Release v0.4.0 - HD Thumbnails, YouTube 2024+ Support, and yt-dlp Integration

Major Features:
- HD video thumbnails (hq720.jpg) with automatic fallback to lower qualities
- HD channel avatars (240x240 instead of 88x88)
- YouTube 2024+ lockupViewModel support for channel playlists
- youtubei/v1/browse API integration for channel playlist tabs
- yt-dlp integration for multi-language audio and subtitles

Bug Fixes:
- Fixed undefined `abort` import in playlist.py
- Fixed undefined functions in proto.py (encode_varint, bytes_to_hex, succinct_encode)
- Fixed missing `traceback` import in proto_debug.py
- Fixed blurry playlist thumbnails using default.jpg instead of HD versions
- Fixed channel playlists page using deprecated pbj=1 format

Improvements:
- Automatic thumbnail fallback system (hq720 → sddefault → hqdefault → mqdefault → default)
- JavaScript thumbnail_fallback() handler for 404 errors
- Better thumbnail quality across all pages (watch, channel, playlist, subscriptions)
- Consistent HD avatar display for all channel items
- Settings system automatically adds new settings without breaking user config

Files Modified:
- youtube/watch.py - HD thumbnails for related videos and playlist items
- youtube/channel.py - HD thumbnails for channel playlists, youtubei API integration
- youtube/playlist.py - HD thumbnails, fixed abort import
- youtube/util.py - HD thumbnail URLs, avatar HD upgrade, prefix_url improvements
- youtube/comments.py - HD video thumbnail
- youtube/subscriptions.py - HD thumbnails, fixed abort import
- youtube/yt_data_extract/common.py - lockupViewModel support, extract_lockup_view_model_info()
- youtube/yt_data_extract/everything_else.py - HD playlist thumbnails
- youtube/proto.py - Fixed undefined function references
- youtube/proto_debug.py - Added traceback import
- youtube/static/js/common.js - thumbnail_fallback() handler
- youtube/templates/*.html - Added onerror handlers for thumbnail fallback
- youtube/version.py - Bump to v0.4.0

Technical Details:
- All thumbnail URLs now use hq720.jpg (1280x720) when available
- Fallback handled client-side via JavaScript onerror handler
- Server-side avatar upgrade via regex in util.prefix_url()
- lockupViewModel parser extracts contentType, metadata, and first_video_id
- Channel playlist tabs now use youtubei/v1/browse instead of deprecated pbj=1
- Settings version system ensures backward compatibility

.gitea/workflows/ci.yaml

@@ -0,0 +1,23 @@
+name: CI
+
+on: [push, pull_request]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: 3.11
+
+      - name: Install dependencies
+        run: |
+          pip install --upgrade pip
+          pip install -r requirements-dev.txt
+
+      - name: Run tests
+        run: pytest

.gitea/workflows/git-sync.yaml

@@ -0,0 +1,40 @@
+name: git-sync-with-mirror
+
+on:
+  push:
+    branches: [ master ]
+  workflow_dispatch:
+
+jobs:
+  git-sync:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: git-sync
+        env:
+          git_sync_source_repo: git@git.fridu.us:heckyel/yt-local.git
+          git_sync_destination_repo: ssh://git@c.fridu.us/software/yt-local.git
+        if: env.git_sync_source_repo && env.git_sync_destination_repo
+        uses: astounds/git-sync@v1
+        with:
+          source_repo: git@git.fridu.us:heckyel/yt-local.git
+          source_branch: "master"
+          destination_repo: ssh://git@c.fridu.us/software/yt-local.git
+          destination_branch: "master"
+          source_ssh_private_key: ${{ secrets.GIT_SYNC_SOURCE_SSH_PRIVATE_KEY }}
+          destination_ssh_private_key:  ${{ secrets.GIT_SYNC_DESTINATION_SSH_PRIVATE_KEY }}
+
+      - name: git-sync-sourcehut
+        env:
+          git_sync_source_repo: git@git.fridu.us:heckyel/yt-local.git
+          git_sync_destination_repo: git@git.sr.ht:~heckyel/yt-local
+        if: env.git_sync_source_repo && env.git_sync_destination_repo
+        uses: astounds/git-sync@v1
+        with:
+          source_repo: git@git.fridu.us:heckyel/yt-local.git
+          source_branch: "master"
+          destination_repo: git@git.sr.ht:~heckyel/yt-local
+          destination_branch: "master"
+          source_ssh_private_key: ${{ secrets.GIT_SYNC_SOURCE_SSH_PRIVATE_KEY }}
+          destination_ssh_private_key:  ${{ secrets.GIT_SYNC_DESTINATION_SSH_PRIVATE_KEY }}
+        continue-on-error: true

.gitignore

@@ -1,5 +1,128 @@
+# Byte-compiled / optimized / DLL files
 __pycache__/
+*.py[cod]
 *$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+Pipfile.lock
+
+# PEP 582
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+*venv*
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# Project specific
 debug/
 data/
 python/
@@ -11,5 +134,17 @@ get-pip.py
 latest-dist.zip
 *.7z
 *.zip
-*venv*
+
+# Editor specific
 flycheck_*
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+.DS_Store
+
+# Temporary files
+*.tmp
+*.bak
+*.orig

Makefile

@@ -0,0 +1,210 @@
+# yt-local Makefile
+# Automated tasks for development, translations, and maintenance
+
+.PHONY: help install dev clean test i18n-extract i18n-init i18n-update i18n-compile i18n-stats i18n-clean setup-dev lint format backup restore
+
+# Variables
+PYTHON := python3
+PIP := pip3
+LANG_CODE ?= es
+VENV_DIR := venv
+PROJECT_NAME := yt-local
+
+## Help
+help: ## Show this help message
+	@echo "$(PROJECT_NAME) - Available tasks:"
+	@echo ""
+	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "  %-20s %s\n", $$1, $$2}'
+	@echo ""
+	@echo "Examples:"
+	@echo "  make install          # Install dependencies"
+	@echo "  make dev              # Run development server"
+	@echo "  make i18n-extract     # Extract strings for translation"
+	@echo "  make i18n-init LANG_CODE=fr  # Initialize French"
+	@echo "  make lint             # Check code style"
+
+## Installation and Setup
+install: ## Install project dependencies
+	@echo "[INFO] Installing dependencies..."
+	$(PIP) install -r requirements.txt
+	@echo "[SUCCESS] Dependencies installed"
+
+setup-dev: ## Complete development setup
+	@echo "[INFO] Setting up development environment..."
+	$(PYTHON) -m venv $(VENV_DIR)
+	./$(VENV_DIR)/bin/pip install -r requirements.txt
+	@echo "[SUCCESS] Virtual environment created in $(VENV_DIR)"
+	@echo "[INFO] Activate with: source $(VENV_DIR)/bin/activate"
+
+requirements: ## Update and install requirements
+	@echo "[INFO] Installing/updating requirements..."
+	$(PIP) install --upgrade pip
+	$(PIP) install -r requirements.txt
+	@echo "[SUCCESS] Requirements installed"
+
+## Development
+dev: ## Run development server
+	@echo "[INFO] Starting development server..."
+	@echo "[INFO] Server available at: http://localhost:9010"
+	$(PYTHON) server.py
+
+run: dev ## Alias for dev
+
+## Testing
+test: ## Run tests
+	@echo "[INFO] Running tests..."
+	@if [ -d "tests" ]; then \
+		$(PYTHON) -m pytest -v; \
+	else \
+		echo "[WARN] No tests directory found"; \
+	fi
+
+test-cov: ## Run tests with coverage
+	@echo "[INFO] Running tests with coverage..."
+	@if command -v pytest-cov >/dev/null 2>&1; then \
+		$(PYTHON) -m pytest -v --cov=$(PROJECT_NAME) --cov-report=html; \
+	else \
+		echo "[WARN] pytest-cov not installed. Run: pip install pytest-cov"; \
+	fi
+
+## Internationalization (i18n)
+i18n-extract: ## Extract strings for translation
+	@echo "[INFO] Extracting strings for translation..."
+	$(PYTHON) manage_translations.py extract
+	@echo "[SUCCESS] Strings extracted to translations/messages.pot"
+
+i18n-init: ## Initialize new language (use LANG_CODE=xx)
+	@echo "[INFO] Initializing language: $(LANG_CODE)"
+	$(PYTHON) manage_translations.py init $(LANG_CODE)
+	@echo "[SUCCESS] Language $(LANG_CODE) initialized"
+	@echo "[INFO] Edit: translations/$(LANG_CODE)/LC_MESSAGES/messages.po"
+
+i18n-update: ## Update existing translations
+	@echo "[INFO] Updating existing translations..."
+	$(PYTHON) manage_translations.py update
+	@echo "[SUCCESS] Translations updated"
+
+i18n-compile: ## Compile translations to binary .mo files
+	@echo "[INFO] Compiling translations..."
+	$(PYTHON) manage_translations.py compile
+	@echo "[SUCCESS] Translations compiled"
+
+i18n-stats: ## Show translation statistics
+	@echo "[INFO] Translation statistics:"
+	@echo ""
+	@for lang_dir in translations/*/; do \
+		if [ -d "$$lang_dir" ] && [ "$$lang_dir" != "translations/*/" ]; then \
+			lang=$$(basename "$$lang_dir"); \
+			po_file="$$lang_dir/LC_MESSAGES/messages.po"; \
+			if [ -f "$$po_file" ]; then \
+				total=$$(grep -c "^msgid " "$$po_file" 2>/dev/null || echo "0"); \
+				translated=$$(grep -c "^msgstr \"[^\"]\+\"" "$$po_file" 2>/dev/null || echo "0"); \
+				fuzzy=$$(grep -c "^#, fuzzy" "$$po_file" 2>/dev/null || echo "0"); \
+				if [ "$$total" -gt 0 ]; then \
+					percent=$$((translated * 100 / total)); \
+					echo "  [STAT] $$lang: $$translated/$$total ($$percent%) - Fuzzy: $$fuzzy"; \
+				else \
+					echo "  [STAT] $$lang: No translations yet"; \
+				fi; \
+			fi \
+		fi \
+	done
+	@echo ""
+
+i18n-clean: ## Clean compiled translation files
+	@echo "[INFO] Cleaning compiled .mo files..."
+	find translations/ -name "*.mo" -delete
+	@echo "[SUCCESS] .mo files removed"
+
+i18n-workflow: ## Complete workflow: extract → update → compile
+	@echo "[INFO] Running complete translation workflow..."
+	@make i18n-extract
+	@make i18n-update
+	@make i18n-compile
+	@make i18n-stats
+	@echo "[SUCCESS] Translation workflow completed"
+
+## Code Quality
+lint: ## Check code with flake8
+	@echo "[INFO] Checking code style..."
+	@if command -v flake8 >/dev/null 2>&1; then \
+		flake8 youtube/ --max-line-length=120 --ignore=E501,W503,E402 --exclude=youtube/ytdlp_service.py,youtube/ytdlp_integration.py,youtube/ytdlp_proxy.py; \
+		echo "[SUCCESS] Code style check passed"; \
+	else \
+		echo "[WARN] flake8 not installed (pip install flake8)"; \
+	fi
+
+format: ## Format code with black (if available)
+	@echo "[INFO] Formatting code..."
+	@if command -v black >/dev/null 2>&1; then \
+		black youtube/ --line-length=120 --exclude='ytdlp_.*\.py'; \
+		echo "[SUCCESS] Code formatted"; \
+	else \
+		echo "[WARN] black not installed (pip install black)"; \
+	fi
+
+check-deps: ## Check installed dependencies
+	@echo "[INFO] Checking dependencies..."
+	@$(PYTHON) -c "import flask_babel; print('[OK] Flask-Babel:', flask_babel.__version__)" 2>/dev/null || echo "[ERROR] Flask-Babel not installed"
+	@$(PYTHON) -c "import flask; print('[OK] Flask:', flask.__version__)" 2>/dev/null || echo "[ERROR] Flask not installed"
+	@$(PYTHON) -c "import yt_dlp; print('[OK] yt-dlp:', yt_dlp.__version__)" 2>/dev/null || echo "[ERROR] yt-dlp not installed"
+
+## Maintenance
+backup: ## Create translations backup
+	@echo "[INFO] Creating translations backup..."
+	@timestamp=$$(date +%Y%m%d_%H%M%S); \
+	tar -czf "translations_backup_$$timestamp.tar.gz" translations/ 2>/dev/null || echo "[WARN] No translations to backup"; \
+	if [ -f "translations_backup_$$timestamp.tar.gz" ]; then \
+		echo "[SUCCESS] Backup created: translations_backup_$$timestamp.tar.gz"; \
+	fi
+
+restore: ## Restore translations from backup
+	@echo "[INFO] Restoring translations from backup..."
+	@if ls translations_backup_*.tar.gz 1>/dev/null 2>&1; then \
+		latest_backup=$$(ls -t translations_backup_*.tar.gz | head -1); \
+		tar -xzf "$$latest_backup"; \
+		echo "[SUCCESS] Restored from: $$latest_backup"; \
+	else \
+		echo "[ERROR] No backup files found"; \
+	fi
+
+clean: ## Clean temporary files and caches
+	@echo "[INFO] Cleaning temporary files..."
+	find . -type f -name "*.pyc" -delete
+	find . -type d -name "__pycache__" -delete
+	find . -type f -name "*.mo" -delete
+	find . -type d -name ".pytest_cache" -delete
+	find . -type f -name ".coverage" -delete
+	find . -type d -name "htmlcov" -delete
+	@echo "[SUCCESS] Temporary files removed"
+
+distclean: clean ## Clean everything including venv
+	@echo "[INFO] Cleaning everything..."
+	rm -rf $(VENV_DIR)
+	@echo "[SUCCESS] Complete cleanup done"
+
+## Project Information
+info: ## Show project information
+	@echo "[INFO] $(PROJECT_NAME) - Project information:"
+	@echo ""
+	@echo "  [INFO] Directory: $$(pwd)"
+	@echo "  [INFO] Python: $$($(PYTHON) --version)"
+	@echo "  [INFO] Pip: $$($(PIP) --version | cut -d' ' -f1-2)"
+	@echo ""
+	@echo "  [INFO] Configured languages:"
+	@for lang_dir in translations/*/; do \
+		if [ -d "$$lang_dir" ] && [ "$$lang_dir" != "translations/*/" ]; then \
+			lang=$$(basename "$$lang_dir"); \
+			echo "    - $$lang"; \
+		fi \
+	done
+	@echo ""
+	@echo "  [INFO] Main files:"
+	@echo "    - babel.cfg (i18n configuration)"
+	@echo "    - manage_translations.py (i18n CLI)"
+	@echo "    - youtube/i18n_strings.py (centralized strings)"
+	@echo "    - youtube/ytdlp_service.py (yt-dlp integration)"
+	@echo ""
+
+# Default target
+.DEFAULT_GOAL := help

README.md

@@ -151,7 +151,7 @@ For coding guidelines and an overview of the software architecture, see the [HAC
 
 yt-local is not made to work in public mode, however there is an instance of yt-local in public mode but with less features
 
-- <https://1cd1-93-95-230-133.ngrok-free.app/https://youtube.com>
+- <https://m.fridu.us/https://youtube.com>
 
