor(browser_keyring_name, logger, keyring=keyring)
 
 
 class LinuxChromeCookieDecryptor(ChromeCookieDecryptor):
@@ -598,7 +590,7 @@ def _parse_safari_cookies_record(data, jar, logger):
 
     p.skip_to(record_size, 'space at the end of the record')
 
-    cookie = compat_cookiejar_Cookie(
+    cookie = http.cookiejar.Cookie(
         version=0, name=name, value=value, port=None, port_specified=False,
         domain=domain, domain_specified=bool(domain), domain_initial_dot=domain.startswith('.'),
         path=path, path_specified=bool(path), secure=is_secure, expires=expiration_date, discard=False,
@@ -718,21 +710,19 @@ def _get_kwallet_network_wallet(logger):
     """
     default_wallet = 'kdewallet'
     try:
-        proc = Popen([
+        stdout, _, returncode = Popen.run([
             'dbus-send', '--session', '--print-reply=literal',
             '--dest=org.kde.kwalletd5',
             '/modules/kwalletd5',
             'org.kde.KWallet.networkWallet'
-        ], stdout=subprocess.PIPE, stderr=subprocess.DEVNULL)
+        ], text=True, stdout=subprocess.PIPE, stderr=subprocess.DEVNULL)
 
-        stdout, stderr = proc.communicate_or_kill()
-        if proc.returncode != 0:
+        if returncode:
             logger.warning('failed to read NetworkWallet')
             return default_wallet
         else:
-            network_wallet = stdout.decode().strip()
-            logger.debug(f'NetworkWallet = "{network_wallet}"')
-            return network_wallet
+            logger.debug(f'NetworkWallet = "{stdout.strip()}"')
+            return stdout.strip()
     except Exception as e:
         logger.warning(f'exception while obtaining NetworkWallet: {e}')
         return default_wallet
@@ -750,17 +740,16 @@ def _get_kwallet_password(browser_keyring_name, logger):
     network_wallet = _get_kwallet_network_wallet(logger)
 
     try:
-        proc = Popen([
+        stdout, _, returncode = Popen.run([
             'kwallet-query',
             '--read-password', f'{browser_keyring_name} Safe Storage',
             '--folder', f'{browser_keyring_name} Keys',
             network_wallet
         ], stdout=subprocess.PIPE, stderr=subprocess.DEVNULL)
 
-        stdout, stderr = proc.communicate_or_kill()
-        if proc.returncode != 0:
-            logger.error(f'kwallet-query failed with return code {proc.returncode}. Please consult '
-                         'the kwallet-query man page for details')
+        if returncode:
+            logger.error(f'kwallet-query failed with return code {returncode}. '
+                         'Please consult the kwallet-query man page for details')
             return b''
         else:
             if stdout.lower().startswith(b'failed to read'):
@@ -775,9 +764,7 @@ def _get_kwallet_password(browser_keyring_name, logger):
                 return b''
             else:
                 logger.debug('password found')
-                if stdout[-1:] == b'\n':
-                    stdout = stdout[:-1]
-                return stdout
+                return stdout.rstrip(b'\n')
     except Exception as e:
         logger.warning(f'exception running kwallet-query: {error_to_str(e)}')
         return b''
@@ -824,17 +811,13 @@ def _get_linux_keyring_password(browser_keyring_name, keyring, logger):
 def _get_mac_keyring_password(browser_keyring_name, logger):
     logger.debug('using find-generic-password to obtain password from OSX keychain')
     try:
-        proc = Popen(
+        stdout, _, _ = Popen.run(
             ['security', 'find-generic-password',
              '-w',  # write password to stdout
              '-a', browser_keyring_name,  # match 'account'
              '-s', f'{browser_keyring_name} Safe Storage'],  # match 'service'
             stdout=subprocess.PIPE, stderr=subprocess.DEVNULL)
-
-        stdout, stderr = proc.communicate_or_kill()
-        if stdout[-1:] == b'\n':
-            stdout = stdout[:-1]
-        return stdout
+        return stdout.rstrip(b'\n')
     except Exception as e:
         logger.warning(f'exception running find-generic-password: {error_to_str(e)}')
         return None
@@ -853,7 +836,7 @@ def _get_windows_v10_key(browser_root, logger):
     except KeyError:
         logger.error('no encrypted key in Local State')
         return None
-    encrypted_key = compat_b64decode(base64_key)
+    encrypted_key = base64.b64decode(base64_key)
     prefix = b'DPAPI'
     if not encrypted_key.startswith(prefix):
         logger.error('invalid key')

yt_dlp/dependencies.py

@@ -1,6 +1,6 @@
 # flake8: noqa: F401
 """Imports all optional dependencies for the project.
-An attribute "_yt_dlp__identifier" may be inserted into the module if it uses an ambigious namespace"""
+An attribute "_yt_dlp__identifier" may be inserted into the module if it uses an ambiguous namespace"""
 
 try:
     import brotlicffi as brotli

yt_dlp/downloader/__init__.py

@@ -1,4 +1,3 @@
-from ..compat import compat_str
 from ..utils import NO_DEFAULT, determine_protocol
 
 
@@ -60,10 +59,11 @@ PROTOCOL_MAP = {
 
 def shorten_protocol_name(proto, simplify=False):
     short_protocol_names = {
-        'm3u8_native': 'm3u8_n',
-        'rtmp_ffmpeg': 'rtmp_f',
+        'm3u8_native': 'm3u8',
+        'm3u8': 'm3u8F',
+        'rtmp_ffmpeg': 'rtmpF',
         'http_dash_segments': 'dash',
-        'http_dash_segments_generator': 'dash_g',
+        'http_dash_segments_generator': 'dashG',
         'niconico_dmc': 'dmc',
         'websocket_frag': 'WSfrag',
     }
@@ -71,6 +71,7 @@ def shorten_protocol_name(proto, simplify=False):
         short_protocol_names.update({
             'https': 'http',
             'ftps': 'ftp',
+            'm3u8': 'm3u8',  # Reverse above m3u8 mapping
             'm3u8_native': 'm3u8',
             'http_dash_segments_generator': 'dash',
             'rtmp_ffmpeg': 'rtmp',
@@ -85,13 +86,13 @@ def _get_suitable_downloader(info_dict, protocol, params, default):
     if default is NO_DEFAULT:
         default = HttpFD
 
-    # if (info_dict.get('start_time') or info_dict.get('end_time')) and not info_dict.get('requested_formats') and FFmpegFD.can_download(info_dict):
-    #     return FFmpegFD
+    if (info_dict.get('section_start') or info_dict.get('section_end')) and FFmpegFD.can_download(info_dict):
+        return FFmpegFD
 
     info_dict['protocol'] = protocol
     downloaders = params.get('external_downloader')
     external_downloader = (
-        downloaders if isinstance(downloaders, compat_str) or downloaders is None
+        downloaders if isinstance(downloaders, str) or downloaders is None
         else downloaders.get(shorten_protocol_name(protocol, True), downloaders.get('default')))
 
     if external_downloader is None:

yt_dlp/downloader/common.py

@@ -15,14 +15,18 @@ from ..utils import (
     NUMBER_RE,
     LockingUnsupportedError,
     Namespace,
+    classproperty,
     decodeArgument,
     encodeFilename,
     error_to_compat_str,
+    float_or_none,
     format_bytes,
+    join_nonempty,
     sanitize_open,
     shell_quote,
     timeconvert,
     timetuple_from_msec,
+    try_call,
 )
 
 
@@ -41,6 +45,7 @@ class FileDownloader:
     verbose:            Print additional info to stdout.
     quiet:              Do not print messages to stdout.
     ratelimit:          Download speed limit, in bytes/sec.
+    continuedl:         Attempt to continue downloads if possible
     throttledratelimit: Assume the download is being throttled below this speed (bytes/sec)
     retries:            Number of times to retry for HTTP error 5xx
     file_access_retries:   Number of times to retry on file access error
@@ -64,6 +69,7 @@ class FileDownloader:
                         useful for bypassing bandwidth throttling imposed by
                         a webserver (experimental)
     progress_template:  See YoutubeDL.py
+    retry_sleep_functions: See YoutubeDL.py
 
     Subclasses of this one must re-define the real_download method.
     """
@@ -98,12 +104,16 @@ class FileDownloader:
     def to_screen(self, *args, **kargs):
         self.ydl.to_screen(*args, quiet=self.params.get('quiet'), **kargs)
 
-    @property
-    def FD_NAME(self):
-        return re.sub(r'(?<!^)(?=[A-Z])', '_', type(self).__name__[:-2]).lower()
+    __to_screen = to_screen
+
+    @classproperty
+    def FD_NAME(cls):
+        return re.sub(r'(?<=[a-z])(?=[A-Z])', '_', cls.__name__[:-2]).lower()
 
     @staticmethod
     def format_seconds(seconds):
+        if seconds is None:
+            return ' Unknown'
         time = timetuple_from_msec(seconds * 1000)
         if time.hours > 99:
             return '--:--:--'
@@ -111,6 +121,8 @@ class FileDownloader:
             return '%02d:%02d' % time[1:-1]
         return '%02d:%02d:%02d' % time[:-1]
 
+    format_eta = format_seconds
+
     @staticmethod
     def calc_percent(byte_counter, data_len):
         if data_len is None:
@@ -119,11 +131,7 @@ class FileDownloader:
 
     @staticmethod
     def format_percent(percent):
-        if percent is None:
-            return '---.-%'
-        elif percent == 100:
-            return '100%'
-        return '%6s' % ('%3.1f%%' % percent)
+        return '  N/A%' if percent is None else f'{percent:>5.1f}%'
 