 ## License
 

babel.cfg

@@ -0,0 +1,7 @@
+[python: youtube/**.py]
+keywords = lazy_gettext:1,2 _l:1,2
+[python: server.py]
+[python: settings.py]
+[jinja2: youtube/templates/**.html]
+extensions=jinja2.ext.i18n
+encoding = utf-8

manage_translations.py

@@ -0,0 +1,113 @@
+#!/usr/bin/env python3
+"""
+Translation management script for yt-local
+
+Usage:
+    python manage_translations.py extract  # Extract strings to messages.pot
+    python manage_translations.py init es  # Initialize Spanish translation
+    python manage_translations.py update   # Update all translations
+    python manage_translations.py compile  # Compile translations to .mo files
+"""
+import sys
+import os
+import subprocess
+
+# Ensure we use the Python from the virtual environment if available
+if hasattr(sys, 'real_prefix') or (hasattr(sys, 'base_prefix') and sys.base_prefix != sys.prefix):
+    # Already in venv
+    pass
+else:
+    # Try to activate venv
+    venv_path = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), 'venv')
+    if os.path.exists(venv_path):
+        venv_bin = os.path.join(venv_path, 'bin')
+        if os.path.exists(venv_bin):
+            os.environ['PATH'] = venv_bin + os.pathsep + os.environ['PATH']
+
+
+def run_command(cmd):
+    """Run a shell command and print output"""
+    print(f"Running: {' '.join(cmd)}")
+    # Use the pybabel from the same directory as our Python executable
+    if cmd[0] == 'pybabel':
+        import os
+        pybabel_path = os.path.join(os.path.dirname(sys.executable), 'pybabel')
+        if os.path.exists(pybabel_path):
+            cmd = [pybabel_path] + cmd[1:]
+    result = subprocess.run(cmd, capture_output=True, text=True)
+    if result.stdout:
+        print(result.stdout)
+    if result.stderr:
+        print(result.stderr, file=sys.stderr)
+    return result.returncode
+
+
+def extract():
+    """Extract translatable strings from source code"""
+    print("Extracting translatable strings...")
+    return run_command([
+        'pybabel', 'extract',
+        '-F', 'babel.cfg',
+        '-k', 'lazy_gettext',
+        '-k', '_l',
+        '-o', 'translations/messages.pot',
+        '.'
+    ])
+
+
+def init(language):
+    """Initialize a new language translation"""
+    print(f"Initializing {language} translation...")
+    return run_command([
+        'pybabel', 'init',
+        '-i', 'translations/messages.pot',
+        '-d', 'translations',
+        '-l', language
+    ])
+
+
+def update():
+    """Update existing translations with new strings"""
+    print("Updating translations...")
+    return run_command([
+        'pybabel', 'update',
+        '-i', 'translations/messages.pot',
+        '-d', 'translations'
+    ])
+
+
+def compile_translations():
+    """Compile .po files to .mo files"""
+    print("Compiling translations...")
+    return run_command([
+        'pybabel', 'compile',
+        '-d', 'translations'
+    ])
+
+
+def main():
+    if len(sys.argv) < 2:
+        print(__doc__)
+        sys.exit(1)
+
+    command = sys.argv[1]
+
+    if command == 'extract':
+        sys.exit(extract())
+    elif command == 'init':
+        if len(sys.argv) < 3:
+            print("Error: Please specify a language code (e.g., es, fr, de)")
+            sys.exit(1)
+        sys.exit(init(sys.argv[2]))
+    elif command == 'update':
+        sys.exit(update())
+    elif command == 'compile':
+        sys.exit(compile_translations())
+    else:
+        print(f"Unknown command: {command}")
+        print(__doc__)
+        sys.exit(1)
+
+
+if __name__ == '__main__':
+    main()

requirements-dev.txt

@@ -1,21 +1,5 @@
-blinker==1.7.0
-Brotli==1.1.0
-cachetools==5.3.3
-click==8.1.7
-defusedxml==0.7.1
-Flask==3.0.2
-gevent==24.2.1
-greenlet==3.0.3
-iniconfig==2.0.0
-itsdangerous==2.1.2
-Jinja2==3.1.3
-MarkupSafe==2.1.5
-packaging==24.0
-pluggy==1.4.0
-PySocks==1.7.1
-pytest==8.1.1
-stem==1.8.2
-urllib3==2.2.1
-Werkzeug==3.0.1
-zope.event==5.0
-zope.interface==6.2
+# Include all production requirements
+-r requirements.txt
+
+# Development requirements
+pytest>=6.2.1

requirements.txt

@@ -1,17 +1,12 @@
-blinker==1.7.0
-Brotli==1.1.0
-cachetools==5.3.3
-click==8.1.7
-defusedxml==0.7.1
-Flask==3.0.2
-gevent==24.2.1
-greenlet==3.0.3
-itsdangerous==2.1.2
-Jinja2==3.1.3
-MarkupSafe==2.1.5
-PySocks==1.7.1
-stem==1.8.2
-urllib3==2.2.1
-Werkzeug==3.0.1
-zope.event==5.0
-zope.interface==6.2
+Flask>=1.0.3
+Flask-Babel>=4.0.0
+Babel>=2.12.0
+gevent>=1.2.2
+Brotli>=1.0.7
+PySocks>=1.6.8
+urllib3>=1.24.1
+defusedxml>=0.5.0
+cachetools>=4.0.0
+stem>=1.8.0
+yt-dlp>=2026.01.01
+requests>=2.25.0

server.py

@@ -279,6 +279,16 @@ if __name__ == '__main__':
 
     print('Starting httpserver at http://%s:%s/' %
           (ip_server, settings.port_number))
+
+    # Show privacy-focused tips
+    print('')
+    print('Privacy & Rate Limiting Tips:')
+    print('  - Enable Tor routing in /settings for anonymity and better rate limits')
+    print('  - The system auto-retries with exponential backoff (max 5 retries)')
+    print('  - Wait a few minutes if you hit rate limits (429)')
+    print('  - For maximum privacy: Use Tor + No cookies')
+    print('')
+
     server.serve_forever()
 
 # for uwsgi, gunicorn, etc.

settings.py

@@ -296,6 +296,17 @@ Archive: https://archive.ph/OZQbN''',
         'category': 'interface',
     }),
 
+    ('language', {
+        'type': str,
+        'default': 'en',
+        'comment': 'Interface language',
+        'options': [
+            ('en', 'English'),
+            ('es', 'Español'),
+        ],
+        'category': 'interface',
+    }),
+
     ('embed_page_mode', {
         'type': bool,
         'label': 'Enable embed page',
@@ -322,13 +333,6 @@ Archive: https://archive.ph/OZQbN''',
         'comment': '',
     }),
 
-    ('gather_googlevideo_domains', {
-        'type': bool,
-        'default': False,
-        'comment': '''Developer use to debug 403s''',
-        'hidden': True,
-    }),
-
     ('debugging_save_responses', {
         'type': bool,
         'default': False,
@@ -336,9 +340,18 @@ Archive: https://archive.ph/OZQbN''',
         'hidden': True,
     }),
 
+    ('ytdlp_enabled', {
+        'type': bool,
+        'default': True,
+        'comment': '''Enable yt-dlp integration for multi-language audio and subtitles''',
+        'hidden': False,
+        'label': 'Enable yt-dlp integration',
+        'category': 'playback',
+    }),
+
     ('settings_version', {
         'type': int,
-        'default': 5,
+        'default': 6,
         'comment': '''Do not change, remove, or comment out this value, or else your settings may be lost or corrupted''',
         'hidden': True,
     }),
@@ -419,11 +432,20 @@ def upgrade_to_5(settings_dict):
     return new_settings
 
 
+def upgrade_to_6(settings_dict):
+    new_settings = settings_dict.copy()
+    if 'gather_googlevideo_domains' in new_settings:
+        del new_settings['gather_googlevideo_domains']
+    new_settings['settings_version'] = 6
+    return new_settings
+
+
 upgrade_functions = {
     1: upgrade_to_2,
     2: upgrade_to_3,
     3: upgrade_to_4,
     4: upgrade_to_5,
+    5: upgrade_to_6,
 }
 
 

translations/es/LC_MESSAGES/messages.po

@@ -0,0 +1,74 @@
+# Spanish translations for yt-local.
+# Copyright (C) 2026 yt-local
+# This file is distributed under the same license as the yt-local project.
+#
+msgid ""
+msgstr ""
+"Project-Id-Version: PROJECT VERSION\n"
+"Report-Msgid-Bugs-To: EMAIL@ADDRESS\n"
+"POT-Creation-Date: 2026-03-22 15:05-0500\n"
+"PO-Revision-Date: 2026-03-22 15:06-0500\n"
+"Last-Translator: \n"
+"Language: es\n"
+"Language-Team: es <LL@li.org>\n"
+"Plural-Forms: nplurals=2; plural=(n != 1);\n"
+"MIME-Version: 1.0\n"
+"Content-Type: text/plain; charset=utf-8\n"
+"Content-Transfer-Encoding: 8bit\n"
+"Generated-By: Babel 2.18.0\n"
+
+#: youtube/templates/base.html:38
+msgid "Type to search..."
+msgstr "Escribe para buscar..."
+
+#: youtube/templates/base.html:39
+msgid "Search"
+msgstr "Buscar"
+
+#: youtube/templates/base.html:45
+msgid "Options"
+msgstr "Opciones"
+
+#: youtube/templates/base.html:47
+msgid "Sort by"
+msgstr "Ordenar por"
+
+#: youtube/templates/base.html:50
+msgid "Relevance"
+msgstr "Relevancia"
+
+#: youtube/templates/base.html:54 youtube/templates/base.html:65
+msgid "Upload date"
+msgstr "Fecha de subida"
+
+#: youtube/templates/base.html:58
+msgid "View count"
+msgstr "Número de visualizaciones"
+
+#: youtube/templates/base.html:62
+msgid "Rating"
+msgstr "Calificación"
+
+#: youtube/templates/base.html:68
+msgid "Any"
+msgstr "Cualquiera"
+
+#: youtube/templates/base.html:72
+msgid "Last hour"
+msgstr "Última hora"
+
+#: youtube/templates/base.html:76
+msgid "Today"
+msgstr "Hoy"
+
+#: youtube/templates/base.html:80
+msgid "This week"
+msgstr "Esta semana"
+
+#: youtube/templates/base.html:84
+msgid "This month"
+msgstr "Este mes"
+
+#: youtube/templates/base.html:88
+msgid "This year"
+msgstr "Este año"

translations/messages.pot

@@ -0,0 +1,75 @@
+# Translations template for PROJECT.
+# Copyright (C) 2026 ORGANIZATION
+# This file is distributed under the same license as the PROJECT project.
+# FIRST AUTHOR <EMAIL@ADDRESS>, 2026.
+#
+#, fuzzy
+msgid ""
+msgstr ""
+"Project-Id-Version: PROJECT VERSION\n"
+"Report-Msgid-Bugs-To: EMAIL@ADDRESS\n"
+"POT-Creation-Date: 2026-03-22 15:05-0500\n"
+"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
+"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
+"Language-Team: LANGUAGE <LL@li.org>\n"
+"MIME-Version: 1.0\n"
+"Content-Type: text/plain; charset=utf-8\n"
+"Content-Transfer-Encoding: 8bit\n"
+"Generated-By: Babel 2.18.0\n"
+
+#: youtube/templates/base.html:38
+msgid "Type to search..."
+msgstr ""
+
+#: youtube/templates/base.html:39
+msgid "Search"
+msgstr ""
+
+#: youtube/templates/base.html:45
+msgid "Options"
+msgstr ""
+
+#: youtube/templates/base.html:47
+msgid "Sort by"
+msgstr ""
+
+#: youtube/templates/base.html:50
+msgid "Relevance"
+msgstr ""
+
+#: youtube/templates/base.html:54 youtube/templates/base.html:65
+msgid "Upload date"
+msgstr ""
+
+#: youtube/templates/base.html:58
+msgid "View count"
+msgstr ""
+
+#: youtube/templates/base.html:62
+msgid "Rating"
+msgstr ""
+
+#: youtube/templates/base.html:68
+msgid "Any"
+msgstr ""
+
+#: youtube/templates/base.html:72
+msgid "Last hour"
+msgstr ""
+
+#: youtube/templates/base.html:76
+msgid "Today"
+msgstr ""
+
+#: youtube/templates/base.html:80
+msgid "This week"
+msgstr ""
+
+#: youtube/templates/base.html:84
+msgid "This month"
+msgstr ""
+
+#: youtube/templates/base.html:88
+msgid "This year"
+msgstr ""
+

youtube/__init__.py

@@ -7,12 +7,36 @@ import settings
 import traceback
 import re
 from sys import exc_info
+from flask_babel import Babel
+
 yt_app = flask.Flask(__name__)
 yt_app.config['TEMPLATES_AUTO_RELOAD'] = True
 yt_app.url_map.strict_slashes = False
 # yt_app.jinja_env.trim_blocks = True
 # yt_app.jinja_env.lstrip_blocks = True
 
+# Configure Babel for i18n
+import os
+yt_app.config['BABEL_DEFAULT_LOCALE'] = 'en'
+# Use absolute path for translations directory to avoid issues with package structure changes
+_app_root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+yt_app.config['BABEL_TRANSLATION_DIRECTORIES'] = os.path.join(_app_root, 'translations')
+
+def get_locale():
+    """Determine the best locale based on user preference or browser settings"""
+    # Check if user has a language preference in settings
+    if hasattr(settings, 'language') and settings.language:
+        locale = settings.language
+        print(f'[i18n] Using user preference: {locale}')
+        return locale
+    # Otherwise, use browser's Accept-Language header
+    # Only match languages with available translations
+    locale = request.accept_languages.best_match(['en', 'es'])
+    print(f'[i18n] Using browser language: {locale}')
+    return locale or 'en'
+
+babel = Babel(yt_app, locale_selector=get_locale)
+
 
 yt_app.add_url_rule('/settings', 'settings_page', settings.settings_page, methods=['POST', 'GET'])
 
@@ -113,15 +137,28 @@ def error_page(e):
             error_message += '\n\nExit node IP address: ' + exc_info()[1].ip
         return flask.render_template('error.html', error_message=error_message, slim=slim), 502
     elif exc_info()[0] == util.FetchError and exc_info()[1].error_message:
+        # Handle specific error codes with user-friendly messages
+        error_code = exc_info()[1].code
+        error_msg = exc_info()[1].error_message
+
+        if error_code == '400':
+            error_message = (f'Error: Bad Request (400)\n\n{error_msg}\n\n'
+                           'This usually means the URL or parameters are invalid. '
+                           'Try going back and trying a different option.')
+        elif error_code == '404':
+            error_message = 'Error: The page you are looking for isn\'t here.'
+        else:
+            error_message = f'Error: {error_code} - {error_msg}'
+
         return (flask.render_template(
                     'error.html',
-                    error_message=exc_info()[1].error_message,
+                    error_message=error_message,
                     slim=slim
                 ), 502)
     elif (exc_info()[0] == util.FetchError
         and exc_info()[1].code == '404'
     ):
-        error_message = ('Error: The page you are looking for isn\'t here. ¯\_(ツ)_/¯')
+        error_message = ('Error: The page you are looking for isn\'t here.')
         return flask.render_template('error.html',
                                      error_code=exc_info()[1].code,
                                      error_message=error_message,

youtube/channel.py

@@ -33,53 +33,75 @@ headers_mobile = (
 real_cookie = (('Cookie', 'VISITOR_INFO1_LIVE=8XihrAcN1l4'),)
 generic_cookie = (('Cookie', 'VISITOR_INFO1_LIVE=ST1Ti53r4fU'),)
 
-# added an extra nesting under the 2nd base64 compared to v4
-# added tab support
-# changed offset field to uint id 1
+# FIXED 2026: YouTube changed continuation token structure (from Invidious commit a9f8127)
+# Sort values for YouTube API (from Invidious): 2=popular, 4=newest, 5=oldest
 def channel_ctoken_v5(channel_id, page, sort, tab, view=1):
-    new_sort = (2 if int(sort) == 1 else 1)
+    # Map sort values to YouTube API values (Invidious values)
+    # Input: sort=3 (newest), sort=4 (newest no shorts)
+    # YouTube expects: 4=newest
+    sort_mapping = {'1': 2, '2': 5, '3': 4, '4': 4}  # 4 is newest without shorts
+    new_sort = sort_mapping.get(sort, 4)
+
     offset = 30*(int(page) - 1)
-    if tab == 'videos':
-        tab = 15
-    elif tab == 'shorts':
-        tab = 10
-    elif tab == 'streams':
-        tab = 14
+
+    # Build continuation token using Invidious structure
+    # The structure is: base64(protobuf({
+    #   80226972: {
+    #     2: channel_id,
+    #     3: base64(protobuf({
+    #       110: {
+    #         3: {
+    #           tab: {
+    #             1: {
+    #               1: base64(protobuf({
+    #                 1: base64(protobuf({
+    #                   2: "ST:" + base64(offset_varint)
+    #                 }))
+    #               }))
+    #             },
+    #             2: base64(protobuf({1: UUID}))
+    #             4: sort_value
+    #             8: base64(protobuf({
+    #               1: UUID
+    #               3: sort_value
+    #             }))
+    #           }
+    #         }
+    #       }
+    #     }))
+    #   }
+    # }))
+
+    # UUID placeholder
+    uuid_proto = proto.string(1, "00000000-0000-0000-0000-000000000000")
+
+    # Offset encoding
+    offset_varint = proto.uint(1, offset)
+    offset_encoded = proto.string(2, proto.unpadded_b64encode(offset_varint))
+    offset_wrapper = proto.string(1, proto.unpadded_b64encode(offset_encoded))
+    offset_base = proto.string(1, proto.unpadded_b64encode(offset_wrapper))
+
+    # Sort value varint
+    sort_varint = proto.uint(4, new_sort)
+
+    # Embedded message with UUID and sort
+    embedded_inner = uuid_proto + proto.uint(3, new_sort)
+    embedded_encoded = proto.string(8, proto.unpadded_b64encode(embedded_inner))
+
+    # Combine: uuid_wrapper + sort_varint + embedded
+    tab_inner_content = offset_base + uuid_proto + sort_varint + embedded_encoded
+
+    tab_inner = proto.string(1, proto.unpadded_b64encode(tab_inner_content))
+    tab_wrapper = proto.string(tab, tab_inner)
+
+    inner_container = proto.string(3, tab_wrapper)
+    outer_container = proto.string(110, inner_container)
+
+    encoded_inner = proto.percent_b64encode(outer_container)
+
     pointless_nest = proto.string(80226972,
         proto.string(2, channel_id)
-        + proto.string(3,
-            proto.percent_b64encode(
-                proto.string(110,
-                    proto.string(3,
-                        proto.string(tab,
-                            proto.string(1,
-                                proto.string(1,
-                                    proto.unpadded_b64encode(
-                                        proto.string(1,
-                                        proto.string(1,
-                                            proto.unpadded_b64encode(
-                                                proto.string(2,
-                                                    b"ST:"
-                                                    + proto.unpadded_b64encode(
-                                                        proto.uint(1, offset)
-                                                    )
-                                                )
-                                            )
-                                        )
-                                        )
-                                    )
-                                )
-                                 # targetId, just needs to be present but
-                                 # doesn't need to be correct
-                                + proto.string(2, "63faaff0-0000-23fe-80f0-582429d11c38")
-                            )
-                            # 1 - newest, 2 - popular
-                            + proto.uint(3, new_sort)
-                        )
-                    )
-                )
-            )
-        )
+        + proto.string(3, encoded_inner)
     )
 
     return base64.urlsafe_b64encode(pointless_nest).decode('ascii')
@@ -161,11 +183,6 @@ def channel_ctoken_v4(channel_id, page, sort, tab, view=1):
 
 # SORT:
 # videos:
-#    Popular - 1
-#    Oldest - 2
-#    Newest - 3
-# playlists:
-#    Oldest - 2
 #    Newest - 3
 #    Last video added - 4
 
@@ -292,7 +309,7 @@ def get_number_of_videos_channel(channel_id):
     try:
         response = util.fetch_url(url, headers_mobile,
             debug_name='number_of_videos', report_text='Got number of videos')
-    except urllib.error.HTTPError as e:
+    except (urllib.error.HTTPError, util.FetchError) as e:
         traceback.print_exc()
         print("Couldn't retrieve number of videos")
         return 1000
@@ -389,7 +406,12 @@ def post_process_channel_info(info):
     info['avatar'] = util.prefix_url(info['avatar'])
     info['channel_url'] = util.prefix_url(info['channel_url'])
     for item in info['items']:
-        item['thumbnail'] = "https://i.ytimg.com/vi/{}/hqdefault.jpg".format(item['id'])
+        # For playlists, use first_video_id for thumbnail, not playlist id
+        if item.get('type') == 'playlist' and item.get('first_video_id'):
+            item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['first_video_id'])
+        elif item.get('type') == 'video':
+            item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['id'])
+        # For channels and other types, keep existing thumbnail
         util.prefix_urls(item)
         util.add_extra_html_info(item)
     if info['current_tab'] == 'about':
@@ -398,11 +420,20 @@ def post_process_channel_info(info):
                 info['links'][i] = (text, util.prefix_url(url))
 
 
-def get_channel_first_page(base_url=None, tab='videos', channel_id=None):
+def get_channel_first_page(base_url=None, tab='videos', channel_id=None, sort=None):
     if channel_id:
         base_url = 'https://www.youtube.com/channel/' + channel_id
-    return util.fetch_url(base_url + '/' + tab + '?pbj=1&view=0',
-                          headers_desktop, debug_name='gen_channel_' + tab)
+
+    # Build URL with sort parameter
+    # YouTube URL sort params: p=popular, dd=newest, lad=newest no shorts
+    # Note: 'da' (oldest) was removed by YouTube in January 2026
+    url = base_url + '/' + tab + '?pbj=1&view=0'
+    if sort:
+        # Map sort values to YouTube's URL parameter values
+        sort_map = {'3': 'dd', '4': 'lad'}
+        url += '&sort=' + sort_map.get(sort, 'dd')
+
+    return util.fetch_url(url, headers_desktop, debug_name='gen_channel_' + tab)
 
 
 playlist_sort_codes = {'2': "da", '3': "dd", '4': "lad"}
@@ -416,7 +447,6 @@ def get_channel_page_general_url(base_url, tab, request, channel_id=None):
     page_number = int(request.args.get('page', 1))
     # sort 1: views
     # sort 2: oldest
-    # sort 3: newest
     # sort 4: newest - no shorts (Just a kludge on our end, not internal to yt)
     default_sort = '3' if settings.include_shorts_in_channel else '4'
     sort = request.args.get('sort', default_sort)
@@ -483,17 +513,15 @@ def get_channel_page_general_url(base_url, tab, request, channel_id=None):
         else:
             num_videos_call = (get_number_of_videos_general, base_url)
 
-        # Use ctoken method, which YouTube changes all the time
-        if channel_id and not default_params:
-            if sort == 4:
-                _sort = 3
-            else:
-                _sort = sort
-            page_call = (get_channel_tab, channel_id, page_number, _sort,
-                         tab, view, ctoken)
-        # Use the first-page method, which won't break
+        # For page 1, use the first-page method which won't break
+        # Pass sort parameter directly (2=oldest, 3=newest, etc.)
+        if page_number == 1:
+            # Always use first-page method for page 1 with sort parameter
+            page_call = (get_channel_first_page, base_url, tab, None, sort)
         else:
-            page_call = (get_channel_first_page, base_url, tab)
+            # For page 2+, we can't paginate without continuation tokens
+            # This is a YouTube limitation, not our bug
+            flask.abort(404, 'Pagination not available for this sort option. YouTube removed this feature.')
 
         tasks = (
             gevent.spawn(*num_videos_call),
@@ -512,7 +540,14 @@ def get_channel_page_general_url(base_url, tab, request, channel_id=None):
         })
         continuation=True
     elif tab == 'playlists' and page_number == 1:
-        polymer_json = util.fetch_url(base_url+ '/playlists?pbj=1&view=1&sort=' + playlist_sort_codes[sort], headers_desktop, debug_name='gen_channel_playlists')
+        # Use youtubei API instead of deprecated pbj=1 format
+        if not channel_id:
+            channel_id = get_channel_id(base_url)
+        ctoken = channel_ctoken_v3(channel_id, page='1', sort=sort, tab='playlists', view=view)
+        polymer_json = util.call_youtube_api('web', 'browse', {
+            'continuation': ctoken,
+        })
+        continuation = True
     elif tab == 'playlists':
         polymer_json = get_channel_tab(channel_id, page_number, sort,
                                        'playlists', view)

youtube/comments.py

@@ -53,7 +53,7 @@ def request_comments(ctoken, replies=False):
                 'hl': 'en',
                 'gl': 'US',
                 'clientName': 'MWEB',
-                'clientVersion': '2.20240328.08.00',
+                'clientVersion': '2.20210804.02.00',
             },
         },
         'continuation': ctoken.replace('=', '%3D'),
@@ -78,7 +78,7 @@ def single_comment_ctoken(video_id, comment_id):
 
 def post_process_comments_info(comments_info):
     for comment in comments_info['comments']:
-        comment['author'] = strip_non_ascii(comment['author'])
+        comment['author'] = strip_non_ascii(comment['author']) if comment.get('author') else ""
         comment['author_url'] = concat_or_none(
             '/', comment['author_url'])
         comment['author_avatar'] = concat_or_none(
@@ -150,7 +150,7 @@ def post_process_comments_info(comments_info):
         util.URL_ORIGIN, '/watch?v=', comments_info['video_id'])
     comments_info['video_thumbnail'] = concat_or_none(
         settings.img_prefix, 'https://i.ytimg.com/vi/',
-        comments_info['video_id'], '/hqdefault.jpg'
+        comments_info['video_id'], '/hq720.jpg'
     )
 
 
@@ -189,10 +189,10 @@ def video_comments(video_id, sort=0, offset=0, lc='', secret_key=''):
                 comments_info['error'] += '\n\n' + e.error_message
             comments_info['error'] += '\n\nExit node IP address: %s' % e.ip
         else:
-            comments_info['error'] = 'YouTube blocked the request. IP address: %s' % e.ip
+            comments_info['error'] = 'YouTube blocked the request. Error: %s' % str(e)
 
     except Exception as e:
-        comments_info['error'] = 'YouTube blocked the request. IP address: %s' % e.ip
+        comments_info['error'] = 'YouTube blocked the request. Error: %s' % str(e)
 
     if comments_info.get('error'):
         print('Error retrieving comments for ' + str(video_id) + ':\n' +

youtube/get_app_version/get_app_version.py

@@ -11,17 +11,10 @@ import subprocess
 def app_version():
     def minimal_env_cmd(cmd):
         # make minimal environment
-        env = {}
-        for k in ['SYSTEMROOT', 'PATH']:
-            v = os.environ.get(k)
-            if v is not None:
-                env[k] = v
+        env = {k: os.environ[k] for k in ['SYSTEMROOT', 'PATH'] if k in os.environ}
+        env.update({'LANGUAGE': 'C', 'LANG': 'C', 'LC_ALL': 'C'})
 
-        env['LANGUAGE'] = 'C'
-        env['LANG'] = 'C'
-        env['LC_ALL'] = 'C'
-        out = subprocess.Popen(
-            cmd, stdout=subprocess.PIPE, env=env).communicate()[0]
+        out = subprocess.Popen(cmd, stdout=subprocess.PIPE, env=env).communicate()[0]
         return out
 
     subst_list = {
@@ -31,24 +24,21 @@ def app_version():
     }
 
     if os.system("command -v git > /dev/null 2>&1") != 0:
-        subst_list
-    else:
-        if call(["git", "branch"], stderr=STDOUT,
-                stdout=open(os.devnull, 'w')) != 0:
-            subst_list
-        else:
-            # version
-            describe = minimal_env_cmd(["git", "describe", "--always"])
-            git_revision = describe.strip().decode('ascii')
-            # branch
-            branch = minimal_env_cmd(["git", "branch"])
-            git_branch = branch.strip().decode('ascii').replace('* ', '')
+        return subst_list
 