     @staticmethod
     def calc_eta(start, now, total, current):
@@ -137,12 +145,6 @@ class FileDownloader:
         rate = float(current) / dif
         return int((float(total) - float(current)) / rate)
 
-    @staticmethod
-    def format_eta(eta):
-        if eta is None:
-            return '--:--'
-        return FileDownloader.format_seconds(eta)
-
     @staticmethod
     def calc_speed(start, now, bytes):
         dif = now - start
@@ -152,13 +154,11 @@ class FileDownloader:
 
     @staticmethod
     def format_speed(speed):
-        if speed is None:
-            return '%10s' % '---b/s'
-        return '%10s' % ('%s/s' % format_bytes(speed))
+        return ' Unknown B/s' if speed is None else f'{format_bytes(speed):>10s}/s'
 
     @staticmethod
     def format_retries(retries):
-        return 'inf' if retries == float('inf') else '%.0f' % retries
+        return 'inf' if retries == float('inf') else int(retries)
 
     @staticmethod
     def best_block_size(elapsed_time, bytes):
@@ -232,7 +232,8 @@ class FileDownloader:
                         self.to_screen(
                             f'[download] Unable to {action} file due to file access error. '
                             f'Retrying (attempt {retry} of {self.format_retries(file_access_retries)}) ...')
-                        time.sleep(0.01)
+                        if not self.sleep_retry('file_access', retry):
+                            time.sleep(0.01)
             return inner
         return outer
 
@@ -282,9 +283,9 @@ class FileDownloader:
         elif self.ydl.params.get('logger'):
             self._multiline = MultilineLogger(self.ydl.params['logger'], lines)
         elif self.params.get('progress_with_newline'):
-            self._multiline = BreaklineStatusPrinter(self.ydl._out_files.screen, lines)
+            self._multiline = BreaklineStatusPrinter(self.ydl._out_files.out, lines)
         else:
-            self._multiline = MultilinePrinter(self.ydl._out_files.screen, lines, not self.params.get('quiet'))
+            self._multiline = MultilinePrinter(self.ydl._out_files.out, lines, not self.params.get('quiet'))
         self._multiline.allow_colors = self._multiline._HAVE_FULLCAP and not self.params.get('no_color')
 
     def _finish_multiline_status(self):
@@ -301,7 +302,7 @@ class FileDownloader:
     )
 
     def _report_progress_status(self, s, default_template):
-        for name, style in self.ProgressStyles:
+        for name, style in self.ProgressStyles.items_:
             name = f'_{name}_str'
             if name not in s:
                 continue
@@ -325,63 +326,52 @@ class FileDownloader:
             self._multiline.stream, self._multiline.allow_colors, *args, **kwargs)
 
     def report_progress(self, s):
+        def with_fields(*tups, default=''):
+            for *fields, tmpl in tups:
+                if all(s.get(f) is not None for f in fields):
+                    return tmpl
+            return default
+
         if s['status'] == 'finished':
             if self.params.get('noprogress'):
                 self.to_screen('[download] Download completed')
-            msg_template = '100%%'
-            if s.get('total_bytes') is not None:
-                s['_total_bytes_str'] = format_bytes(s['total_bytes'])
-                msg_template += ' of %(_total_bytes_str)s'
-            if s.get('elapsed') is not None:
-                s['_elapsed_str'] = self.format_seconds(s['elapsed'])
-                msg_template += ' in %(_elapsed_str)s'
-            s['_percent_str'] = self.format_percent(100)
-            self._report_progress_status(s, msg_template)
-            return
+            s.update({
+                '_total_bytes_str': format_bytes(s.get('total_bytes')),
+                '_elapsed_str': self.format_seconds(s.get('elapsed')),
+                '_percent_str': self.format_percent(100),
+            })
+            self._report_progress_status(s, join_nonempty(
+                '100%%',
+                with_fields(('total_bytes', 'of %(_total_bytes_str)s')),
+                with_fields(('elapsed', 'in %(_elapsed_str)s')),
+                delim=' '))
 
         if s['status'] != 'downloading':
             return
 
-        if s.get('eta') is not None:
-            s['_eta_str'] = self.format_eta(s['eta'])
-        else:
-            s['_eta_str'] = 'Unknown'
+        s.update({
+            '_eta_str': self.format_eta(s.get('eta')),
+            '_speed_str': self.format_speed(s.get('speed')),
+            '_percent_str': self.format_percent(try_call(
+                lambda: 100 * s['downloaded_bytes'] / s['total_bytes'],
+                lambda: 100 * s['downloaded_bytes'] / s['total_bytes_estimate'],
+                lambda: s['downloaded_bytes'] == 0 and 0)),
+            '_total_bytes_str': format_bytes(s.get('total_bytes')),
+            '_total_bytes_estimate_str': format_bytes(s.get('total_bytes_estimate')),
+            '_downloaded_bytes_str': format_bytes(s.get('downloaded_bytes')),
+            '_elapsed_str': self.format_seconds(s.get('elapsed')),
+        })
 
-        if s.get('total_bytes') and s.get('downloaded_bytes') is not None:
-            s['_percent_str'] = self.format_percent(100 * s['downloaded_bytes'] / s['total_bytes'])
-        elif s.get('total_bytes_estimate') and s.get('downloaded_bytes') is not None:
-            s['_percent_str'] = self.format_percent(100 * s['downloaded_bytes'] / s['total_bytes_estimate'])
-        else:
-            if s.get('downloaded_bytes') == 0:
-                s['_percent_str'] = self.format_percent(0)
-            else:
-                s['_percent_str'] = 'Unknown %'
+        msg_template = with_fields(
+            ('total_bytes', '%(_percent_str)s of %(_total_bytes_str)s at %(_speed_str)s ETA %(_eta_str)s'),
+            ('total_bytes_estimate', '%(_percent_str)s of ~%(_total_bytes_estimate_str)s at %(_speed_str)s ETA %(_eta_str)s'),
+            ('downloaded_bytes', 'elapsed', '%(_downloaded_bytes_str)s at %(_speed_str)s (%(_elapsed_str)s)'),
+            ('downloaded_bytes', '%(_downloaded_bytes_str)s at %(_speed_str)s'),
+            default='%(_percent_str)s at %(_speed_str)s ETA %(_eta_str)s')
 
-        if s.get('speed') is not None:
-            s['_speed_str'] = self.format_speed(s['speed'])
-        else:
-            s['_speed_str'] = 'Unknown speed'
-
-        if s.get('total_bytes') is not None:
-            s['_total_bytes_str'] = format_bytes(s['total_bytes'])
-            msg_template = '%(_percent_str)s of %(_total_bytes_str)s at %(_speed_str)s ETA %(_eta_str)s'
-        elif s.get('total_bytes_estimate') is not None:
-            s['_total_bytes_estimate_str'] = format_bytes(s['total_bytes_estimate'])
-            msg_template = '%(_percent_str)s of ~%(_total_bytes_estimate_str)s at %(_speed_str)s ETA %(_eta_str)s'
-        else:
-            if s.get('downloaded_bytes') is not None:
-                s['_downloaded_bytes_str'] = format_bytes(s['downloaded_bytes'])
-                if s.get('elapsed'):
-                    s['_elapsed_str'] = self.format_seconds(s['elapsed'])
-                    msg_template = '%(_downloaded_bytes_str)s at %(_speed_str)s (%(_elapsed_str)s)'
-                else:
-                    msg_template = '%(_downloaded_bytes_str)s at %(_speed_str)s'
-            else:
-                msg_template = '%(_percent_str)s at %(_speed_str)s ETA %(_eta_str)s'
-        if s.get('fragment_index') and s.get('fragment_count'):
-            msg_template += ' (frag %(fragment_index)s/%(fragment_count)s)'
-        elif s.get('fragment_index'):
-            msg_template += ' (frag %(fragment_index)s)'
+        msg_template += with_fields(
+            ('fragment_index', 'fragment_count', ' (frag %(fragment_index)s/%(fragment_count)s)'),
+            ('fragment_index', ' (frag %(fragment_index)s)'))
         self._report_progress_status(s, msg_template)
 
     def report_resuming_byte(self, resume_len):
@@ -390,14 +380,23 @@ class FileDownloader:
 
     def report_retry(self, err, count, retries):
         """Report retry in case of HTTP error 5xx"""
-        self.to_screen(
+        self.__to_screen(
             '[download] Got server HTTP error: %s. Retrying (attempt %d of %s) ...'
             % (error_to_compat_str(err), count, self.format_retries(retries)))
+        self.sleep_retry('http', count)
 
     def report_unable_to_resume(self):
         """Report it was impossible to resume download."""
         self.to_screen('[download] Unable to resume')
 
+    def sleep_retry(self, retry_type, count):
+        sleep_func = self.params.get('retry_sleep_functions', {}).get(retry_type)
+        delay = float_or_none(sleep_func(n=count - 1)) if sleep_func else None
+        if delay:
+            self.__to_screen(f'Sleeping {delay:.2f} seconds ...')
+            time.sleep(delay)
+        return sleep_func is not None
+
     @staticmethod
     def supports_manifest(manifest):
         """ Whether the downloader can download the fragments from the manifest.

yt_dlp/downloader/dash.py

@@ -1,7 +1,7 @@
 import time
 
+from . import get_suitable_downloader
 from .fragment import FragmentFD
-from ..downloader import get_suitable_downloader
 from ..utils import urljoin
 
 
@@ -73,6 +73,7 @@ class DashSegmentsFD(FragmentFD):
 
             yield {
                 'frag_index': frag_index,
+                'fragment_count': fragment.get('fragment_count'),
                 'index': i,
                 'url': fragment_url,
             }

yt_dlp/downloader/external.py

@@ -1,3 +1,4 @@
+import enum
 import os.path
 import re
 import subprocess
@@ -5,7 +6,7 @@ import sys
 import time
 
 from .fragment import FragmentFD
-from ..compat import compat_setenv, compat_str
+from ..compat import functools
 from ..postprocessor.ffmpeg import EXT_TO_OUT_FORMATS, FFmpegPostProcessor
 from ..utils import (
     Popen,
@@ -24,9 +25,15 @@ from ..utils import (
 )
 
 
+class Features(enum.Enum):
+    TO_STDOUT = enum.auto()
+    MULTIPLE_FORMATS = enum.auto()
+
+
 class ExternalFD(FragmentFD):
     SUPPORTED_PROTOCOLS = ('http', 'https', 'ftp', 'ftps')
-    can_download_to_stdout = False
+    SUPPORTED_FEATURES = ()
+    _CAPTURE_STDERR = True
 
     def real_download(self, filename, info_dict):
         self.report_destination(filename)
@@ -74,7 +81,7 @@ class ExternalFD(FragmentFD):
     def EXE_NAME(cls):
         return cls.get_basename()
 
-    @property
+    @functools.cached_property
     def exe(self):
         return self.EXE_NAME
 
@@ -90,9 +97,11 @@ class ExternalFD(FragmentFD):
 
     @classmethod
     def supports(cls, info_dict):
-        return (
-            (cls.can_download_to_stdout or not info_dict.get('to_stdout'))
-            and info_dict['protocol'] in cls.SUPPORTED_PROTOCOLS)
+        return all((
+            not info_dict.get('to_stdout') or Features.TO_STDOUT in cls.SUPPORTED_FEATURES,
+            '+' not in info_dict['protocol'] or Features.MULTIPLE_FORMATS in cls.SUPPORTED_FEATURES,
+            all(proto in cls.SUPPORTED_PROTOCOLS for proto in info_dict['protocol'].split('+')),
+        ))
 
     @classmethod
     def can_download(cls, info_dict, path=None):
@@ -119,29 +128,31 @@ class ExternalFD(FragmentFD):
         self._debug_cmd(cmd)
 
         if 'fragments' not in info_dict:
-            p = Popen(cmd, stderr=subprocess.PIPE)
-            _, stderr = p.communicate_or_kill()
-            if p.returncode != 0:
-                self.to_stderr(stderr.decode('utf-8', 'replace'))
-            return p.returncode
+            _, stderr, returncode = Popen.run(
+                cmd, text=True, stderr=subprocess.PIPE if self._CAPTURE_STDERR else None)
+            if returncode and stderr:
+                self.to_stderr(stderr)
+            return returncode
 
         fragment_retries = self.params.get('fragment_retries', 0)
         skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
 
         count = 0
         while count <= fragment_retries:
-            p = Popen(cmd, stderr=subprocess.PIPE)
-            _, stderr = p.communicate_or_kill()
-            if p.returncode == 0:
+            _, stderr, returncode = Popen.run(cmd, text=True, stderr=subprocess.PIPE)
+            if not returncode:
                 break
+
             # TODO: Decide whether to retry based on error code
             # https://aria2.github.io/manual/en/html/aria2c.html#exit-status
-            self.to_stderr(stderr.decode('utf-8', 'replace'))
+            if stderr:
+                self.to_stderr(stderr)
             count += 1
             if count <= fragment_retries:
                 self.to_screen(
                     '[%s] Got error. Retrying fragments (attempt %d of %s)...'
                     % (self.get_basename(), count, self.format_retries(fragment_retries)))
+                self.sleep_retry('fragment', count)
         if count > fragment_retries:
             if not skip_unavailable_fragments:
                 self.report_error('Giving up after %s fragment retries' % fragment_retries)
@@ -170,6 +181,7 @@ class ExternalFD(FragmentFD):
 
 class CurlFD(ExternalFD):
     AVAILABLE_OPT = '-V'
+    _CAPTURE_STDERR = False  # curl writes the progress to stderr
 
     def _make_cmd(self, tmpfilename, info_dict):
         cmd = [self.exe, '--location', '-o', tmpfilename, '--compressed']
@@ -194,16 +206,6 @@ class CurlFD(ExternalFD):
         cmd += ['--', info_dict['url']]
         return cmd
 
-    def _call_downloader(self, tmpfilename, info_dict):
-        cmd = [encodeArgument(a) for a in self._make_cmd(tmpfilename, info_dict)]
-
-        self._debug_cmd(cmd)
-
-        # curl writes the progress to stderr so don't capture it.
-        p = Popen(cmd)
-        p.communicate_or_kill()
-        return p.returncode
-
 
 class AxelFD(ExternalFD):
     AVAILABLE_OPT = '-V'
@@ -322,7 +324,7 @@ class HttpieFD(ExternalFD):
 
 class FFmpegFD(ExternalFD):
     SUPPORTED_PROTOCOLS = ('http', 'https', 'ftp', 'ftps', 'm3u8', 'm3u8_native', 'rtsp', 'rtmp', 'rtmp_ffmpeg', 'mms', 'http_dash_segments')
-    can_download_to_stdout = True
+    SUPPORTED_FEATURES = (Features.TO_STDOUT, Features.MULTIPLE_FORMATS)
 
     @classmethod
     def available(cls, path=None):
@@ -330,10 +332,6 @@ class FFmpegFD(ExternalFD):
         # Fixme: This may be wrong when --ffmpeg-location is used
         return FFmpegPostProcessor().available
 
-    @classmethod
-    def supports(cls, info_dict):
-        return all(proto in cls.SUPPORTED_PROTOCOLS for proto in info_dict['protocol'].split('+'))
-
     def on_process_started(self, proc, stdin):
         """ Override this in subclasses  """
         pass
@@ -378,13 +376,6 @@ class FFmpegFD(ExternalFD):
             # http://trac.ffmpeg.org/ticket/6125#comment:10
             args += ['-seekable', '1' if seekable else '0']
 
-        # start_time = info_dict.get('start_time') or 0
-        # if start_time:
-        #     args += ['-ss', compat_str(start_time)]
-        # end_time = info_dict.get('end_time')
-        # if end_time:
-        #     args += ['-t', compat_str(end_time - start_time)]
-
         http_headers = None
         if info_dict.get('http_headers'):
             youtubedl_headers = handle_youtubedl_headers(info_dict['http_headers'])
@@ -411,8 +402,8 @@ class FFmpegFD(ExternalFD):
             # We could switch to the following code if we are able to detect version properly
             # args += ['-http_proxy', proxy]
             env = os.environ.copy()
-            compat_setenv('HTTP_PROXY', proxy, env=env)
-            compat_setenv('http_proxy', proxy, env=env)
+            env['HTTP_PROXY'] = proxy
+            env['http_proxy'] = proxy
 
         protocol = info_dict.get('protocol')
 
@@ -442,25 +433,31 @@ class FFmpegFD(ExternalFD):
             if isinstance(conn, list):
                 for entry in conn:
                     args += ['-rtmp_conn', entry]
-            elif isinstance(conn, compat_str):
+            elif isinstance(conn, str):
                 args += ['-rtmp_conn', conn]
 
+        start_time, end_time = info_dict.get('section_start') or 0, info_dict.get('section_end')
+
         for i, url in enumerate(urls):
-            # We need to specify headers for each http input stream
-            # otherwise, it will only be applied to the first.
-            # https://github.com/yt-dlp/yt-dlp/issues/2696
             if http_headers is not None and re.match(r'^https?://', url):
                 args += http_headers
+            if start_time:
+                args += ['-ss', str(start_time)]
+            if end_time:
+                args += ['-t', str(end_time - start_time)]
+
             args += self._configuration_args((f'_i{i + 1}', '_i')) + ['-i', url]
 
-        args += ['-c', 'copy']
+        if not (start_time or end_time) or not self.params.get('force_keyframes_at_cuts'):
+            args += ['-c', 'copy']
+
         if info_dict.get('requested_formats') or protocol == 'http_dash_segments':
             for (i, fmt) in enumerate(info_dict.get('requested_formats') or [info_dict]):
                 stream_number = fmt.get('manifest_stream_number', 0)
                 args.extend(['-map', f'{i}:{stream_number}'])
 
         if self.params.get('test', False):
-            args += ['-fs', compat_str(self._TEST_FILE_SIZE)]
+            args += ['-fs', str(self._TEST_FILE_SIZE)]
 
         ext = info_dict['ext']
         if protocol in ('m3u8', 'm3u8_native'):
@@ -495,24 +492,23 @@ class FFmpegFD(ExternalFD):
         args.append(encodeFilename(ffpp._ffmpeg_filename_argument(tmpfilename), True))
         self._debug_cmd(args)
 