-            subst_list = {
-                "version": __version__,
-                "branch": git_branch,
-                "commit": git_revision
-            }
+    if call(["git", "branch"], stderr=STDOUT, stdout=open(os.devnull, 'w')) != 0:
+        return subst_list
+
+    describe = minimal_env_cmd(["git", "describe", "--tags", "--always"])
+    git_revision = describe.strip().decode('ascii')
+
+    branch = minimal_env_cmd(["git", "branch"])
+    git_branch = branch.strip().decode('ascii').replace('* ', '')
+
+    subst_list.update({
+        "branch": git_branch,
+        "commit": git_revision
+    })
 
     return subst_list
 

youtube/i18n_strings.py

@@ -0,0 +1,112 @@
+#!/usr/bin/env python3
+"""
+Centralized i18n strings for yt-local
+
+This file contains static strings that need to be translated but are used
+dynamically in templates or generated content. By importing this module,
+these strings get extracted by babel for translation.
+"""
+
+from flask_babel import lazy_gettext as _l
+
+# Settings categories
+CATEGORY_NETWORK = _l('Network')
+CATEGORY_PLAYBACK = _l('Playback')
+CATEGORY_INTERFACE = _l('Interface')
+
+# Common setting labels
+ROUTE_TOR = _l('Route Tor')
+DEFAULT_SUBTITLES_MODE = _l('Default subtitles mode')
+AV1_CODEC_RANKING = _l('AV1 Codec Ranking')
+VP8_VP9_CODEC_RANKING = _l('VP8/VP9 Codec Ranking')
+H264_CODEC_RANKING = _l('H.264 Codec Ranking')
+USE_INTEGRATED_SOURCES = _l('Use integrated sources')
+ROUTE_IMAGES = _l('Route images')
+ENABLE_COMMENTS_JS = _l('Enable comments.js')
+ENABLE_SPONSORBLOCK = _l('Enable SponsorBlock')
+ENABLE_EMBED_PAGE = _l('Enable embed page')
+
+# Setting names (auto-generated from setting keys)
+RELATED_VIDEOS_MODE = _l('Related videos mode')
+COMMENTS_MODE = _l('Comments mode')
+ENABLE_COMMENT_AVATARS = _l('Enable comment avatars')
+DEFAULT_COMMENT_SORTING = _l('Default comment sorting')
+THEATER_MODE = _l('Theater mode')
+AUTOPLAY_VIDEOS = _l('Autoplay videos')
+DEFAULT_RESOLUTION = _l('Default resolution')
+USE_VIDEO_PLAYER = _l('Use video player')
+USE_VIDEO_DOWNLOAD = _l('Use video download')
+PROXY_IMAGES = _l('Proxy images')
+THEME = _l('Theme')
+FONT = _l('Font')
+LANGUAGE = _l('Language')
+EMBED_PAGE_MODE = _l('Embed page mode')
+
+# Common option values
+OFF = _l('Off')
+ON = _l('On')
+DISABLED = _l('Disabled')
+ENABLED = _l('Enabled')
+ALWAYS_SHOWN = _l('Always shown')
+SHOWN_BY_CLICKING_BUTTON = _l('Shown by clicking button')
+NATIVE = _l('Native')
+NATIVE_WITH_HOTKEYS = _l('Native with hotkeys')
+PLYR = _l('Plyr')
+
+# Theme options
+LIGHT = _l('Light')
+GRAY = _l('Gray')
+DARK = _l('Dark')
+
+# Font options
+BROWSER_DEFAULT = _l('Browser default')
+LIBERATION_SERIF = _l('Liberation Serif')
+ARIAL = _l('Arial')
+VERDANA = _l('Verdana')
+TAHOMA = _l('Tahoma')
+
+# Search and filter options
+SORT_BY = _l('Sort by')
+RELEVANCE = _l('Relevance')
+UPLOAD_DATE = _l('Upload date')
+VIEW_COUNT = _l('View count')
+RATING = _l('Rating')
+
+# Time filters
+ANY = _l('Any')
+LAST_HOUR = _l('Last hour')
+TODAY = _l('Today')
+THIS_WEEK = _l('This week')
+THIS_MONTH = _l('This month')
+THIS_YEAR = _l('This year')
+
+# Content types
+TYPE = _l('Type')
+VIDEO = _l('Video')
+CHANNEL = _l('Channel')
+PLAYLIST = _l('Playlist')
+MOVIE = _l('Movie')
+SHOW = _l('Show')
+
+# Duration filters
+DURATION = _l('Duration')
+SHORT_DURATION = _l('Short (< 4 minutes)')
+LONG_DURATION = _l('Long (> 20 minutes)')
+
+# Actions
+SEARCH = _l('Search')
+DOWNLOAD = _l('Download')
+SUBSCRIBE = _l('Subscribe')
+UNSUBSCRIBE = _l('Unsubscribe')
+IMPORT = _l('Import')
+EXPORT = _l('Export')
+SAVE = _l('Save')
+CHECK = _l('Check')
+MUTE = _l('Mute')
+UNMUTE = _l('Unmute')
+
+# Common UI elements
+OPTIONS = _l('Options')
+SETTINGS = _l('Settings')
+ERROR = _l('Error')
+LOADING = _l('loading...')

youtube/playlist.py

@@ -8,7 +8,7 @@ import json
 import string
 import gevent
 import math
-from flask import request
+from flask import request, abort
 import flask
 
 
@@ -107,7 +107,7 @@ def get_playlist_page():
         util.prefix_urls(item)
         util.add_extra_html_info(item)
         if 'id' in item:
-            item['thumbnail'] = f"{settings.img_prefix}https://i.ytimg.com/vi/{item['id']}/hqdefault.jpg"
+            item['thumbnail'] = f"{settings.img_prefix}https://i.ytimg.com/vi/{item['id']}/hq720.jpg"
 
         item['url'] += '&list=' + playlist_id
         if item['index']:
@@ -115,7 +115,7 @@ def get_playlist_page():
 
     video_count = yt_data_extract.deep_get(info, 'metadata', 'video_count')
     if video_count is None:
-        video_count = 40
+        video_count = 1000
 
     return flask.render_template(
         'playlist.html',

youtube/proto.py

@@ -113,12 +113,12 @@ def read_protobuf(data):
             length = read_varint(data)
             value = data.read(length)
         elif wire_type == 3:
-            end_bytes = encode_varint((field_number << 3) | 4)
+            end_bytes = varint_encode((field_number << 3) | 4)
             value = read_group(data, end_bytes)
         elif wire_type == 5:
             value = data.read(4)
         else:
-            raise Exception("Unknown wire type: " + str(wire_type) + ", Tag: " + bytes_to_hex(succinct_encode(tag)) + ", at position " + str(data.tell()))
+            raise Exception("Unknown wire type: " + str(wire_type) + " at position " + str(data.tell()))
         yield (wire_type, field_number, value)
 
 

youtube/proto_debug.py

@@ -97,6 +97,7 @@ import re
 import time
 import json
 import os
+import traceback
 import pprint
 
 

youtube/static/channel.css

@@ -256,7 +256,8 @@ hr {
     padding-top: 6px;
     text-align: center;
     white-space: nowrap;
-    border: none;
+    border: 1px solid;
+    border-color: var(--button-border);
     border-radius: 0.2rem;
 }
 

youtube/static/dark_theme.css

@@ -1,20 +1,22 @@
 :root {
-    --background:           #212121;
+    --background:           #121113;
     --text:                 #FFFFFF;
-    --secondary-hover:      #73828c;
-    --secondary-focus:      #303030;
-    --secondary-inverse:    #FFF;
+    --secondary-hover:      #222222;
+    --secondary-focus:      #121113;
+    --secondary-inverse:    #FFFFFF;
     --primary-background:   #242424;
-    --secondary-background: #424242;
-    --thumb-background:     #757575;
+    --secondary-background: #222222;
+    --thumb-background:     #222222;
     --link:                 #00B0FF;
     --link-visited:         #40C4FF;
-    --border-bg:            #FFFFFF;
-    --buttom:               #dcdcdb;
-    --buttom-text:          #415462;
-    --button-border:        #91918c;
-    --buttom-hover:         #BBB;
-    --search-text:          #FFF;
-    --time-background:      #212121;
-    --time-text:            #FFF;
+    --border-bg:            #222222;
+    --border-bg-settings:   #000000;
+    --border-bg-license:    #000000;
+    --buttom:               #121113;
+    --buttom-text:          #FFFFFF;
+    --button-border:        #222222;
+    --buttom-hover:         #222222;
+    --search-text:          #FFFFFF;
+    --time-background:      #121113;
+    --time-text:            #FFFFFF;
 }

youtube/static/gray_theme.css

@@ -1,19 +1,21 @@
 :root {
-    --background:            #2d3743;
+    --background:            #2D3743;
     --text:                  #FFFFFF;
-    --secondary-hover:       #73828c;
+    --secondary-hover:       #73828C;
     --secondary-focus:       rgba(115, 130, 140, 0.125);
     --secondary-inverse:     #FFFFFF;
-    --primary-background:    #2d3743;
+    --primary-background:    #2D3743;
     --secondary-background:  #102027;
     --thumb-background:      #35404D;
-    --link:                  #22aaff;
-    --link-visited:          #7755ff;
+    --link:                  #22AAFF;
+    --link-visited:          #7755FF;
     --border-bg:             #FFFFFF;
-    --buttom:                #DCDCDC;
-    --buttom-text:           #415462;
-    --button-border:         #91918c;
-    --buttom-hover:          #BBBBBB;
+    --border-bg-settings:    #FFFFFF;
+    --border-bg-license:     #FFFFFF;
+    --buttom:                #2D3743;
+    --buttom-text:           #FFFFFF;
+    --button-border:         #102027;
+    --buttom-hover:          #102027;
     --search-text:           #FFFFFF;
     --time-background:       #212121;
     --time-text:             #FFFFFF;

youtube/static/js/av-merge.js

@@ -20,6 +20,29 @@
 // TODO: Call abort to cancel in-progress appends?
 
 
+// Buffer sizes for different systems
+const BUFFER_CONFIG = {
+    default: 50 * 10**6,  // 50 megabytes
+    webOS: 20 * 10**6,    // 20 megabytes WebOS (LG)
+    samsungTizen: 20 * 10**6, // 20 megabytes Samsung Tizen OS
+    androidTV: 30 * 10**6, // 30 megabytes Android TV
+    desktop: 50 * 10**6,  // 50 megabytes PC/Mac
+};
+
+function detectSystem() {
+    const userAgent = navigator.userAgent.toLowerCase();
+    if (/webos|lg browser/i.test(userAgent)) {
+        return "webOS";
+    } else if (/tizen/i.test(userAgent)) {
+        return "samsungTizen";
+    } else if (/android tv|smart-tv/i.test(userAgent)) {
+        return "androidTV";
+    } else if (/firefox|chrome|safari|edge/i.test(userAgent)) {
+        return "desktop";
+    } else {
+        return "default";
+    }
+}
 
 function AVMerge(video, srcInfo, startTime){
     this.audioSource = null;
@@ -164,6 +187,8 @@ AVMerge.prototype.printDebuggingInfo = function() {
 }
 
 function Stream(avMerge, source, startTime, avRatio) {
+    const selectedSystem = detectSystem();
+    let baseBufferTarget = BUFFER_CONFIG[selectedSystem] || BUFFER_CONFIG.default;
     this.avMerge = avMerge;
     this.video = avMerge.video;
     this.url = source['url'];
@@ -173,10 +198,11 @@ function Stream(avMerge, source, startTime, avRatio) {
     this.mimeCodec = source['mime_codec']
     this.streamType = source['acodec'] ? 'audio' : 'video';
     if (this.streamType == 'audio') {
-        this.bufferTarget = avRatio*50*10**6;
+        this.bufferTarget = avRatio * baseBufferTarget;
     } else {
-        this.bufferTarget = 50*10**6; // 50 megabytes
+        this.bufferTarget = baseBufferTarget;
     }
+    console.info(`Detected system: ${selectedSystem}. Applying bufferTarget of ${this.bufferTarget} bytes to ${this.streamType}.`);
 
     this.initRange = source['init_range'];
     this.indexRange = source['index_range'];

youtube/static/js/common.js

@@ -114,3 +114,60 @@ function copyTextToClipboard(text) {
 window.addEventListener('DOMContentLoaded', function() {
     cur_track_idx = getDefaultTranscriptTrackIdx();
 });
+
+/**
+ * Thumbnail fallback handler
+ * Tries lower quality thumbnails when higher quality fails (404)
+ * Priority: hq720.jpg -> sddefault.jpg -> hqdefault.jpg -> mqdefault.jpg -> default.jpg
+ */
+function thumbnail_fallback(img) {
+    const src = img.src || img.dataset.src;
+    if (!src) return;
+
+    // Handle YouTube video thumbnails
+    if (src.includes('/i.ytimg.com/')) {
+        // Extract video ID from URL
+        const match = src.match(/\/vi\/([^/]+)/);
+        if (!match) return;
+
+        const videoId = match[1];
+        const imgPrefix = settings_img_prefix || '';
+
+        // Define fallback order (from highest to lowest quality)
+        const fallbacks = [
+            'hq720.jpg',
+            'sddefault.jpg',
+            'hqdefault.jpg',
+            'mqdefault.jpg',
+            'default.jpg'
+        ];
+
+        // Find current quality and try next fallback
+        for (let i = 0; i < fallbacks.length; i++) {
+            if (src.includes(fallbacks[i])) {
+                // Try next quality
+                if (i < fallbacks.length - 1) {
+                    const newSrc = imgPrefix + 'https://i.ytimg.com/vi/' + videoId + '/' + fallbacks[i + 1];
+                    if (img.dataset.src) {
+                        img.dataset.src = newSrc;
+                    } else {
+                        img.src = newSrc;
+                    }
+                }
+                break;
+            }
+        }
+    }
+    // Handle YouTube channel avatars (ggpht.com)
+    else if (src.includes('ggpht.com') || src.includes('yt3.ggpht.com')) {
+        // Try to increase avatar size (s88 -> s240)
+        const newSrc = src.replace(/=s\d+-c-k/, '=s240-c-k-c0x00ffffff-no-rj');
+        if (newSrc !== src) {
+            if (img.dataset.src) {
+                img.dataset.src = newSrc;
+            } else {
+                img.src = newSrc;
+            }
+        }
+    }
+}

youtube/static/js/plyr-start.js

@@ -58,7 +58,7 @@
     },
   });
 
-  const player = new Plyr(document.getElementById('js-video-player'), {
+  const playerOptions = {
     // Learning about autoplay permission https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Permissions-Policy/autoplay#syntax
     autoplay: autoplayActive,
     disableContextMenu: false,
@@ -117,5 +117,20 @@
     tooltips: {
       controls: true,
     },
+  }
+
+  const player = new Plyr(document.getElementById('js-video-player'), playerOptions);
+
+  // disable double click to fullscreen
+  // https://github.com/sampotts/plyr/issues/1370#issuecomment-528966795
+  player.eventListeners.forEach(function(eventListener) {
+    if(eventListener.type === 'dblclick') {
+        eventListener.element.removeEventListener(eventListener.type, eventListener.callback, eventListener.options);
+    }
   });
+
+  // Add .started property, true after the playback has been started
+  // Needed so controls won't be hidden before playback has started
+  player.started = false;
+  player.once('playing', function(){this.started = true});
 })();

youtube/static/js/watch.js

@@ -5,8 +5,9 @@ function changeQuality(selection) {
     let videoPaused = video.paused;
     let videoSpeed = video.playbackRate;
     let srcInfo;
-    if (avMerge)
+    if (avMerge && typeof avMerge.close === 'function') {
         avMerge.close();
+    }
     if (selection.type == 'uni'){
         srcInfo = data['uni_sources'][selection.index];
         video.src = srcInfo.url;

youtube/static/license.css

@@ -181,7 +181,7 @@ label[for=options-toggle-cbox] {
 
 .table td,.table th {
     padding: 10px 10px;
-    border: 1px solid var(--secondary-background);
+    border: 1px solid var(--border-bg-license);
     text-align: center;
 }
 

youtube/static/light_theme.css

@@ -10,9 +10,11 @@
     --link:                 #212121;
     --link-visited:         #808080;
     --border-bg:            #212121;
-    --buttom:               #DCDCDC;
+    --border-bg-settings:   #91918C;
+    --border-bg-license:    #91918C;
+    --buttom:               #FFFFFF;
     --buttom-text:          #212121;
-    --button-border:        #91918c;
+    --button-border:        #91918C;
     --buttom-hover:         #BBBBBB;
     --search-text:          #212121;
     --time-background:      #212121;

youtube/static/modules/plyr/custom_plyr.css

@@ -37,3 +37,41 @@ e.g. Firefox playback speed options */
     max-height: 320px;
     overflow-y: auto;
 }
+
+/*
+* Custom styles similar to youtube
+*/
+.plyr__controls {
+    display: flex;
+    justify-content: center;
+}
+
+.plyr__progress__container {
+    position: absolute;
+    bottom: 0;
+    width: 100%;
+    margin-bottom: -10px;
+}
+
+.plyr__controls .plyr__controls__item:first-child {
+    margin-left: 0;
+    margin-right: 0;
+    z-index: 5;
+}
+
+.plyr__controls .plyr__controls__item.plyr__volume {
+    margin-left: auto;
+}
+
+.plyr__controls .plyr__controls__item.plyr__progress__container {
+    padding-left: 10px;
+    padding-right: 10px;
+}
+
+.plyr__progress input[type="range"] {
+    margin-bottom: 50px;
+}
+
+/*
+* End custom styles
+*/

youtube/static/settings.css

@@ -155,7 +155,7 @@ label[for=options-toggle-cbox] {
 }
 
 .settings-form > h2 {
-    border-bottom: 2px solid var(--border-bg);
+    border-bottom: 2px solid var(--border-bg-settings);
     padding-bottom: 0.5rem;
 }
 

youtube/static/watch.css

@@ -128,6 +128,29 @@ header {
     background-color: var(--buttom-hover);
 }
 
+.live-url-choices {
+    background-color: var(--thumb-background);
+    margin: 1rem 0;
+    padding: 1rem;
+}
+
+.playability-error {
+    position: relative;
+    box-sizing: border-box;
+    height: 30vh;
+    margin: 1rem 0;
+}
+
+.playability-error > span {
+    display: flex;
+    background-color: var(--thumb-background);
+    height: 100%;
+    object-fit: cover;
+    justify-content: center;
+    align-items: center;
+    text-align: center;
+}
+
 .playlist {
     display: grid;
     grid-gap: 4px;
@@ -622,6 +645,9 @@ figure.sc-video {
         max-height: 80vh;
         overflow-y: scroll;
     }
+    .playability-error {
+        height: 60vh;
+    }
     .playlist {
         display: grid;
         grid-gap: 1px;

youtube/subscriptions.py

@@ -1089,12 +1089,12 @@ def serve_subscription_thumbnail(thumbnail):
             f.close()
             return flask.Response(image, mimetype='image/jpeg')
 