-        proc = Popen(args, stdin=subprocess.PIPE, env=env)
-        if url in ('-', 'pipe:'):
-            self.on_process_started(proc, proc.stdin)
-        try:
-            retval = proc.wait()
-        except BaseException as e:
-            # subprocces.run would send the SIGKILL signal to ffmpeg and the
-            # mp4 file couldn't be played, but if we ask ffmpeg to quit it
-            # produces a file that is playable (this is mostly useful for live
-            # streams). Note that Windows is not affected and produces playable
-            # files (see https://github.com/ytdl-org/youtube-dl/issues/8300).
-            if isinstance(e, KeyboardInterrupt) and sys.platform != 'win32' and url not in ('-', 'pipe:'):
-                proc.communicate_or_kill(b'q')
-            else:
-                proc.kill()
-                proc.wait()
-            raise
-        return retval
+        with Popen(args, stdin=subprocess.PIPE, env=env) as proc:
+            if url in ('-', 'pipe:'):
+                self.on_process_started(proc, proc.stdin)
+            try:
+                retval = proc.wait()
+            except BaseException as e:
+                # subprocces.run would send the SIGKILL signal to ffmpeg and the
+                # mp4 file couldn't be played, but if we ask ffmpeg to quit it
+                # produces a file that is playable (this is mostly useful for live
+                # streams). Note that Windows is not affected and produces playable
+                # files (see https://github.com/ytdl-org/youtube-dl/issues/8300).
+                if isinstance(e, KeyboardInterrupt) and sys.platform != 'win32' and url not in ('-', 'pipe:'):
+                    proc.communicate_or_kill(b'q')
+                else:
+                    proc.kill(timeout=None)
+                raise
+            return retval
 
 
 class AVconvFD(FFmpegFD):

yt_dlp/downloader/f4m.py

@@ -1,17 +1,13 @@
+import base64
 import io
 import itertools
+import struct
 import time
+import urllib.error
+import urllib.parse
 
 from .fragment import FragmentFD
-from ..compat import (
-    compat_b64decode,
-    compat_etree_fromstring,
-    compat_struct_pack,
-    compat_struct_unpack,
-    compat_urllib_error,
-    compat_urllib_parse_urlparse,
-    compat_urlparse,
-)
+from ..compat import compat_etree_fromstring
 from ..utils import fix_xml_ampersands, xpath_text
 
 
@@ -35,13 +31,13 @@ class FlvReader(io.BytesIO):
 
     # Utility functions for reading numbers and strings
     def read_unsigned_long_long(self):
-        return compat_struct_unpack('!Q', self.read_bytes(8))[0]
+        return struct.unpack('!Q', self.read_bytes(8))[0]
 
     def read_unsigned_int(self):
-        return compat_struct_unpack('!I', self.read_bytes(4))[0]
+        return struct.unpack('!I', self.read_bytes(4))[0]
 
     def read_unsigned_char(self):
-        return compat_struct_unpack('!B', self.read_bytes(1))[0]
+        return struct.unpack('!B', self.read_bytes(1))[0]
 
     def read_string(self):
         res = b''
@@ -203,11 +199,11 @@ def build_fragments_list(boot_info):
 
 
 def write_unsigned_int(stream, val):
-    stream.write(compat_struct_pack('!I', val))
+    stream.write(struct.pack('!I', val))
 
 
 def write_unsigned_int_24(stream, val):
-    stream.write(compat_struct_pack('!I', val)[1:])
+    stream.write(struct.pack('!I', val)[1:])
 
 
 def write_flv_header(stream):
@@ -301,12 +297,12 @@ class F4mFD(FragmentFD):
         # 1. http://live-1-1.rutube.ru/stream/1024/HDS/SD/C2NKsS85HQNckgn5HdEmOQ/1454167650/S-s604419906/move/four/dirs/upper/1024-576p.f4m
         bootstrap_url = node.get('url')
         if bootstrap_url:
-            bootstrap_url = compat_urlparse.urljoin(
+            bootstrap_url = urllib.parse.urljoin(
                 base_url, bootstrap_url)
             boot_info = self._get_bootstrap_from_url(bootstrap_url)
         else:
             bootstrap_url = None
-            bootstrap = compat_b64decode(node.text)
+            bootstrap = base64.b64decode(node.text)
             boot_info = read_bootstrap_info(bootstrap)
         return boot_info, bootstrap_url
 
@@ -336,14 +332,14 @@ class F4mFD(FragmentFD):
         # Prefer baseURL for relative URLs as per 11.2 of F4M 3.0 spec.
         man_base_url = get_base_url(doc) or man_url
 
-        base_url = compat_urlparse.urljoin(man_base_url, media.attrib['url'])
+        base_url = urllib.parse.urljoin(man_base_url, media.attrib['url'])
         bootstrap_node = doc.find(_add_ns('bootstrapInfo'))
         boot_info, bootstrap_url = self._parse_bootstrap_node(
             bootstrap_node, man_base_url)
         live = boot_info['live']
         metadata_node = media.find(_add_ns('metadata'))
         if metadata_node is not None:
-            metadata = compat_b64decode(metadata_node.text)
+            metadata = base64.b64decode(metadata_node.text)
         else:
             metadata = None
 
@@ -371,7 +367,7 @@ class F4mFD(FragmentFD):
             if not live:
                 write_metadata_tag(dest_stream, metadata)
 
-        base_url_parsed = compat_urllib_parse_urlparse(base_url)
+        base_url_parsed = urllib.parse.urlparse(base_url)
 
         self._start_frag_download(ctx, info_dict)
 
@@ -391,9 +387,10 @@ class F4mFD(FragmentFD):
                 query.append(info_dict['extra_param_to_segment_url'])
             url_parsed = base_url_parsed._replace(path=base_url_parsed.path + name, query='&'.join(query))
             try:
-                success, down_data = self._download_fragment(ctx, url_parsed.geturl(), info_dict)
+                success = self._download_fragment(ctx, url_parsed.geturl(), info_dict)
                 if not success:
                     return False
+                down_data = self._read_fragment(ctx)
                 reader = FlvReader(down_data)
                 while True:
                     try:
@@ -410,7 +407,7 @@ class F4mFD(FragmentFD):
                     if box_type == b'mdat':
                         self._append_fragment(ctx, box_data)
                         break
-            except compat_urllib_error.HTTPError as err:
+            except urllib.error.HTTPError as err:
                 if live and (err.code == 404 or err.code == 410):
                     # We didn't keep up with the live window. Continue
                     # with the next available fragment.

yt_dlp/downloader/fragment.py

@@ -4,12 +4,14 @@ import http.client
 import json
 import math
 import os
+import struct
 import time
+import urllib.error
 
 from .common import FileDownloader
 from .http import HttpFD
 from ..aes import aes_cbc_decrypt_bytes, unpad_pkcs7
-from ..compat import compat_os_name, compat_struct_pack, compat_urllib_error
+from ..compat import compat_os_name
 from ..utils import (
     DownloadError,
     encodeFilename,
@@ -23,11 +25,7 @@ class HttpQuietDownloader(HttpFD):
     def to_screen(self, *args, **kargs):
         pass
 
-    console_title = to_screen
-
-    def report_retry(self, err, count, retries):
-        super().to_screen(
-            f'[download] Got server HTTP error: {err}. Retrying (attempt {count} of {self.format_retries(retries)}) ...')
+    to_console_title = to_screen
 
 
 class FragmentFD(FileDownloader):
@@ -70,6 +68,7 @@ class FragmentFD(FileDownloader):
         self.to_screen(
             '\r[download] Got server HTTP error: %s. Retrying fragment %d (attempt %d of %s) ...'
             % (error_to_compat_str(err), frag_index, count, self.format_retries(retries)))
+        self.sleep_retry('fragment', count)
 
     def report_skip_fragment(self, frag_index, err=None):
         err = f' {err};' if err else ''
@@ -168,18 +167,11 @@ class FragmentFD(FileDownloader):
             total_frags_str = 'unknown (live)'
         self.to_screen(f'[{self.FD_NAME}] Total fragments: {total_frags_str}')
         self.report_destination(ctx['filename'])
-        dl = HttpQuietDownloader(
-            self.ydl,
-            {
-                'continuedl': self.params.get('continuedl', True),
-                'quiet': self.params.get('quiet'),
-                'noprogress': True,
-                'ratelimit': self.params.get('ratelimit'),
-                'retries': self.params.get('retries', 0),
-                'nopart': self.params.get('nopart', False),
-                'test': False,
-            }
-        )
+        dl = HttpQuietDownloader(self.ydl, {
+            **self.params,
+            'noprogress': True,
+            'test': False,
+        })
         tmpfilename = self.temp_name(ctx['filename'])
         open_mode = 'wb'
         resume_len = 0
@@ -252,6 +244,9 @@ class FragmentFD(FileDownloader):
             if s['status'] not in ('downloading', 'finished'):
                 return
 
+            if not total_frags and ctx.get('fragment_count'):
+                state['fragment_count'] = ctx['fragment_count']
+
             if ctx_id is not None and s.get('ctx_id') != ctx_id:
                 return
 
@@ -355,7 +350,7 @@ class FragmentFD(FileDownloader):
             decrypt_info = fragment.get('decrypt_info')
             if not decrypt_info or decrypt_info['METHOD'] != 'AES-128':
                 return frag_content
-            iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', fragment['media_sequence'])
+            iv = decrypt_info.get('IV') or struct.pack('>8xq', fragment['media_sequence'])
             decrypt_info['KEY'] = decrypt_info.get('KEY') or _get_key(info_dict.get('_decryption_key_url') or decrypt_info['URI'])
             # Don't decrypt the content in tests since the data is explicitly truncated and it's not to a valid block
             # size (see https://github.com/ytdl-org/youtube-dl/pull/27660). Tests only care that the correct data downloaded,
@@ -460,10 +455,11 @@ class FragmentFD(FileDownloader):
             fatal, count = is_fatal(fragment.get('index') or (frag_index - 1)), 0
             while count <= fragment_retries:
                 try:
+                    ctx['fragment_count'] = fragment.get('fragment_count')
                     if self._download_fragment(ctx, fragment['url'], info_dict, headers):
                         break
                     return
-                except (compat_urllib_error.HTTPError, http.client.IncompleteRead) as err:
+                except (urllib.error.HTTPError, http.client.IncompleteRead) as err:
                     # Unavailable (possibly temporary) fragments may be served.
                     # First we try to retry then either skip or abort.
                     # See https://github.com/ytdl-org/youtube-dl/issues/10165,
@@ -506,12 +502,20 @@ class FragmentFD(FileDownloader):
 
             self.report_warning('The download speed shown is only of one thread. This is a known issue and patches are welcome')
             with tpe or concurrent.futures.ThreadPoolExecutor(max_workers) as pool:
-                for fragment, frag_index, frag_filename in pool.map(_download_fragment, fragments):
-                    ctx['fragment_filename_sanitized'] = frag_filename
-                    ctx['fragment_index'] = frag_index
-                    result = append_fragment(decrypt_fragment(fragment, self._read_fragment(ctx)), frag_index, ctx)
-                    if not result:
-                        return False
+                try:
+                    for fragment, frag_index, frag_filename in pool.map(_download_fragment, fragments):
+                        ctx.update({
+                            'fragment_filename_sanitized': frag_filename,
+                            'fragment_index': frag_index,
+                        })
+                        if not append_fragment(decrypt_fragment(fragment, self._read_fragment(ctx)), frag_index, ctx):
+                            return False
+                except KeyboardInterrupt:
+                    self._finish_multiline_status()
+                    self.report_error(
+                        'Interrupted by user. Waiting for all threads to shutdown...', is_error=False, tb=False)
+                    pool.shutdown(wait=False)
+                    raise
         else:
             for fragment in fragments:
                 if not interrupt_trigger[0]:

yt_dlp/downloader/hls.py

@@ -1,13 +1,13 @@
 import binascii
 import io
 import re
+import urllib.parse
 
+from . import get_suitable_downloader
 from .external import FFmpegFD
 from .fragment import FragmentFD
 from .. import webvtt
-from ..compat import compat_urlparse
 from ..dependencies import Cryptodome_AES
-from ..downloader import get_suitable_downloader
 from ..utils import bug_reports_message, parse_m3u8_attributes, update_url_query
 
 
@@ -61,12 +61,18 @@ class HlsFD(FragmentFD):
         s = urlh.read().decode('utf-8', 'ignore')
 
         can_download, message = self.can_download(s, info_dict, self.params.get('allow_unplayable_formats')), None
-        if can_download and not Cryptodome_AES and '#EXT-X-KEY:METHOD=AES-128' in s:
-            if FFmpegFD.available():
+        if can_download:
+            has_ffmpeg = FFmpegFD.available()
+            no_crypto = not Cryptodome_AES and '#EXT-X-KEY:METHOD=AES-128' in s
+            if no_crypto and has_ffmpeg:
                 can_download, message = False, 'The stream has AES-128 encryption and pycryptodomex is not available'
-            else:
+            elif no_crypto:
                 message = ('The stream has AES-128 encryption and neither ffmpeg nor pycryptodomex are available; '
                            'Decryption will be performed natively, but will be extremely slow')
+            elif info_dict.get('extractor_key') == 'Generic' and re.search(r'(?m)#EXT-X-MEDIA-SEQUENCE:(?!0$)', s):
+                install_ffmpeg = '' if has_ffmpeg else 'install ffmpeg and '
+                message = ('Live HLS streams are not supported by the native downloader. If this is a livestream, '
+                           f'please {install_ffmpeg}add "--downloader ffmpeg --hls-use-mpegts" to your command')
         if not can_download:
             has_drm = re.search('|'.join([
                 r'#EXT-X-FAXS-CM:',  # Adobe Flash Access
@@ -140,7 +146,7 @@ class HlsFD(FragmentFD):
         extra_query = None
         extra_param_to_segment_url = info_dict.get('extra_param_to_segment_url')
         if extra_param_to_segment_url:
-            extra_query = compat_urlparse.parse_qs(extra_param_to_segment_url)
+            extra_query = urllib.parse.parse_qs(extra_param_to_segment_url)
         i = 0
         media_sequence = 0
         decrypt_info = {'METHOD': 'NONE'}
@@ -162,7 +168,7 @@ class HlsFD(FragmentFD):
                     frag_url = (
                         line
                         if re.match(r'^https?://', line)
-                        else compat_urlparse.urljoin(man_url, line))
+                        else urllib.parse.urljoin(man_url, line))
                     if extra_query:
                         frag_url = update_url_query(frag_url, extra_query)
 
@@ -187,7 +193,7 @@ class HlsFD(FragmentFD):
                     frag_url = (
                         map_info.get('URI')
                         if re.match(r'^https?://', map_info.get('URI'))
-                        else compat_urlparse.urljoin(man_url, map_info.get('URI')))
+                        else urllib.parse.urljoin(man_url, map_info.get('URI')))
                     if extra_query:
                         frag_url = update_url_query(frag_url, extra_query)
 
@@ -215,7 +221,7 @@ class HlsFD(FragmentFD):
                         if 'IV' in decrypt_info:
                             decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:].zfill(32))
                         if not re.match(r'^https?://', decrypt_info['URI']):
-                            decrypt_info['URI'] = compat_urlparse.urljoin(
+                            decrypt_info['URI'] = urllib.parse.urljoin(
                                 man_url, decrypt_info['URI'])
                         if extra_query:
                             decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_query)

yt_dlp/downloader/http.py

@@ -1,11 +1,12 @@
+import http.client
 import os
 import random
 import socket
 import ssl
 import time
+import urllib.error
 
 from .common import FileDownloader
-from ..compat import compat_http_client, compat_urllib_error
 from ..utils import (
     ContentTooShortError,
     ThrottledDownload,
@@ -24,7 +25,7 @@ RESPONSE_READ_EXCEPTIONS = (
     socket.timeout,  # compat: py < 3.10
     ConnectionError,
     ssl.SSLError,
-    compat_http_client.HTTPException
+    http.client.HTTPException
 )
 
 
@@ -136,20 +137,18 @@ class HttpFD(FileDownloader):
                 if has_range:
                     content_range = ctx.data.headers.get('Content-Range')
                     content_range_start, content_range_end, content_len = parse_http_range(content_range)
-                    if content_range_start is not None and range_start == content_range_start:
-                        # Content-Range is present and matches requested Range, resume is possible
-                        accept_content_len = (
+                    # Content-Range is present and matches requested Range, resume is possible
+                    if range_start == content_range_start and (
                             # Non-chunked download
                             not ctx.chunk_size
                             # Chunked download and requested piece or
                             # its part is promised to be served
                             or content_range_end == range_end
-                            or content_len < range_end)
-                        if accept_content_len:
-                            ctx.content_len = content_len
-                            if content_len or req_end:
-                                ctx.data_len = min(content_len or req_end, req_end or content_len) - (req_start or 0)
-                            return
+                            or content_len < range_end):
+                        ctx.content_len = content_len
+                        if content_len or req_end:
+                            ctx.data_len = min(content_len or req_end, req_end or content_len) - (req_start or 0)
+                        return
                     # Content-Range is either not present or invalid. Assuming remote webserver is
                     # trying to send the whole file, resume is not possible, so wiping the local file
                     # and performing entire redownload
@@ -157,7 +156,7 @@ class HttpFD(FileDownloader):
                     ctx.resume_len = 0
                     ctx.open_mode = 'wb'
                 ctx.data_len = ctx.content_len = int_or_none(ctx.data.info().get('Content-length', None))
-            except compat_urllib_error.HTTPError as err:
+            except urllib.error.HTTPError as err:
                 if err.code == 416:
                     # Unable to resume (requested range not satisfiable)
                     try:
@@ -165,7 +164,7 @@ class HttpFD(FileDownloader):
                         ctx.data = self.ydl.urlopen(
                             sanitized_Request(url, request_data, headers))
                         content_length = ctx.data.info()['Content-Length']
-                    except compat_urllib_error.HTTPError as err:
+                    except urllib.error.HTTPError as err:
                         if err.code < 500 or err.code >= 600:
                             raise
                     else:
@@ -198,7 +197,7 @@ class HttpFD(FileDownloader):
                     # Unexpected HTTP error
                     raise
                 raise RetryDownload(err)
-            except compat_urllib_error.URLError as err:
+            except urllib.error.URLError as err:
                 if isinstance(err.reason, ssl.CertificateError):
                     raise
                 raise RetryDownload(err)

yt_dlp/downloader/ism.py

@@ -2,9 +2,9 @@ import binascii
 import io
 import struct
 import time
+import urllib.error
 
 from .fragment import FragmentFD
-from ..compat import compat_urllib_error
 
 u8 = struct.Struct('>B')
 u88 = struct.Struct('>Bx')
@@ -268,7 +268,7 @@ class IsmFD(FragmentFD):
                         extra_state['ism_track_written'] = True
                     self._append_fragment(ctx, frag_content)
                     break
-                except compat_urllib_error.HTTPError as err:
+                except urllib.error.HTTPError as err:
                     count += 1
                     if count <= fragment_retries:
                         self.report_retry_fragment(err, frag_index, count, fragment_retries)

yt_dlp/downloader/niconico.py

@@ -1,8 +1,7 @@
 import threading
 
+from . import get_suitable_downloader
 from .common import FileDownloader
-from ..downloader import get_suitable_downloader
-from ..extractor.niconico import NiconicoIE
 from ..utils import sanitized_Request
 
 
@@ -10,8 +9,9 @@ class NiconicoDmcFD(FileDownloader):
     """ Downloading niconico douga from DMC with heartbeat """
 
     def real_download(self, filename, info_dict):
-        self.to_screen('[%s] Downloading from DMC' % self.FD_NAME)
+        from ..extractor.niconico import NiconicoIE
 