-    url = f"https://i.ytimg.com/vi/{video_id}/hqdefault.jpg"
+    url = f"https://i.ytimg.com/vi/{video_id}/hq720.jpg"
     try:
         image = util.fetch_url(url, report_text="Saved thumbnail: " + video_id)
     except urllib.error.HTTPError as e:
         print("Failed to download thumbnail for " + video_id + ": " + str(e))
-        abort(e.code)
+        flask.abort(e.code)
     try:
         f = open(thumbnail_path, 'wb')
     except FileNotFoundError:

youtube/templates/base.html

@@ -26,6 +26,12 @@
              // @license-end
             </script>
         {% endif %}
+        <script>
+            // @license magnet:?xt=urn:btih:0b31508aeb0634b347b8270c7bee4d411b5d4109&dn=agpl-3.0.txt AGPL-v3-or-Later
+            // Image prefix for thumbnails
+            let settings_img_prefix = "{{ settings.img_prefix or '' }}";
+            // @license-end
+        </script>
     </head>
 
     <body>
@@ -35,57 +41,57 @@
             </nav>
             <form class="form" id="site-search" action="/youtube.com/results">
                 <input type="search" name="search_query" class="search-box" value="{{ search_box_value }}"
-                       {{ "autofocus" if (request.path in ("/", "/results") or error_message) else "" }} required placeholder="Type to search...">
-                <button type="submit" value="Search" class="search-button">Search</button>
+                       {{ "autofocus" if (request.path in ("/", "/results") or error_message) else "" }} required placeholder="{{ _('Type to search...') }}">
+                <button type="submit" value="Search" class="search-button">{{ _('Search') }}</button>
                 <!-- options -->
                 <div class="dropdown">
                     <!-- hidden box -->
                     <input id="options-toggle-cbox" class="opt-box" type="checkbox">
                     <!-- end hidden box -->
-                    <label class="dropdown-label" for="options-toggle-cbox">Options</label>
+                    <label class="dropdown-label" for="options-toggle-cbox">{{ _('Options') }}</label>
                     <div class="dropdown-content">
-                        <h3>Sort by</h3>
+                        <h3>{{ _('Sort by') }}</h3>
                         <div class="option">
                             <input type="radio" id="sort_relevance" name="sort" value="0">
-                            <label for="sort_relevance">Relevance</label>
+                            <label for="sort_relevance">{{ _('Relevance') }}</label>
                         </div>
                         <div class="option">
                             <input type="radio" id="sort_upload_date" name="sort" value="2">
-                            <label for="sort_upload_date">Upload date</label>
+                            <label for="sort_upload_date">{{ _('Upload date') }}</label>
                         </div>
                         <div class="option">
                             <input type="radio" id="sort_view_count" name="sort" value="3">
-                            <label for="sort_view_count">View count</label>
+                            <label for="sort_view_count">{{ _('View count') }}</label>
                         </div>
                         <div class="option">
                             <input type="radio" id="sort_rating" name="sort" value="1">
-                            <label for="sort_rating">Rating</label>
+                            <label for="sort_rating">{{ _('Rating') }}</label>
                         </div>
 
-                        <h3>Upload date</h3>
+                        <h3>{{ _('Upload date') }}</h3>
                         <div class="option">
                             <input type="radio" id="time_any" name="time" value="0">
-                            <label for="time_any">Any</label>
+                            <label for="time_any">{{ _('Any') }}</label>
                         </div>
                         <div class="option">
                             <input type="radio" id="time_last_hour" name="time" value="1">
-                            <label for="time_last_hour">Last hour</label>
+                            <label for="time_last_hour">{{ _('Last hour') }}</label>
                         </div>
                         <div class="option">
                             <input type="radio" id="time_today" name="time" value="2">
-                            <label for="time_today">Today</label>
+                            <label for="time_today">{{ _('Today') }}</label>
                         </div>
                         <div class="option">
                             <input type="radio" id="time_this_week" name="time" value="3">
-                            <label for="time_this_week">This week</label>
+                            <label for="time_this_week">{{ _('This week') }}</label>
                         </div>
                         <div class="option">
                             <input type="radio" id="time_this_month" name="time" value="4">
-                            <label for="time_this_month">This month</label>
+                            <label for="time_this_month">{{ _('This month') }}</label>
                         </div>
                         <div class="option">
                             <input type="radio" id="time_this_year" name="time" value="5">
-                            <label for="time_this_year">This year</label>
+                            <label for="time_this_year">{{ _('This year') }}</label>
                         </div>
 
                         <h3>Type</h3>

youtube/templates/channel.html

@@ -81,10 +81,10 @@
         <!-- new-->
         <div id="links-metadata">
                 {% if current_tab in ('videos', 'shorts', 'streams') %}
-                    {% set sorts = [('1', 'views'), ('2', 'oldest'), ('3', 'newest'), ('4', 'newest - no shorts'),] %}
+                    {% set sorts = [('3', 'newest'), ('4', 'newest - no shorts')] %}
                     <div id="number-of-results">{{ number_of_videos }} videos</div>
                 {% elif current_tab == 'playlists' %}
-                    {% set sorts = [('2', 'oldest'), ('3', 'newest'), ('4', 'last video added')] %}
+                    {% set sorts = [('3', 'newest'), ('4', 'last video added')] %}
                     {% if items %}
                         <h2 class="page-number">Page {{ page_number }}</h2>
                     {% else %}

youtube/templates/common_elements.html

@@ -23,11 +23,11 @@
                 <a class="thumbnail-box" href="{{ info['url'] }}" title="{{ info['title'] }}">
                     <div class="thumbnail {% if info['type'] == 'channel' %} channel {% endif %}">
                         {% if lazy_load %}
-                            <img class="thumbnail-img lazy" alt="&#x20;" data-src="{{ info['thumbnail'] }}">
+                            <img class="thumbnail-img lazy" alt="&#x20;" data-src="{{ info['thumbnail'] }}" onerror="thumbnail_fallback(this)">
                         {% elif info['type'] == 'channel' %}
-                            <img class="thumbnail-img channel" alt="&#x20;" src="{{ info['thumbnail'] }}">
+                            <img class="thumbnail-img channel" alt="&#x20;" src="{{ info['thumbnail'] }}" onerror="thumbnail_fallback(this)">
                         {% else %}
-                            <img class="thumbnail-img" alt="&#x20;" src="{{ info['thumbnail'] }}">
+                            <img class="thumbnail-img" alt="&#x20;" src="{{ info['thumbnail'] }}" onerror="thumbnail_fallback(this)">
                         {% endif %}
 
                         {% if info['type'] != 'channel' %}

youtube/templates/settings.html

@@ -31,11 +31,19 @@
                                     <input type="number" id="{{ 'setting_' + setting_name }}" name="{{ setting_name }}" value="{{ value }}" step="1">
                                 {% endif %}
                             {% elif setting_info['type'].__name__ == 'float' %}
-
+                                <input type="number" id="{{ 'setting_' + setting_name }}" name="{{ setting_name }}" value="{{ value }}" step="0.01">
                             {% elif setting_info['type'].__name__ == 'str' %}
-                                <input type="text" id="{{ 'setting_' + setting_name }}" name="{{ setting_name }}" value="{{ value }}">
+                                {% if 'options' is in(setting_info) %}
+                                    <select id="{{ 'setting_' + setting_name }}" name="{{ setting_name }}">
+                                        {% for option in setting_info['options'] %}
+                                            <option value="{{ option[0] }}" {{ 'selected' if option[0] == value else '' }}>{{ option[1] }}</option>
+                                        {% endfor %}
+                                    </select>
+                                {% else %}
+                                    <input type="text" id="{{ 'setting_' + setting_name }}" name="{{ setting_name }}" value="{{ value }}">
+                                {% endif %}
                             {% else %}
-                                <span>Error: Unknown setting type: setting_info['type'].__name__</span>
+                                <span>Error: Unknown setting type: {{ setting_info['type'].__name__ }}</span>
                             {% endif %}
                         </li>
                     {% endif %}

youtube/templates/watch.html

@@ -85,6 +85,16 @@
                             <option value='{"type": "pair", "index": {{ loop.index0}}}' {{ 'selected' if loop.index0 == pair_idx and using_pair_sources else '' }} >{{ src_pair['quality_string'] }}</option>
                         {% endfor %}
                     </select>
+
+                    {% if audio_tracks and audio_tracks|length > 1 %}
+                    <select id="audio-language-select" autocomplete="off" title="Audio language">
+                        {% for track in audio_tracks %}
+                            <option value="{{ track.get('track_id', track['language']) }}" {{ 'selected' if loop.index0 == 0 else '' }}>
+                                🔊 {{ track['language_name'] }}{% if track.get('is_default') %} (Default){% endif %}
+                            </option>
+                        {% endfor %}
+                    </select>
+                    {% endif %}
                 {% endif %}
             </div>
             <input class="v-checkbox" name="video_info_list" value="{{ video_info }}" form="playlist-edit" type="checkbox">
@@ -246,6 +256,38 @@
      let storyboard_url = {{ storyboard_url | tojson }};
      // @license-end
     </script>
+
+    <!-- Audio language selector handler -->
+    <script>
+     // @license magnet:?xt=urn:btih:0b31508aeb0634b347b8270c7bee4d411b5d4109&dn=agpl-3.0.txt AGPL-v3-or-Later
+     (function() {
+         'use strict';
+         const audioSelect = document.getElementById('audio-language-select');
+         const qualitySelect = document.getElementById('quality-select');
+
+         if (audioSelect && qualitySelect) {
+             audioSelect.addEventListener('change', function() {
+                 const selectedAudio = this.value;
+                 const selectedQuality = qualitySelect.value;
+
+                 // Parse current quality selection
+                 let qualityData;
+                 try {
+                     qualityData = JSON.parse(selectedQuality);
+                 } catch(e) {
+                     return;
+                 }
+
+                 // Reload video with new audio language
+                 const currentUrl = new URL(window.location.href);
+                 currentUrl.searchParams.set('audio_lang', selectedAudio);
+                 window.location.href = currentUrl.toString();
+             });
+         }
+     }());
+     // @license-end
+    </script>
+
     <script src="/youtube.com/static/js/common.js"></script>
     <script src="/youtube.com/static/js/transcript-table.js"></script>
     {% if settings.use_video_player == 2 %}

youtube/util.py

@@ -1,4 +1,5 @@
 from datetime import datetime
+import logging
 import settings
 import socks
 import sockshandler
@@ -18,6 +19,8 @@ import gevent.queue
 import gevent.lock
 import collections
 import stem
+
+logger = logging.getLogger(__name__)
 import stem.control
 import traceback
 
@@ -302,72 +305,144 @@ def fetch_url_response(url, headers=(), timeout=15, data=None,
 def fetch_url(url, headers=(), timeout=15, report_text=None, data=None,
               cookiejar_send=None, cookiejar_receive=None, use_tor=True,
               debug_name=None):
-    while True:
-        start_time = time.monotonic()
+    """
+    Fetch URL with exponential backoff retry logic for rate limiting.
 
-        response, cleanup_func = fetch_url_response(
-            url, headers, timeout=timeout, data=data,
-            cookiejar_send=cookiejar_send, cookiejar_receive=cookiejar_receive,
-            use_tor=use_tor)
-        response_time = time.monotonic()
+    Retries:
+    - 429 Too Many Requests: Exponential backoff (1s, 2s, 4s, 8s, 16s)
+    - 503 Service Unavailable: Exponential backoff
+    - 302 Redirect to Google Sorry: Treated as rate limit
 
-        content = response.read()
+    Max retries: 5 attempts with exponential backoff
+    """
+    import random
 
-        read_finish = time.monotonic()
+    max_retries = 5
+    base_delay = 1.0  # Base delay in seconds
 
-        cleanup_func(response)  # release_connection for urllib3
-        content = decode_content(
-            content,
-            response.getheader('Content-Encoding', default='identity'))
+    for attempt in range(max_retries):
+        try:
+            start_time = time.monotonic()
 
-        if (settings.debugging_save_responses
-                and debug_name is not None and content):
-            save_dir = os.path.join(settings.data_dir, 'debug')
-            if not os.path.exists(save_dir):
-                os.makedirs(save_dir)
+            response, cleanup_func = fetch_url_response(
+                url, headers, timeout=timeout, data=data,
+                cookiejar_send=cookiejar_send, cookiejar_receive=cookiejar_receive,
+                use_tor=use_tor)
+            response_time = time.monotonic()
 
-            with open(os.path.join(save_dir, debug_name), 'wb') as f:
-                f.write(content)
+            content = response.read()
 
-        if response.status == 429 or (
-            response.status == 302 and (response.getheader('Location') == url
-                or response.getheader('Location').startswith(
-                       'https://www.google.com/sorry/index'
-                   )
-            )
-        ):
-            print(response.status, response.reason, response.headers)
-            ip = re.search(
-                br'IP address: ((?:[\da-f]*:)+[\da-f]+|(?:\d+\.)+\d+)',
-                content)
-            ip = ip.group(1).decode('ascii') if ip else None
-            if not ip:
-                ip = re.search(r'IP=((?:\d+\.)+\d+)',
-                               response.getheader('Set-Cookie') or '')
-                ip = ip.group(1) if ip else None
+            read_finish = time.monotonic()
 
-            # don't get new identity if we're not using Tor
-            if not use_tor:
-                raise FetchError('429', reason=response.reason, ip=ip)
+            cleanup_func(response)  # release_connection for urllib3
+            content = decode_content(
+                content,
+                response.headers.get('Content-Encoding', default='identity'))
 