+        self.to_screen('[%s] Downloading from DMC' % self.FD_NAME)
         ie = NiconicoIE(self.ydl)
         info_dict, heartbeat_info_dict = ie._get_heartbeat_info(info_dict)
 

yt_dlp/downloader/rtmp.py

@@ -4,7 +4,6 @@ import subprocess
 import time
 
 from .common import FileDownloader
-from ..compat import compat_str
 from ..utils import (
     Popen,
     check_executable,
@@ -92,8 +91,7 @@ class RtmpFD(FileDownloader):
                     self.to_screen('')
                 return proc.wait()
             except BaseException:  # Including KeyboardInterrupt
-                proc.kill()
-                proc.wait()
+                proc.kill(timeout=None)
                 raise
 
         url = info_dict['url']
@@ -144,7 +142,7 @@ class RtmpFD(FileDownloader):
         if isinstance(conn, list):
             for entry in conn:
                 basic_args += ['--conn', entry]
-        elif isinstance(conn, compat_str):
+        elif isinstance(conn, str):
             basic_args += ['--conn', conn]
         if protocol is not None:
             basic_args += ['--protocol', protocol]

yt_dlp/downloader/youtube_live_chat.py

@@ -1,9 +1,8 @@
 import json
 import time
+import urllib.error
 
 from .fragment import FragmentFD
-from ..compat import compat_urllib_error
-from ..extractor.youtube import YoutubeBaseInfoExtractor as YT_BaseIE
 from ..utils import RegexNotFoundError, dict_get, int_or_none, try_get
 
 
@@ -26,7 +25,9 @@ class YoutubeLiveChatFD(FragmentFD):
             'total_frags': None,
         }
 
-        ie = YT_BaseIE(self.ydl)
+        from ..extractor.youtube import YoutubeBaseInfoExtractor
+
+        ie = YoutubeBaseInfoExtractor(self.ydl)
 
         start_time = int(time.time() * 1000)
 
@@ -127,7 +128,7 @@ class YoutubeLiveChatFD(FragmentFD):
                     elif info_dict['protocol'] == 'youtube_live_chat':
                         continuation_id, offset, click_tracking_params = parse_actions_live(live_chat_continuation)
                     return True, continuation_id, offset, click_tracking_params
-                except compat_urllib_error.HTTPError as err:
+                except urllib.error.HTTPError as err:
                     count += 1
                     if count <= fragment_retries:
                         self.report_retry_fragment(err, frag_index, count, fragment_retries)

yt_dlp/extractor/__init__.py

@@ -1,32 +1,15 @@
-import contextlib
-import os
+from ..compat.compat_utils import passthrough_module
 
-from ..utils import load_plugins
-
-_LAZY_LOADER = False
-if not os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
-    with contextlib.suppress(ImportError):
-        from .lazy_extractors import *  # noqa: F403
-        from .lazy_extractors import _ALL_CLASSES
-        _LAZY_LOADER = True
-
-if not _LAZY_LOADER:
-    from .extractors import *  # noqa: F403
-    _ALL_CLASSES = [  # noqa: F811
-        klass
-        for name, klass in globals().items()
-        if name.endswith('IE') and name != 'GenericIE'
-    ]
-    _ALL_CLASSES.append(GenericIE)  # noqa: F405
-
-_PLUGIN_CLASSES = load_plugins('extractor', 'IE', globals())
-_ALL_CLASSES = list(_PLUGIN_CLASSES.values()) + _ALL_CLASSES
+passthrough_module(__name__, '.extractors')
+del passthrough_module
 
 
 def gen_extractor_classes():
     """ Return a list of supported extractors.
     The order does matter; the first extractor matched is the one handling the URL.
     """
+    from .extractors import _ALL_CLASSES
+
     return _ALL_CLASSES
 
 
@@ -39,10 +22,12 @@ def gen_extractors():
 
 def list_extractor_classes(age_limit=None):
     """Return a list of extractors that are suitable for the given age, sorted by extractor name"""
+    from .generic import GenericIE
+
     yield from sorted(filter(
-        lambda ie: ie.is_suitable(age_limit) and ie != GenericIE,  # noqa: F405
+        lambda ie: ie.is_suitable(age_limit) and ie != GenericIE,
         gen_extractor_classes()), key=lambda ie: ie.IE_NAME.lower())
-    yield GenericIE  # noqa: F405
+    yield GenericIE
 
 
 def list_extractors(age_limit=None):
@@ -52,4 +37,6 @@ def list_extractors(age_limit=None):
 
 def get_info_extractor(ie_name):
     """Returns the info extractor class with the given ie_name"""
-    return globals()[ie_name + 'IE']
+    from . import extractors
+
+    return getattr(extractors, f'{ie_name}IE')

yt_dlp/extractor/abematv.py

@@ -7,16 +7,17 @@ import json
 import re
 import struct
 import time
+import urllib.parse
+import urllib.request
 import urllib.response
 import uuid
 
 from .common import InfoExtractor
 from ..aes import aes_ecb_decrypt
-from ..compat import compat_urllib_parse_urlparse, compat_urllib_request
 from ..utils import (
     ExtractorError,
     bytes_to_intlist,
-    decode_base,
+    decode_base_n,
     int_or_none,
     intlist_to_bytes,
     request_to_url,
@@ -33,7 +34,7 @@ def add_opener(ydl, handler):
     ''' Add a handler for opening URLs, like _download_webpage '''
     # https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L426
     # https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L605
-    assert isinstance(ydl._opener, compat_urllib_request.OpenerDirector)
+    assert isinstance(ydl._opener, urllib.request.OpenerDirector)
     ydl._opener.add_handler(handler)
 
 
@@ -46,7 +47,7 @@ def remove_opener(ydl, handler):
     # https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L426
     # https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L605
     opener = ydl._opener
-    assert isinstance(ydl._opener, compat_urllib_request.OpenerDirector)
+    assert isinstance(ydl._opener, urllib.request.OpenerDirector)
     if isinstance(handler, (type, tuple)):
         find_cp = lambda x: isinstance(x, handler)
     else:
@@ -96,20 +97,20 @@ def remove_opener(ydl, handler):
         opener.handlers[:] = [x for x in opener.handlers if not find_cp(x)]
 
 
-class AbemaLicenseHandler(compat_urllib_request.BaseHandler):
+class AbemaLicenseHandler(urllib.request.BaseHandler):
     handler_order = 499
     STRTABLE = '123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz'
     HKEY = b'3AF0298C219469522A313570E8583005A642E73EDD58E3EA2FB7339D3DF1597E'
 
     def __init__(self, ie: 'AbemaTVIE'):
-        # the protcol that this should really handle is 'abematv-license://'
+        # the protocol that this should really handle is 'abematv-license://'
         # abematv_license_open is just a placeholder for development purposes
         # ref. https://github.com/python/cpython/blob/f4c03484da59049eb62a9bf7777b963e2267d187/Lib/urllib/request.py#L510
         setattr(self, 'abematv-license_open', getattr(self, 'abematv_license_open'))
         self.ie = ie
 
     def _get_videokey_from_ticket(self, ticket):
-        to_show = self.ie._downloader.params.get('verbose', False)
+        to_show = self.ie.get_param('verbose', False)
         media_token = self.ie._get_media_token(to_show=to_show)
 
         license_response = self.ie._download_json(
@@ -123,7 +124,7 @@ class AbemaLicenseHandler(compat_urllib_request.BaseHandler):
                 'Content-Type': 'application/json',
             })
 
-        res = decode_base(license_response['k'], self.STRTABLE)
+        res = decode_base_n(license_response['k'], table=self.STRTABLE)
         encvideokey = bytes_to_intlist(struct.pack('>QQ', res >> 64, res & 0xffffffffffffffff))
 
         h = hmac.new(
@@ -136,7 +137,7 @@ class AbemaLicenseHandler(compat_urllib_request.BaseHandler):
 
     def abematv_license_open(self, url):
         url = request_to_url(url)
-        ticket = compat_urllib_parse_urlparse(url).netloc
+        ticket = urllib.parse.urlparse(url).netloc
         response_data = self._get_videokey_from_ticket(ticket)
         return urllib.response.addinfourl(io.BytesIO(response_data), headers={
             'Content-Length': len(response_data),
@@ -311,7 +312,7 @@ class AbemaTVIE(AbemaTVBaseIE):
 
     def _real_extract(self, url):
         # starting download using infojson from this extractor is undefined behavior,
-        # and never be fixed in the future; you must trigger downloads by directly specifing URL.
+        # and never be fixed in the future; you must trigger downloads by directly specifying URL.
         # (unless there's a way to hook before downloading by extractor)
         video_id, video_type = self._match_valid_url(url).group('id', 'type')
         headers = {

yt_dlp/extractor/adobepass.py

@@ -1,3 +1,4 @@
+import getpass
 import json
 import re
 import time
@@ -5,19 +6,15 @@ import urllib.error
 import xml.etree.ElementTree as etree
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urlparse,
-    compat_getpass
-)
+from ..compat import compat_urlparse
 from ..utils import (
-    unescapeHTML,
-    urlencode_postdata,
-    unified_timestamp,
-    ExtractorError,
     NO_DEFAULT,
+    ExtractorError,
+    unescapeHTML,
+    unified_timestamp,
+    urlencode_postdata,
 )
 
-
 MSO_INFO = {
     'DTV': {
         'name': 'DIRECTV',
@@ -1431,7 +1428,7 @@ class AdobePassIE(InfoExtractor):
         guid = xml_text(resource, 'guid') if '<' in resource else resource
         count = 0
         while count < 2:
-            requestor_info = self._downloader.cache.load(self._MVPD_CACHE, requestor_id) or {}
+            requestor_info = self.cache.load(self._MVPD_CACHE, requestor_id) or {}
             authn_token = requestor_info.get('authn_token')
             if authn_token and is_expired(authn_token, 'simpleTokenExpires'):
                 authn_token = None
@@ -1506,7 +1503,7 @@ class AdobePassIE(InfoExtractor):
                             'send_confirm_link': False,
                             'send_token': True
                         }))
-                    philo_code = compat_getpass('Type auth code you have received [Return]: ')
+                    philo_code = getpass.getpass('Type auth code you have received [Return]: ')
                     self._download_webpage(
                         'https://idp.philo.com/auth/update/login_code', video_id, 'Submitting token', data=urlencode_postdata({
                             'token': philo_code
@@ -1726,12 +1723,12 @@ class AdobePassIE(InfoExtractor):
                         raise_mvpd_required()
                     raise
                 if '<pendingLogout' in session:
-                    self._downloader.cache.store(self._MVPD_CACHE, requestor_id, {})
+                    self.cache.store(self._MVPD_CACHE, requestor_id, {})
                     count += 1
                     continue
                 authn_token = unescapeHTML(xml_text(session, 'authnToken'))
                 requestor_info['authn_token'] = authn_token
-                self._downloader.cache.store(self._MVPD_CACHE, requestor_id, requestor_info)
+                self.cache.store(self._MVPD_CACHE, requestor_id, requestor_info)
 
             authz_token = requestor_info.get(guid)
             if authz_token and is_expired(authz_token, 'simpleTokenTTL'):
@@ -1747,14 +1744,14 @@ class AdobePassIE(InfoExtractor):
                         'userMeta': '1',
                     }), headers=mvpd_headers)
                 if '<pendingLogout' in authorize:
-                    self._downloader.cache.store(self._MVPD_CACHE, requestor_id, {})
+                    self.cache.store(self._MVPD_CACHE, requestor_id, {})
                     count += 1
                     continue
                 if '<error' in authorize:
                     raise ExtractorError(xml_text(authorize, 'details'), expected=True)
                 authz_token = unescapeHTML(xml_text(authorize, 'authzToken'))
                 requestor_info[guid] = authz_token
-                self._downloader.cache.store(self._MVPD_CACHE, requestor_id, requestor_info)
+                self.cache.store(self._MVPD_CACHE, requestor_id, requestor_info)
 
             mvpd_headers.update({
                 'ap_19': xml_text(authn_token, 'simpleSamlNameID'),
@@ -1770,7 +1767,7 @@ class AdobePassIE(InfoExtractor):
                     'hashed_guid': 'false',
                 }), headers=mvpd_headers)
             if '<pendingLogout' in short_authorize:
-                self._downloader.cache.store(self._MVPD_CACHE, requestor_id, {})
+                self.cache.store(self._MVPD_CACHE, requestor_id, {})
                 count += 1
                 continue
             return short_authorize

yt_dlp/extractor/animelab.py

@@ -1,270 +0,0 @@
-from .common import InfoExtractor
-
-from ..utils import (
-    ExtractorError,
-    urlencode_postdata,
-    int_or_none,
-    str_or_none,
-    determine_ext,
-)
-
-from ..compat import compat_HTTPError
-
-
-class AnimeLabBaseIE(InfoExtractor):
-    _LOGIN_URL = 'https://www.animelab.com/login'
-    _NETRC_MACHINE = 'animelab'
-    _LOGGED_IN = False
-
-    def _is_logged_in(self, login_page=None):
-        if not self._LOGGED_IN:
-            if not login_page:
-                login_page = self._download_webpage(self._LOGIN_URL, None, 'Downloading login page')
-            AnimeLabBaseIE._LOGGED_IN = 'Sign In' not in login_page
-        return self._LOGGED_IN
-
-    def _perform_login(self, username, password):
-        if self._is_logged_in():
-            return
-
-        login_form = {
-            'email': username,
-            'password': password,
-        }
-
-        try:
-            response = self._download_webpage(
-                self._LOGIN_URL, None, 'Logging in', 'Wrong login info',
-                data=urlencode_postdata(login_form),
-                headers={'Content-Type': 'application/x-www-form-urlencoded'})
-        except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
-                raise ExtractorError('Unable to log in (wrong credentials?)', expected=True)
-            raise
-
-        if not self._is_logged_in(response):
-            raise ExtractorError('Unable to login (cannot verify if logged in)')
-
-    def _real_initialize(self):
-        if not self._is_logged_in():
-            self.raise_login_required('Login is required to access any AnimeLab content')
-
-
-class AnimeLabIE(AnimeLabBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?animelab\.com/player/(?P<id>[^/]+)'
-
-    _TEST = {
-        'url': 'https://www.animelab.com/player/fullmetal-alchemist-brotherhood-episode-42',
-        'md5': '05bde4b91a5d1ff46ef5b94df05b0f7f',
-        'info_dict': {
-            'id': '383',
-            'ext': 'mp4',
-            'display_id': 'fullmetal-alchemist-brotherhood-episode-42',
-            'title': 'Fullmetal Alchemist: Brotherhood - Episode 42 - Signs of a Counteroffensive',
-            'description': 'md5:103eb61dd0a56d3dfc5dbf748e5e83f4',
-            'series': 'Fullmetal Alchemist: Brotherhood',
-            'episode': 'Signs of a Counteroffensive',
-            'episode_number': 42,
-            'duration': 1469,
-            'season': 'Season 1',
-            'season_number': 1,
-            'season_id': '38',
-        },
-        'params': {
-            # Ensure the same video is downloaded whether the user is premium or not
-            'format': '[format_id=21711_yeshardsubbed_ja-JP][height=480]',
-        },
-    }
-
-    def _real_extract(self, url):
-        display_id = self._match_id(url)
-
-        # unfortunately we can get different URLs for the same formats
-        # e.g. if we are using a "free" account so no dubs available
-        # (so _remove_duplicate_formats is not effective)
-        # so we use a dictionary as a workaround
-        formats = {}
-        for language_option_url in ('https://www.animelab.com/player/%s/subtitles',
-                                    'https://www.animelab.com/player/%s/dubbed'):
-            actual_url = language_option_url % display_id
-            webpage = self._download_webpage(actual_url, display_id, 'Downloading URL ' + actual_url)
-
-            video_collection = self._parse_json(self._search_regex(r'new\s+?AnimeLabApp\.VideoCollection\s*?\((.*?)\);', webpage, 'AnimeLab VideoCollection'), display_id)
-            position = int_or_none(self._search_regex(r'playlistPosition\s*?=\s*?(\d+)', webpage, 'Playlist Position'))
-
-            raw_data = video_collection[position]['videoEntry']
-
-            video_id = str_or_none(raw_data['id'])
-
-            # create a title from many sources (while grabbing other info)
-            # TODO use more fallback sources to get some of these
-            series = raw_data.get('showTitle')
-            video_type = raw_data.get('videoEntryType', {}).get('name')
-            episode_number = raw_data.get('episodeNumber')
-            episode_name = raw_data.get('name')
-
-            title_parts = (series, video_type, episode_number, episode_name)
-            if None not in title_parts:
-                title = '%s - %s %s - %s' % title_parts
-            else:
-                title = episode_name
-
-            description = raw_data.get('synopsis') or self._og_search_description(webpage, default=None)
-
-            duration = int_or_none(raw_data.get('duration'))
-
-            thumbnail_data = raw_data.get('images', [])
-            thumbnails = []
-            for thumbnail in thumbnail_data:
-                for instance in thumbnail['imageInstances']:
-                    image_data = instance.get('imageInfo', {})
-                    thumbnails.append({
-                        'id': str_or_none(image_data.get('id')),
-                        'url': image_data.get('fullPath'),
-                        'width': image_data.get('width'),
-                        'height': image_data.get('height'),
-                    })
-
-            season_data = raw_data.get('season', {}) or {}
-            season = str_or_none(season_data.get('name'))
-            season_number = int_or_none(season_data.get('seasonNumber'))
-            season_id = str_or_none(season_data.get('id'))
-
-            for video_data in raw_data['videoList']:
-                current_video_list = {}
-                current_video_list['language'] = video_data.get('language', {}).get('languageCode')
-
-                is_hardsubbed = video_data.get('hardSubbed')
-
-                for video_instance in video_data['videoInstances']:
-                    httpurl = video_instance.get('httpUrl')
-                    url = httpurl if httpurl else video_instance.get('rtmpUrl')
-                    if url is None:
-                        # this video format is unavailable to the user (not premium etc.)
-                        continue
-
-                    current_format = current_video_list.copy()
-
-                    format_id_parts = []
-
-                    format_id_parts.append(str_or_none(video_instance.get('id')))
-
-                    if is_hardsubbed is not None:
-                        if is_hardsubbed:
-                            format_id_parts.append('yeshardsubbed')
-                        else:
-                            format_id_parts.append('nothardsubbed')
-
-                    format_id_parts.append(current_format['language'])
-
-                    format_id = '_'.join([x for x in format_id_parts if x is not None])
-
-                    ext = determine_ext(url)
-                    if ext == 'm3u8':
-                        for format_ in self._extract_m3u8_formats(
-                                url, video_id, m3u8_id=format_id, fatal=False):
-                            formats[format_['format_id']] = format_
-                        continue
-                    elif ext == 'mpd':
-                        for format_ in self._extract_mpd_formats(
-                                url, video_id, mpd_id=format_id, fatal=False):
-                            formats[format_['format_id']] = format_
-                        continue
-
-                    current_format['url'] = url
-                    quality_data = video_instance.get('videoQuality')
-                    if quality_data:
-                        quality = quality_data.get('name') or quality_data.get('description')
-                    else:
-                        quality = None
-
-                    height = None
-                    if quality:
-                        height = int_or_none(self._search_regex(r'(\d+)p?$', quality, 'Video format height', default=None))
-
-                    if height is None:
-                        self.report_warning('Could not get height of video')
-                    else:
-                        current_format['height'] = height
-                    current_format['format_id'] = format_id
-
-                    formats[current_format['format_id']] = current_format
-
-        formats = list(formats.values())
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'display_id': display_id,
-            'title': title,
-            'description': description,
-            'series': series,
-            'episode': episode_name,
-            'episode_number': int_or_none(episode_number),
-            'thumbnails': thumbnails,
-            'duration': duration,
-            'formats': formats,
-            'season': season,
-            'season_number': season_number,
-            'season_id': season_id,
-        }
-
-
-class AnimeLabShowsIE(AnimeLabBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?animelab\.com/shows/(?P<id>[^/]+)'
-
-    _TEST = {
-        'url': 'https://www.animelab.com/shows/attack-on-titan',
-        'info_dict': {
-            'id': '45',
-            'title': 'Attack on Titan',
-            'description': 'md5:989d95a2677e9309368d5cf39ba91469',
-        },
-        'playlist_count': 59,
-        'skip': 'All AnimeLab content requires authentication',
-    }
-
-    def _real_extract(self, url):
-        _BASE_URL = 'http://www.animelab.com'
-        _SHOWS_API_URL = '/api/videoentries/show/videos/'
-        display_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, display_id, 'Downloading requested URL')
-
-        show_data_str = self._search_regex(r'({"id":.*}),\svideoEntry', webpage, 'AnimeLab show data')
-        show_data = self._parse_json(show_data_str, display_id)
-
-        show_id = str_or_none(show_data.get('id'))
-        title = show_data.get('name')
-        description = show_data.get('shortSynopsis') or show_data.get('longSynopsis')
-
-        entries = []
-        for season in show_data['seasons']:
-            season_id = season['id']
-            get_data = urlencode_postdata({
-                'seasonId': season_id,
-                'limit': 1000,
-            })
-            # despite using urlencode_postdata, we are sending a GET request
-            target_url = _BASE_URL + _SHOWS_API_URL + show_id + "?" + get_data.decode('utf-8')
-            response = self._download_webpage(
-                target_url,
-                None, 'Season id %s' % season_id)
-
-            season_data = self._parse_json(response, display_id)
-
-            for video_data in season_data['list']:
-                entries.append(self.url_result(
-                    _BASE_URL + '/player/' + video_data['slug'], 'AnimeLab',
-                    str_or_none(video_data.get('id')), video_data.get('name')
-                ))
-
-        return {
-            '_type': 'playlist',
-            'id': show_id,
-            'title': title,
-            'description': description,
-            'entries': entries,
-        }
-
-# TODO implement myqueue