-            print('Error: YouTube blocked the request because the Tor exit node is overutilized. Exit node IP address: %s' % ip)
+            if (settings.debugging_save_responses
+                    and debug_name is not None
+                    and content):
+                save_dir = os.path.join(settings.data_dir, 'debug')
+                if not os.path.exists(save_dir):
+                    os.makedirs(save_dir)
 
-            # get new identity
-            error = tor_manager.new_identity(start_time)
-            if error:
-                raise FetchError(
-                    '429', reason=response.reason, ip=ip,
-                    error_message='Automatic circuit change: ' + error)
-            else:
-                continue # retry now that we have new identity
+                with open(os.path.join(save_dir, debug_name), 'wb') as f:
+                    f.write(content)
 
-        elif response.status >= 400:
-            raise FetchError(str(response.status), reason=response.reason,
-                             ip=None)
-        break
+            # Check for rate limiting (429) or redirect to Google Sorry
+            if response.status == 429 or (
+                response.status == 302 and (response.getheader('Location') == url
+                    or response.getheader('Location').startswith(
+                           'https://www.google.com/sorry/index'
+                       )
+                    )
+            ):
+                logger.info(f'Rate limit response: {response.status} {response.reason}')
+                ip = re.search(
+                    br'IP address: ((?:[\da-f]*:)+[\da-f]+|(?:\d+\.)+\d+)',
+                    content)
+                ip = ip.group(1).decode('ascii') if ip else None
+                if not ip:
+                    ip = re.search(r'IP=((?:\d+\.)+\d+)',
+                                   response.getheader('Set-Cookie') or '')
+                    ip = ip.group(1) if ip else None
+
+                # If this is the last attempt, raise error
+                if attempt >= max_retries - 1:
+                    if not use_tor or not settings.route_tor:
+                        logger.warning(f'YouTube returned 429 but Tor is not enabled. Consider enabling Tor routing.')
+                        raise FetchError('429', reason=response.reason, ip=ip)
+
+                    logger.error(f'YouTube blocked request - Tor exit node overutilized. Exit IP: {ip}')
+
+                    # get new identity
+                    error = tor_manager.new_identity(start_time)
+                    if error:
+                        raise FetchError(
+                            '429', reason=response.reason, ip=ip,
+                            error_message='Automatic circuit change: ' + error)
+                    else:
+                        continue  # retry with new identity
+
+                # Calculate delay with exponential backoff and jitter
+                delay = (base_delay * (2 ** attempt)) + random.uniform(0, 1)
+                logger.info(f'Rate limited (429). Waiting {delay:.1f}s before retry {attempt + 1}/{max_retries}...')
+                time.sleep(delay)
+                continue  # retry
+
+            # Check for client errors (400, 404) - don't retry these
+            if response.status == 400:
+                logger.error(f'Bad Request (400) - Invalid parameters or URL: {url[:100]}')
+                raise FetchError('400', reason='Bad Request - Invalid parameters or URL format', ip=None)
+
+            if response.status == 404:
+                logger.warning(f'Not Found (404): {url[:100]}')
+                raise FetchError('404', reason='Not Found', ip=None)
+
+            # Check for other server errors (503, 502, 504)
+            if response.status in (502, 503, 504):
+                if attempt >= max_retries - 1:
+                    logger.error(f'Server error {response.status} after {max_retries} retries')
+                    raise FetchError(str(response.status), reason=response.reason, ip=None)
+
+                # Exponential backoff for server errors
+                delay = (base_delay * (2 ** attempt)) + random.uniform(0, 1)
+                logger.warning(f'Server error ({response.status}). Waiting {delay:.1f}s before retry {attempt + 1}/{max_retries}...')
+                time.sleep(delay)
+                continue
+
+            # Success - break out of retry loop
+            break
+
+        except urllib3.exceptions.MaxRetryError as e:
+            # If this is the last attempt, raise the error
+            if attempt >= max_retries - 1:
+                exception_cause = e.__context__.__context__
+                if (isinstance(exception_cause, socks.ProxyConnectionError)
+                        and settings.route_tor):
+                    msg = ('Failed to connect to Tor. Check that Tor is open and '
+                           'that your internet connection is working.\n\n'
+                           + str(e))
+                    logger.error(f'Tor connection failed: {msg}')
+                    raise FetchError('502', reason='Bad Gateway',
+                                     error_message=msg)
+                elif isinstance(e.__context__,
+                                urllib3.exceptions.NewConnectionError):
+                    msg = 'Failed to establish a connection.\n\n' + str(e)
+                    logger.error(f'Connection failed: {msg}')
+                    raise FetchError(
+                        '502', reason='Bad Gateway',
+                         error_message=msg)
+                else:
+                    raise
+
+            # Wait and retry
+            delay = (base_delay * (2 ** attempt)) + random.uniform(0, 1)
+            logger.warning(f'Connection error. Waiting {delay:.1f}s before retry {attempt + 1}/{max_retries}...')
+            time.sleep(delay)
 
     if report_text:
-        print(report_text, '    Latency:', round(response_time - start_time, 3), '    Read time:', round(read_finish - response_time,3))
+        logger.info(f'{report_text} - Latency: {round(response_time - start_time, 3)}s - Read time: {round(read_finish - response_time, 3)}s')
 
     return content
 
@@ -394,23 +469,22 @@ def head(url, use_tor=False, report_text=None, max_redirects=10):
             round(time.monotonic() - start_time, 3))
     return response
 
-
-mobile_user_agent = 'Mozilla/5.0 (Linux; Android 14) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.6312.80 Mobile Safari/537.36'
+mobile_user_agent = 'Mozilla/5.0 (Linux; Android 7.0; Redmi Note 4 Build/NRD90M) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Mobile Safari/537.36'
 mobile_ua = (('User-Agent', mobile_user_agent),)
-desktop_user_agent = 'Mozilla/5.0 (Windows NT 10.0; rv:124.0) Gecko/20100101 Firefox/124.0'
+desktop_user_agent = 'Mozilla/5.0 (Windows NT 6.1; rv:52.0) Gecko/20100101 Firefox/52.0'
 desktop_ua = (('User-Agent', desktop_user_agent),)
 json_header = (('Content-Type', 'application/json'),)
 desktop_xhr_headers = (
     ('Accept', '*/*'),
     ('Accept-Language', 'en-US,en;q=0.5'),
     ('X-YouTube-Client-Name', '1'),
-    ('X-YouTube-Client-Version', '2.20240327.00.00'),
+    ('X-YouTube-Client-Version', '2.20240304.00.00'),
 ) + desktop_ua
 mobile_xhr_headers = (
     ('Accept', '*/*'),
     ('Accept-Language', 'en-US,en;q=0.5'),
-    ('X-YouTube-Client-Name', '1'),
-    ('X-YouTube-Client-Version', '2.20240328.08.00'),
+    ('X-YouTube-Client-Name', '2'),
+    ('X-YouTube-Client-Version', '2.20240304.08.00'),
 ) + mobile_ua
 
 
@@ -462,7 +536,7 @@ class RateLimitedQueue(gevent.queue.Queue):
 
 
 def download_thumbnail(save_directory, video_id):
-    url = f"https://i.ytimg.com/vi/{video_id}/hqdefault.jpg"
+    url = f"https://i.ytimg.com/vi/{video_id}/hq720.jpg"
     save_location = os.path.join(save_directory, video_id + ".jpg")
     try:
         thumbnail = fetch_url(url, report_text="Saved thumbnail: " + video_id)
@@ -502,9 +576,40 @@ def video_id(url):
     return urllib.parse.parse_qs(url_parts.query)['v'][0]
 
 
-# default, sddefault, mqdefault, hqdefault, hq720
-def get_thumbnail_url(video_id):
-    return f"{settings.img_prefix}https://i.ytimg.com/vi/{video_id}/hqdefault.jpg"
+def get_thumbnail_url(video_id, quality='hq720'):
+    """Get thumbnail URL with fallback to lower quality if needed.
+
+    Args:
+        video_id: YouTube video ID
+        quality: Preferred quality ('maxres', 'hq720', 'sd', 'hq', 'mq', 'default')
+
+    Returns:
+        Tuple of (best_available_url, quality_used)
+    """
+    # Quality priority order (highest to lowest)
+    quality_order = {
+        'maxres': ['maxresdefault.jpg', 'sddefault.jpg', 'hqdefault.jpg'],
+        'hq720': ['hq720.jpg', 'sddefault.jpg', 'hqdefault.jpg'],
+        'sd': ['sddefault.jpg', 'hqdefault.jpg'],
+        'hq': ['hqdefault.jpg', 'mqdefault.jpg'],
+        'mq': ['mqdefault.jpg', 'default.jpg'],
+        'default': ['default.jpg'],
+    }
+
+    qualities = quality_order.get(quality, quality_order['hq720'])
+    base_url = f"{settings.img_prefix}https://i.ytimg.com/vi/{video_id}/"
+
+    # For now, return the highest quality URL
+    # The browser will handle 404s gracefully with alt text
+    return base_url + qualities[0], qualities[0]
+
+
+def get_best_thumbnail_url(video_id):
+    """Get the best available thumbnail URL for a video.
+
+    Tries hq720 first (for HD videos), falls back to sddefault for SD videos.
+    """
+    return get_thumbnail_url(video_id, quality='hq720')[0]
 
 
 def seconds_to_timestamp(seconds):
@@ -538,6 +643,12 @@ def prefix_url(url):
     if url is None:
         return None
     url = url.lstrip('/')     # some urls have // before them, which has a special meaning
+
+    # Increase resolution for YouTube channel avatars
+    if url and ('ggpht.com' in url or 'yt3.ggpht.com' in url):
+        # Replace size parameter with higher resolution (s240 instead of s88)
+        url = re.sub(r'=s\d+-c-k', '=s240-c-k-c0x00ffffff-no-rj', url)
+
     return '/' + url
 
 
@@ -674,12 +785,12 @@ INNERTUBE_CLIENTS = {
                 'hl': 'en',
                 'gl': 'US',
                 'clientName': 'ANDROID',
-                'clientVersion': '19.12.36',
+                'clientVersion': '19.09.36',
                 'osName': 'Android',
-                'osVersion': '14',
-                'androidSdkVersion': 34,
+                'osVersion': '12',
+                'androidSdkVersion': 31,
                 'platform': 'MOBILE',
-                'userAgent': 'com.google.android.youtube/19.12.36 (Linux; U; Android 14; US) gzip'
+                'userAgent': 'com.google.android.youtube/19.09.36 (Linux; U; Android 12; US) gzip'
             },
             # https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-887739287
             #'thirdParty': {
@@ -690,6 +801,45 @@ INNERTUBE_CLIENTS = {
         'REQUIRE_JS_PLAYER': False,
     },
 
+    'android-test-suite': {
+        'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
+        'INNERTUBE_CONTEXT': {
+            'client': {
+                'hl': 'en',
+                'gl': 'US',
+                'clientName': 'ANDROID_TESTSUITE',
+                'clientVersion': '1.9',
+                'osName': 'Android',
+                'osVersion': '12',
+                'androidSdkVersion': 31,
+                'platform': 'MOBILE',
+                'userAgent': 'com.google.android.youtube/1.9 (Linux; U; Android 12; US) gzip'
+            },
+            # https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-887739287
+            #'thirdParty': {
+            #    'embedUrl': 'https://google.com',  # Can be any valid URL
+            #}
+        },
+        'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
+        'REQUIRE_JS_PLAYER': False,
+    },
+
+    'ios': {
+        'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
+        'INNERTUBE_CONTEXT': {
+            'client': {
+                'hl': 'en',
+                'gl': 'US',
+                'clientName': 'IOS',
+                'clientVersion': '19.09.3',
+                'deviceModel': 'iPhone14,3',
+                'userAgent': 'com.google.ios.youtube/19.09.3 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
+            }
+        },
+        'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
+        'REQUIRE_JS_PLAYER': False
+    },
+
     # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
     # See: https://github.com/zerodytrash/YouTube-Internal-Clients
     'tv_embedded': {
@@ -717,14 +867,62 @@ INNERTUBE_CLIENTS = {
         'INNERTUBE_CONTEXT': {
             'client': {
                 'clientName': 'WEB',
-                'clientVersion': '2.20240327.00.00',
+                'clientVersion': '2.20220801.00.00',
                 'userAgent': desktop_user_agent,
             }
         },
         'INNERTUBE_CONTEXT_CLIENT_NAME': 1
     },
+    'android_vr': {
+        'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
+        'INNERTUBE_CONTEXT': {
+            'client': {
+                'clientName': 'ANDROID_VR',
+                'clientVersion': '1.60.19',
+                'deviceMake': 'Oculus',
+                'deviceModel': 'Quest 3',
+                'androidSdkVersion': 32,
+                'userAgent': 'com.google.android.apps.youtube.vr.oculus/1.60.19 (Linux; U; Android 12L; eureka-user Build/SQ3A.220605.009.A1) gzip',
+                'osName': 'Android',
+                'osVersion': '12L',
+            },
+        },
+        'INNERTUBE_CONTEXT_CLIENT_NAME': 28,
+        'REQUIRE_JS_PLAYER': False,
+    },
 }
 
+def get_visitor_data():
+    visitor_data = None
+    visitor_data_cache = os.path.join(settings.data_dir, 'visitorData.txt')
+    if not os.path.exists(settings.data_dir):
+        os.makedirs(settings.data_dir)
+    if os.path.isfile(visitor_data_cache):
+        with open(visitor_data_cache, 'r') as file:
+            print('Getting visitor_data from cache')
+            visitor_data = file.read()
+        max_age = 12*3600
+        file_age = time.time() - os.path.getmtime(visitor_data_cache)
+        if file_age > max_age:
+            print('visitor_data cache is too old. Removing file...')
+            os.remove(visitor_data_cache)
+        return visitor_data
+
+    print('Fetching youtube homepage to get visitor_data')
+    yt_homepage = 'https://www.youtube.com'
+    yt_resp = fetch_url(yt_homepage, headers={'User-Agent': mobile_user_agent}, report_text='Getting youtube homepage')
+    visitor_data_re = r'''"visitorData":\s*?"(.+?)"'''
+    visitor_data_match = re.search(visitor_data_re, yt_resp.decode())
+    if visitor_data_match:
+        visitor_data = visitor_data_match.group(1)
+        print(f'Got visitor_data: {len(visitor_data)}')
+        with open(visitor_data_cache, 'w') as file:
+            print('Saving visitor_data cache...')
+            file.write(visitor_data)
+        return visitor_data
+    else:
+        print('Unable to get visitor_data value')
+    return visitor_data
 
 def call_youtube_api(client, api, data):
     client_params = INNERTUBE_CLIENTS[client]
@@ -732,12 +930,17 @@ def call_youtube_api(client, api, data):
     key = client_params['INNERTUBE_API_KEY']
     host = client_params.get('INNERTUBE_HOST') or 'www.youtube.com'
     user_agent = context['client'].get('userAgent') or mobile_user_agent
+    visitor_data = get_visitor_data()
 
     url = 'https://' + host + '/youtubei/v1/' + api + '?key=' + key
+    if visitor_data:
+        context['client'].update({'visitorData': visitor_data})
     data['context'] = context
 
     data = json.dumps(data)
     headers = (('Content-Type', 'application/json'),('User-Agent', user_agent))
+    if visitor_data:
+        headers = ( *headers, ('X-Goog-Visitor-Id', visitor_data ))
     response = fetch_url(
         url, data=data, headers=headers,
         debug_name='youtubei_' + api + '_' + client,
@@ -748,6 +951,8 @@ def call_youtube_api(client, api, data):
 
 def strip_non_ascii(string):
     ''' Returns the string without non ASCII characters'''
+    if string is None:
+        return ""
     stripped = (c for c in string if 0 < ord(c) < 127)
     return ''.join(stripped)
 

youtube/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '0.2.11'
+__version__ = 'v0.4.0'

youtube/watch.py

@@ -6,6 +6,9 @@ import settings
 
 from flask import request
 import flask
+import logging
+
+logger = logging.getLogger(__name__)
 
 import json
 import gevent
@@ -343,7 +346,6 @@ def _add_to_error(info, key, additional_message):
 def fetch_player_response(client, video_id):
     return util.call_youtube_api(client, 'player', {
         'videoId': video_id,
-        'params': 'CgIQBg',
     })
 