yt_dlp/extractor/archiveorg.py

@@ -1,36 +1,34 @@
-import re
 import json
+import re
+import urllib.parse
+
 from .common import InfoExtractor
-from .youtube import YoutubeIE, YoutubeBaseInfoExtractor
-from ..compat import (
-    compat_urllib_parse_unquote,
-    compat_urllib_parse_unquote_plus,
-    compat_HTTPError
-)
+from .youtube import YoutubeBaseInfoExtractor, YoutubeIE
+from ..compat import compat_HTTPError, compat_urllib_parse_unquote
 from ..utils import (
+    KNOWN_EXTENSIONS,
+    ExtractorError,
+    HEADRequest,
     bug_reports_message,
     clean_html,
     dict_get,
     extract_attributes,
-    ExtractorError,
     get_element_by_id,
-    HEADRequest,
     int_or_none,
     join_nonempty,
-    KNOWN_EXTENSIONS,
     merge_dicts,
     mimetype2ext,
     orderedSet,
     parse_duration,
     parse_qs,
-    str_to_int,
     str_or_none,
+    str_to_int,
     traverse_obj,
     try_get,
     unified_strdate,
     unified_timestamp,
+    url_or_none,
     urlhandle_detect_ext,
-    url_or_none
 )
 
 
@@ -143,7 +141,7 @@ class ArchiveOrgIE(InfoExtractor):
         return json.loads(extract_attributes(element)['value'])
 
     def _real_extract(self, url):
-        video_id = compat_urllib_parse_unquote_plus(self._match_id(url))
+        video_id = urllib.parse.unquote_plus(self._match_id(url))
         identifier, entry_id = (video_id.split('/', 1) + [None])[:2]
 
         # Archive.org metadata API doesn't clearly demarcate playlist entries
@@ -442,9 +440,10 @@ class YoutubeWebArchiveIE(InfoExtractor):
             'only_matching': True
         },
     ]
-    _YT_INITIAL_DATA_RE = r'(?:(?:(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;)|%s)' % YoutubeBaseInfoExtractor._YT_INITIAL_DATA_RE
-    _YT_INITIAL_PLAYER_RESPONSE_RE = r'(?:(?:(?:window\s*\[\s*["\']ytInitialPlayerResponse["\']\s*\]|ytInitialPlayerResponse)\s*=[(\s]*({.+?})[)\s]*;)|%s)' % YoutubeBaseInfoExtractor._YT_INITIAL_PLAYER_RESPONSE_RE
-    _YT_INITIAL_BOUNDARY_RE = r'(?:(?:var\s+meta|</script|\n)|%s)' % YoutubeBaseInfoExtractor._YT_INITIAL_BOUNDARY_RE
+    _YT_INITIAL_DATA_RE = YoutubeBaseInfoExtractor._YT_INITIAL_DATA_RE
+    _YT_INITIAL_PLAYER_RESPONSE_RE = fr'''(?x)
+        (?:window\s*\[\s*["\']ytInitialPlayerResponse["\']\s*\]|ytInitialPlayerResponse)\s*=[(\s]*|
+        {YoutubeBaseInfoExtractor._YT_INITIAL_PLAYER_RESPONSE_RE}'''
 
     _YT_DEFAULT_THUMB_SERVERS = ['i.ytimg.com']  # thumbnails most likely archived on these servers
     _YT_ALL_THUMB_SERVERS = orderedSet(
@@ -474,11 +473,6 @@ class YoutubeWebArchiveIE(InfoExtractor):
         elif not isinstance(res, list) or len(res) != 0:
             self.report_warning('Error while parsing CDX API response' + bug_reports_message())
 
-    def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
-        return self._parse_json(self._search_regex(
-            (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
-             regex), webpage, name, default='{}'), video_id, fatal=False)
-
     def _extract_webpage_title(self, webpage):
         page_title = self._html_extract_title(webpage, default='')
         # YouTube video pages appear to always have either 'YouTube -' as prefix or '- YouTube' as suffix.
@@ -488,10 +482,11 @@ class YoutubeWebArchiveIE(InfoExtractor):
 
     def _extract_metadata(self, video_id, webpage):
         search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None)) if webpage else (lambda x: None))
-        player_response = self._extract_yt_initial_variable(
-            webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE, video_id, 'initial player response') or {}
-        initial_data = self._extract_yt_initial_variable(
-            webpage, self._YT_INITIAL_DATA_RE, video_id, 'initial player response') or {}
+        player_response = self._search_json(
+            self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response',
+            video_id, default={})
+        initial_data = self._search_json(
+            self._YT_INITIAL_DATA_RE, webpage, 'initial data', video_id, default={})
 
         initial_data_video = traverse_obj(
             initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'videoPrimaryInfoRenderer'),

yt_dlp/extractor/arnes.py

@@ -90,7 +90,7 @@ class ArnesIE(InfoExtractor):
             'timestamp': parse_iso8601(video.get('creationTime')),
             'channel': channel.get('name'),
             'channel_id': channel_id,
-            'channel_url': format_field(channel_id, template=f'{self._BASE_URL}/?channel=%s'),
+            'channel_url': format_field(channel_id, None, f'{self._BASE_URL}/?channel=%s'),
             'duration': float_or_none(video.get('duration'), 1000),
             'view_count': int_or_none(video.get('views')),
             'tags': video.get('hashtags'),

yt_dlp/extractor/atscaleconf.py

@@ -0,0 +1,34 @@
+import re
+
+from .common import InfoExtractor
+
+
+class AtScaleConfEventIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?atscaleconference\.com/events/(?P<id>[^/&$?]+)'
+
+    _TESTS = [{
+        'url': 'https://atscaleconference.com/events/data-scale-spring-2022/',
+        'playlist_mincount': 13,
+        'info_dict': {
+            'id': 'data-scale-spring-2022',
+            'title': 'Data @Scale Spring 2022',
+            'description': 'md5:7d7ca1c42ac9c6d8a785092a1aea4b55'
+        },
+    }, {
+        'url': 'https://atscaleconference.com/events/video-scale-2021/',
+        'playlist_mincount': 14,
+        'info_dict': {
+            'id': 'video-scale-2021',
+            'title': 'Video @Scale 2021',
+            'description': 'md5:7d7ca1c42ac9c6d8a785092a1aea4b55'
+        },
+    }]
+
+    def _real_extract(self, url):
+        id = self._match_id(url)
+        webpage = self._download_webpage(url, id)
+
+        return self.playlist_from_matches(
+            re.findall(r'data-url\s*=\s*"(https?://(?:www\.)?atscaleconference\.com/videos/[^"]+)"', webpage),
+            ie='Generic', playlist_id=id,
+            title=self._og_search_title(webpage), description=self._og_search_description(webpage))