 
@@ -368,42 +370,42 @@ def fetch_watch_page_info(video_id, playlist_id, index):
     watch_page = watch_page.decode('utf-8')
     return yt_data_extract.extract_watch_info_from_html(watch_page)
 
+
 def extract_info(video_id, use_invidious, playlist_id=None, index=None):
+    primary_client = 'android_vr'
+    fallback_client = 'ios'
+    last_resort_client = 'tv_embedded'
+
     tasks = (
         # Get video metadata from here
         gevent.spawn(fetch_watch_page_info, video_id, playlist_id, index),
-
-        # Get video URLs by spoofing as android client because its urls don't
-        # require decryption
-        # The URLs returned with WEB for videos requiring decryption
-        # couldn't be decrypted with the base.js from the web page for some
-        # reason
-        # https://github.com/yt-dlp/yt-dlp/issues/574#issuecomment-887171136
-
-        # Update 4/26/23, these URLs will randomly start returning 403
-        # mid-playback and I'm not sure why
-        gevent.spawn(fetch_player_response, 'android', video_id)
+        gevent.spawn(fetch_player_response, primary_client, video_id)
     )
     gevent.joinall(tasks)
     util.check_gevent_exceptions(*tasks)
-    info, player_response = tasks[0].value, tasks[1].value
+
+    info = tasks[0].value or {}
+    player_response = tasks[1].value or {}
 
     yt_data_extract.update_with_new_urls(info, player_response)
 
-    # Age restricted video, retry
-    if info['age_restricted'] or info['player_urls_missing']:
-        if info['age_restricted']:
-            print('Age restricted video, retrying')
-        else:
-            print('Player urls missing, retrying')
-        player_response = fetch_player_response('tv_embedded', video_id)
+    # Fallback to 'ios' if no valid URLs are found
+    if not info.get('formats') or info.get('player_urls_missing'):
+        print(f"No URLs found in '{primary_client}', attempting with '{fallback_client}'.")
+        player_response = fetch_player_response(fallback_client, video_id) or {}
+        yt_data_extract.update_with_new_urls(info, player_response)
+
+    # Final attempt with 'tv_embedded' if there are still no URLs
+    if not info.get('formats') or info.get('player_urls_missing'):
+        print(f"No URLs found in '{fallback_client}', attempting with '{last_resort_client}'")
+        player_response = fetch_player_response(last_resort_client, video_id) or {}
         yt_data_extract.update_with_new_urls(info, player_response)
 
     # signature decryption
-    decryption_error = decrypt_signatures(info, video_id)
-    if decryption_error:
-        decryption_error = 'Error decrypting url signatures: ' + decryption_error
-        info['playability_error'] = decryption_error
+    if info.get('formats'):
+        decryption_error = decrypt_signatures(info, video_id)
+        if decryption_error:
+            info['playability_error'] = 'Error decrypting url signatures: ' + decryption_error
 
     # check if urls ready (non-live format) in former livestream
     # urls not ready if all of them have no filesize
@@ -417,21 +419,21 @@ def extract_info(video_id, use_invidious, playlist_id=None, index=None):
 
     # livestream urls
     # sometimes only the livestream urls work soon after the livestream is over
-    if (info['hls_manifest_url']
-        and (info['live'] or not info['formats'] or not info['urls_ready'])
-    ):
-        manifest = util.fetch_url(info['hls_manifest_url'],
-            debug_name='hls_manifest.m3u8',
-            report_text='Fetched hls manifest'
-        ).decode('utf-8')
-
-        info['hls_formats'], err = yt_data_extract.extract_hls_formats(manifest)
-        if not err:
-            info['playability_error'] = None
-        for fmt in info['hls_formats']:
-            fmt['video_quality'] = video_quality_string(fmt)
-    else:
-        info['hls_formats'] = []
+    info['hls_formats'] = []
+    if info.get('hls_manifest_url') and (info.get('live') or not info.get('formats') or not info['urls_ready']):
+        try:
+            manifest = util.fetch_url(info['hls_manifest_url'],
+                debug_name='hls_manifest.m3u8',
+                report_text='Fetched hls manifest'
+            ).decode('utf-8')
+            info['hls_formats'], err = yt_data_extract.extract_hls_formats(manifest)
+            if not err:
+                info['playability_error'] = None
+            for fmt in info['hls_formats']:
+                fmt['video_quality'] = video_quality_string(fmt)
+        except Exception as e:
+            print(f"Error obteniendo HLS manifest: {e}")
+            info['hls_formats'] = []
 
     # check for 403. Unnecessary for tor video routing b/c ip address is same
     info['invidious_used'] = False
@@ -626,7 +628,12 @@ def get_watch_page(video_id=None):
 
     # prefix urls, and other post-processing not handled by yt_data_extract
     for item in info['related_videos']:
-        item['thumbnail'] = "https://i.ytimg.com/vi/{}/hqdefault.jpg".format(item['id']) # set HQ relateds thumbnail videos
+        # For playlists, use first_video_id for thumbnail, not playlist id
+        if item.get('type') == 'playlist' and item.get('first_video_id'):
+            item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['first_video_id'])
+        elif item.get('type') == 'video':
+            item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['id'])
+        # For other types, keep existing thumbnail or skip
         util.prefix_urls(item)
         util.add_extra_html_info(item)
     for song in info['music_list']:
@@ -634,6 +641,9 @@ def get_watch_page(video_id=None):
     if info['playlist']:
         playlist_id = info['playlist']['id']
         for item in info['playlist']['items']:
+            # Set high quality thumbnail for playlist videos
+            if item.get('type') == 'video' and item.get('id'):
+                item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['id'])
             util.prefix_urls(item)
             util.add_extra_html_info(item)
             if playlist_id:
@@ -660,12 +670,6 @@ def get_watch_page(video_id=None):
             '/videoplayback',
             '/videoplayback/name/' + filename)
 
-    if settings.gather_googlevideo_domains:
-        with open(os.path.join(settings.data_dir, 'googlevideo-domains.txt'), 'a+', encoding='utf-8') as f:
-            url = info['formats'][0]['url']
-            subdomain = url[0:url.find(".googlevideo.com")]
-            f.write(subdomain + "\n")
-
     download_formats = []
 
     for format in (info['formats'] + info['hls_formats']):
@@ -692,6 +696,30 @@ def get_watch_page(video_id=None):
     pair_sources = source_info['pair_sources']
     uni_idx, pair_idx = source_info['uni_idx'], source_info['pair_idx']
 
+    # Extract audio tracks using yt-dlp for multi-language support
+    audio_tracks = []
+    try:
+        from youtube import ytdlp_integration
+        logger.info(f'Extracting audio tracks for video: {video_id}')
+        ytdlp_info = ytdlp_integration.extract_video_info_ytdlp(video_id)
+        audio_tracks = ytdlp_info.get('audio_tracks', [])
+
+        if audio_tracks:
+            logger.info(f'✓ Found {len(audio_tracks)} audio tracks:')
+            for i, track in enumerate(audio_tracks[:10], 1):  # Log first 10
+                logger.info(f'  [{i}] {track["language_name"]} ({track["language"]}) - '
+                          f'bitrate: {track.get("audio_bitrate", "N/A")}k, '
+                          f'codec: {track.get("acodec", "N/A")}, '
+                          f'format_id: {track.get("format_id", "N/A")}')
+            if len(audio_tracks) > 10:
+                logger.info(f'  ... and {len(audio_tracks) - 10} more')
+        else:
+            logger.warning(f'No audio tracks found for video {video_id}')
+
+    except Exception as e:
+        logger.error(f'Failed to extract audio tracks: {e}', exc_info=True)
+        audio_tracks = []
+
     pair_quality = yt_data_extract.deep_get(pair_sources, pair_idx, 'quality')
     uni_quality = yt_data_extract.deep_get(uni_sources, uni_idx, 'quality')
 
@@ -815,7 +843,9 @@ def get_watch_page(video_id=None):
             'playlist': info['playlist'],
             'related': info['related_videos'],
             'playability_error': info['playability_error'],
+            'audio_tracks': audio_tracks,
         },
+        audio_tracks = audio_tracks,
         font_family = youtube.font_choices[settings.font], # for embed page
         **source_info,
         using_pair_sources = using_pair_sources,
@@ -824,9 +854,17 @@ def get_watch_page(video_id=None):
 
 @yt_app.route('/api/<path:dummy>')
 def get_captions(dummy):
-    result = util.fetch_url('https://www.youtube.com' + request.full_path)
-    result = result.replace(b"align:start position:0%", b"")
-    return result
+    try:
+        result = util.fetch_url('https://www.youtube.com' + request.full_path)
+        result = result.replace(b"align:start position:0%", b"")
+        return result
+    except util.FetchError as e:
+        # Return empty captions gracefully instead of error page
+        logger.warning(f'Failed to fetch captions: {e}')
+        return flask.Response(b'WEBVTT\n\n', mimetype='text/vtt', status=200)
+    except Exception as e:
+        logger.error(f'Unexpected error fetching captions: {e}')
+        return flask.Response(b'WEBVTT\n\n', mimetype='text/vtt', status=200)
 
 
 times_reg = re.compile(r'^\d\d:\d\d:\d\d\.\d\d\d --> \d\d:\d\d:\d\d\.\d\d\d.*$')
@@ -891,3 +929,18 @@ def get_transcript(caption_path):
 
     return flask.Response(result.encode('utf-8'),
                           mimetype='text/plain;charset=UTF-8')
+
+
+# ============================================================================
+# yt-dlp Integration Routes
+# ============================================================================
+
+@yt_app.route('/ytl-api/video-with-audio/<video_id>')
+def proxy_video_with_audio(video_id):
+    """
+    Proxy para servir video con audio específico usando yt-dlp
+    """
+    from youtube import ytdlp_proxy
+    audio_lang = request.args.get('lang', 'en')
+    max_quality = int(request.args.get('quality', 720))
+    return ytdlp_proxy.stream_video_with_audio(video_id, audio_lang, max_quality)

youtube/yt_data_extract/common.py

@@ -226,6 +226,89 @@ def check_missing_keys(object, *key_sequences):
 
     return None
 
+
+def extract_lockup_view_model_info(item, additional_info={}):
+    """Extract info from new lockupViewModel format (YouTube 2024+)"""
+    info = {'error': None}
+
+    content_type = item.get('contentType', '')
+    content_id = item.get('contentId', '')
+
+    # Extract title from metadata
+    metadata = item.get('metadata', {})
+    lockup_metadata = metadata.get('lockupMetadataViewModel', {})
+    title_data = lockup_metadata.get('title', {})
+    info['title'] = title_data.get('content', '')
+
+    # Determine type based on contentType
+    if 'PLAYLIST' in content_type:
+        info['type'] = 'playlist'
+        info['playlist_type'] = 'playlist'
+        info['id'] = content_id
+        info['video_count'] = None
+        info['first_video_id'] = None
+
+        # Try to get video count from metadata
+        metadata_rows = lockup_metadata.get('metadata', {})
+        for row in metadata_rows.get('contentMetadataViewModel', {}).get('metadataRows', []):
+            for part in row.get('metadataParts', []):
+                text = part.get('text', {}).get('content', '')
+                if 'video' in text.lower():
+                    info['video_count'] = extract_int(text)
+    elif 'VIDEO' in content_type:
+        info['type'] = 'video'
+        info['id'] = content_id
+        info['view_count'] = None
+        info['approx_view_count'] = None
+        info['time_published'] = None
+        info['duration'] = None
+
+        # Extract duration/other info from metadata rows
+        metadata_rows = lockup_metadata.get('metadata', {})
+        for row in metadata_rows.get('contentMetadataViewModel', {}).get('metadataRows', []):
+            for part in row.get('metadataParts', []):
+                text = part.get('text', {}).get('content', '')
+                if 'view' in text.lower():
+                    info['approx_view_count'] = extract_approx_int(text)
+                elif 'ago' in text.lower():
+                    info['time_published'] = text
+    elif 'CHANNEL' in content_type:
+        info['type'] = 'channel'
+        info['id'] = content_id
+        info['approx_subscriber_count'] = None
+    else:
+        info['type'] = 'unsupported'
+        return info
+
+    # Extract thumbnail from contentImage
+    content_image = item.get('contentImage', {})
+    collection_thumb = content_image.get('collectionThumbnailViewModel', {})
+    primary_thumb = collection_thumb.get('primaryThumbnail', {})
+    thumb_vm = primary_thumb.get('thumbnailViewModel', {})
+    image_sources = thumb_vm.get('image', {}).get('sources', [])
+    if image_sources:
+        info['thumbnail'] = image_sources[0].get('url', '')
+    else:
+        info['thumbnail'] = ''
+
+    # Extract author info if available
+    info['author'] = None
+    info['author_id'] = None
+    info['author_url'] = None
+
+    # Try to get first video ID from inline player data
+    item_playback = item.get('itemPlayback', {})
+    inline_player = item_playback.get('inlinePlayerData', {})
+    on_select = inline_player.get('onSelect', {})
+    innertube_cmd = on_select.get('innertubeCommand', {})
+    watch_endpoint = innertube_cmd.get('watchEndpoint', {})
+    if watch_endpoint.get('videoId'):
+        info['first_video_id'] = watch_endpoint.get('videoId')
+
+    info.update(additional_info)
+    return info
+
+
 def extract_item_info(item, additional_info={}):
     if not item:
         return {'error': 'No item given'}
@@ -243,6 +326,10 @@ def extract_item_info(item, additional_info={}):
         info['type'] = 'unsupported'
         return info
 
+    # Handle new lockupViewModel format (YouTube 2024+)
+    if type == 'lockupViewModel':
+        return extract_lockup_view_model_info(item, additional_info)
+
     # type looks like e.g. 'compactVideoRenderer' or 'gridVideoRenderer'
     # camelCase split, https://stackoverflow.com/a/37697078
     type_parts = [s.lower() for s in re.sub(r'([A-Z][a-z]+)', r' \1', type).split()]
@@ -441,6 +528,9 @@ _item_types = {
     'channelRenderer',
     'compactChannelRenderer',
     'gridChannelRenderer',
+
+    # New viewModel format (YouTube 2024+)
+    'lockupViewModel',
 }
 
 def _traverse_browse_renderer(renderer):

youtube/yt_data_extract/everything_else.py

@@ -229,7 +229,7 @@ def extract_playlist_metadata(polymer_json):
     if metadata['first_video_id'] is None:
         metadata['thumbnail'] = None
     else:
-        metadata['thumbnail'] = f"https://i.ytimg.com/vi/{metadata['first_video_id']}/hqdefault.jpg"
+        metadata['thumbnail'] = f"https://i.ytimg.com/vi/{metadata['first_video_id']}/hq720.jpg"
 
     metadata['video_count'] = extract_int(header.get('numVideosText'))
     metadata['description'] = extract_str(header.get('descriptionText'), default='')

youtube/ytdlp_integration.py

@@ -0,0 +1,78 @@
+#!/usr/bin/env python3
+"""
+yt-dlp integration wrapper for backward compatibility.
+
+This module now uses the centralized ytdlp_service for all operations.
+"""
+import logging
+from youtube.ytdlp_service import (
+    extract_video_info,
+    get_language_name,
+    clear_cache,
+    get_cache_info,
+)
+
+logger = logging.getLogger(__name__)
+
+
+def extract_video_info_ytdlp(video_id):
+    """
+    Extract video information using yt-dlp (with caching).
+
+    This is a wrapper around ytdlp_service.extract_video_info()
+    for backward compatibility.
+
+    Args:
+        video_id: YouTube video ID
+
+    Returns:
+        Dictionary with audio_tracks, formats, title, duration
+    """
+    logger.debug(f'Extracting video info (legacy API): {video_id}')
+
+    info = extract_video_info(video_id)
+
+    # Convert to legacy format for backward compatibility
+    return {
+        'audio_tracks': info.get('audio_tracks', []),
+        'all_audio_formats': info.get('formats', []),
+        'formats': info.get('formats', []),
+        'title': info.get('title', ''),
+        'duration': info.get('duration', 0),
+        'error': info.get('error'),
+    }
+
+
+def get_audio_formats_for_language(video_id, language='en'):
+    """
+    Get available audio formats for a specific language.
+
+    Args:
+        video_id: YouTube video ID
+        language: Language code (default: 'en')
+
+    Returns:
+        List of audio format dicts
+    """
+    info = extract_video_info_ytdlp(video_id)
+
+    if 'error' in info:
+        logger.warning(f'Cannot get audio formats: {info["error"]}')
+        return []
+
+    audio_formats = []
+    for track in info.get('audio_tracks', []):
+        if track['language'] == language:
+            audio_formats.append(track)
+
+    logger.debug(f'Found {len(audio_formats)} {language} audio formats')
+    return audio_formats
+
+
+__all__ = [
+    'extract_video_info_ytdlp',
+    'get_audio_formats_for_language',
+    'get_language_name',
+    'clear_cache',
+    'get_cache_info',
+]

youtube/ytdlp_proxy.py

@@ -0,0 +1,99 @@
+#!/usr/bin/env python3
+"""
+Proxy for serving videos with specific audio using yt-dlp.
+
+This module provides streaming functionality for unified formats
+with specific audio languages.
+"""
+import logging
+from flask import Response, request, stream_with_context
+import urllib.request
+import urllib.error
+from youtube.ytdlp_service import find_best_unified_format
+
+logger = logging.getLogger(__name__)
+
+
+def stream_video_with_audio(video_id: str, audio_language: str = 'en', max_quality: int = 720):
+    """
+    Stream video with specific audio language.
+
+    Args:
+        video_id: YouTube video ID
+        audio_language: Preferred audio language (default: 'en')
+        max_quality: Maximum video height (default: 720)
+
+    Returns:
+        Flask Response with video stream, or 404 if not available
+    """
+    logger.info(f'Stream request: {video_id} | audio={audio_language} | quality={max_quality}p')
+
+    # Find best unified format
+    best_format = find_best_unified_format(video_id, audio_language, max_quality)
+
+    if not best_format:
+        logger.info(f'No suitable unified format found, returning 404 to trigger fallback')
+        return Response('No suitable unified format available', status=404)
+
+    url = best_format.get('url')
+    if not url:
+        logger.error('Format found but no URL available')
+        return Response('Format URL not available', status=500)
+
+    logger.debug(f'Streaming from: {url[:80]}...')
+
+    # Stream the video
+    try:
+        req = urllib.request.Request(url)
+        req.add_header('User-Agent', 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36')
+        req.add_header('Accept', '*/*')
+
+        # Add Range header if client requests it
+        if 'Range' in request.headers:
+            req.add_header('Range', request.headers['Range'])
+            logger.debug(f'Range request: {request.headers["Range"]}')
+
+        resp = urllib.request.urlopen(req, timeout=60)
+
+        def generate():
+            """Generator for streaming video chunks."""
+            try:
+                while True:
+                    chunk = resp.read(65536)  # 64KB chunks
+                    if not chunk:
+                        break
+                    yield chunk
+            except Exception as e:
+                logger.error(f'Stream error: {e}')
+                raise
+
+        # Build response headers
+        response_headers = {
+            'Content-Type': resp.headers.get('Content-Type', 'video/mp4'),
+            'Access-Control-Allow-Origin': '*',
+        }
+
+        # Copy important headers
+        for header in ['Content-Length', 'Content-Range', 'Accept-Ranges']:
+            if header in resp.headers:
+                response_headers[header] = resp.headers[header]
+
+        status_code = resp.getcode()
+        logger.info(f'Streaming started: {status_code}')
+
+        return Response(
+            stream_with_context(generate()),
+            status=status_code,
+            headers=response_headers,
+            direct_passthrough=True
+        )
+
+    except urllib.error.HTTPError as e:
+        logger.error(f'HTTP error streaming: {e.code} {e.reason}')
+        return Response(f'Error: {e.code} {e.reason}', status=e.code)
+    except urllib.error.URLError as e:
+        logger.error(f'URL error streaming: {e.reason}')
+        return Response(f'Network error: {e.reason}', status=502)
+    except Exception as e:
+        logger.error(f'Streaming error: {e}', exc_info=True)
+        return Response(f'Error: {e}', status=500)

youtube/ytdlp_service.py

@@ -0,0 +1,393 @@
+#!/usr/bin/env python3
+"""
+Centralized yt-dlp integration with caching, logging, and error handling.
+
+This module provides a clean interface for yt-dlp functionality:
+- Multi-language audio track extraction
+- Subtitle extraction
+- Age-restricted video support
+
+All yt-dlp usage should go through this module for consistency.
+"""
+import logging
+from functools import lru_cache
+from typing import Dict, List, Optional, Any
+import yt_dlp
+import settings
+
+logger = logging.getLogger(__name__)
+
+# Language name mapping
+LANGUAGE_NAMES = {
+    'en': 'English',
+    'es': 'Español',
+    'fr': 'Français',
+    'de': 'Deutsch',
+    'it': 'Italiano',
+    'pt': 'Português',
+    'ru': 'Русский',
+    'ja': '日本語',
+    'ko': '한국어',
+    'zh': '中文',
+    'ar': 'العربية',
+    'hi': 'हिन्दी',
+    'und': 'Unknown',
+    'zxx': 'No linguistic content',
+}
+
+
+def get_language_name(lang_code: str) -> str:
+    """Convert ISO 639-1/2 language code to readable name."""
+    if not lang_code:
+        return 'Unknown'
+    return LANGUAGE_NAMES.get(lang_code.lower(), lang_code.upper())
+
+
+def _get_ytdlp_config() -> Dict[str, Any]:
+    """Get yt-dlp configuration from settings."""
+    config = {
+        'quiet': True,
+        'no_warnings': True,
+        'extract_flat': False,
+        'format': 'best',
+        'skip_download': True,
+        'socket_timeout': 30,
+        'extractor_retries': 3,
+        'http_chunk_size': 10485760,  # 10MB
+    }
+
+    # Configure Tor proxy if enabled
+    if settings.route_tor:
+        config['proxy'] = 'socks5://127.0.0.1:9150'
+        logger.debug('Tor proxy enabled for yt-dlp')
+
+    # Use cookies if available
+    import os
+    cookies_file = 'youtube_cookies.txt'
+    if os.path.exists(cookies_file):
+        config['cookiefile'] = cookies_file
+        logger.debug('Using cookies file for yt-dlp')
+
+    return config
+
+
+@lru_cache(maxsize=128)
+def extract_video_info(video_id: str) -> Dict[str, Any]:
+    """
+    Extract video information using yt-dlp with caching.
+
+    Args:
+        video_id: YouTube video ID
+
+    Returns:
+        Dictionary with video information including audio tracks
+
+    Caching:
+        Results are cached to avoid repeated requests to YouTube.
+        Cache size is limited to prevent memory issues.
+    """
+    # Check if yt-dlp is enabled
+    if not getattr(settings, 'ytdlp_enabled', True):
+        logger.debug('yt-dlp integration is disabled')
+        return {'error': 'yt-dlp disabled', 'audio_tracks': []}
+
+    url = f'https://www.youtube.com/watch?v={video_id}'
+    ydl_opts = _get_ytdlp_config()
+
+    try:
+        logger.debug(f'Extracting video info: {video_id}')
+
+        with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+            info = ydl.extract_info(url, download=False)
+
+            if not info:
+                logger.warning(f'No info returned for video: {video_id}')
+                return {'error': 'No info returned', 'audio_tracks': []}
+
+            logger.info(f'Extracted {len(info.get("formats", []))} total formats')
+
+            # Extract audio tracks grouped by language
+            audio_tracks = _extract_audio_tracks(info)
+
+            return {
+                'video_id': video_id,
+                'title': info.get('title', ''),
+                'duration': info.get('duration', 0),
+                'audio_tracks': audio_tracks,
+                'formats': info.get('formats', []),
+                'subtitles': info.get('subtitles', {}),
+                'automatic_captions': info.get('automatic_captions', {}),
+            }
+
+    except yt_dlp.utils.DownloadError as e:
+        logger.error(f'yt-dlp download error for {video_id}: {e}')
+        return {'error': str(e), 'audio_tracks': []}
+    except Exception as e:
+        logger.error(f'yt-dlp extraction error for {video_id}: {e}', exc_info=True)
+        return {'error': str(e), 'audio_tracks': []}
+
+
+def _extract_audio_tracks(info: Dict[str, Any]) -> List[Dict[str, Any]]:
+    """
+    Extract audio tracks from video info, grouped by language.
+
+    Returns a list of unique audio tracks (one per language),
+    keeping the highest quality for each language.
+    """
+    audio_by_language = {}
+    all_formats = info.get('formats', [])
+
+    logger.debug(f'Processing {len(all_formats)} formats to extract audio tracks')
+
+    for fmt in all_formats:
+        # Only audio-only formats
+        has_audio = fmt.get('acodec') and fmt.get('acodec') != 'none'
+        has_video = fmt.get('vcodec') and fmt.get('vcodec') != 'none'
+
+        if not has_audio or has_video:
+            continue
+
+        # Extract language information
+        lang = (
+            fmt.get('language') or
+            fmt.get('audio_language') or
+            fmt.get('lang') or
+            'und'
+        )
+
+        # Get language name
+        lang_name = (
+            fmt.get('language_name') or
+            fmt.get('lang_name') or
+            get_language_name(lang)
+        )
+
+        # Get bitrate
+        bitrate = fmt.get('abr') or fmt.get('tbr') or 0
+
+        # Create track info
+        track_info = {
+            'language': lang,
+            'language_name': lang_name,
+            'format_id': str(fmt.get('format_id', '')),
+            'itag': str(fmt.get('format_id', '')),
+            'ext': fmt.get('ext'),
+            'acodec': fmt.get('acodec'),
+            'audio_bitrate': int(bitrate) if bitrate else 0,
+            'audio_sample_rate': fmt.get('asr'),
+            'url': fmt.get('url'),
+            'filesize': fmt.get('filesize'),
+        }
+
+        # Keep best quality per language
+        lang_key = lang.lower()
+        if lang_key not in audio_by_language:
+            audio_by_language[lang_key] = track_info
+            logger.debug(f'  Added {lang} ({lang_name}) - {bitrate}k')
+        else:
+            current_bitrate = audio_by_language[lang_key].get('audio_bitrate', 0)
+            if bitrate > current_bitrate:
+                logger.debug(f'  Updated {lang} ({lang_name}): {current_bitrate}k → {bitrate}k')
+                audio_by_language[lang_key] = track_info
+
+    # Convert to list and sort
+    audio_tracks = list(audio_by_language.values())
+
+    # Sort: English first, then by bitrate (descending)
+    audio_tracks.sort(
+        key=lambda x: (
+            0 if x['language'] == 'en' else 1,
+            -x.get('audio_bitrate', 0)
+        )
+    )
+
+    logger.info(f'Extracted {len(audio_tracks)} unique audio languages')
+    for track in audio_tracks[:5]:  # Log first 5
+        logger.info(f'  → {track["language_name"]} ({track["language"]}): {track["audio_bitrate"]}k')
+
+    return audio_tracks
+
+
+def get_subtitle_url(video_id: str, lang: str = 'en') -> Optional[str]:
+    """
+    Get subtitle URL for a specific language.
+
+    Args:
+        video_id: YouTube video ID
+        lang: Language code (default: 'en')
+
+    Returns:
+        URL to subtitle file, or None if not available
+    """
+    info = extract_video_info(video_id)
+
+    if 'error' in info:
+        logger.warning(f'Cannot get subtitles: {info["error"]}')
+        return None
+
+    # Try manual subtitles first
+    subtitles = info.get('subtitles', {})
+    if lang in subtitles:
+        for sub in subtitles[lang]:
+            if sub.get('ext') == 'vtt':
+                logger.debug(f'Found manual {lang} subtitle')
+                return sub.get('url')
+
+    # Try automatic captions
+    auto_captions = info.get('automatic_captions', {})
+    if lang in auto_captions:
+        for sub in auto_captions[lang]:
+            if sub.get('ext') == 'vtt':
+                logger.debug(f'Found automatic {lang} subtitle')
+                return sub.get('url')
+
+    logger.debug(f'No {lang} subtitle found')
+    return None
+
+
+def find_best_unified_format(
+    video_id: str,
+    audio_language: str = 'en',
+    max_quality: int = 720
+) -> Optional[Dict[str, Any]]:
+    """
+    Find best unified (video+audio) format for specific language and quality.
+
+    Args:
+        video_id: YouTube video ID
+        audio_language: Preferred audio language
+        max_quality: Maximum video height (e.g., 720, 1080)
+
+    Returns:
+        Format dict if found, None otherwise
+    """
+    info = extract_video_info(video_id)
+
+    if 'error' in info or not info.get('formats'):
+        return None
+
+    # Quality thresholds (minimum acceptable height as % of requested)
+    thresholds = {
+        2160: 0.85,
+        1440: 0.80,
+        1080: 0.70,
+        720: 0.70,
+        480: 0.60,
+        360: 0.50,
+    }
+
+    # Get threshold for requested quality
+    threshold = 0.70
+    for q, t in thresholds.items():
+        if max_quality >= q:
+            threshold = t
+            break
+
+    min_height = int(max_quality * threshold)
+    logger.debug(f'Quality threshold: {threshold:.0%} = min {min_height}p for {max_quality}p')
+
+    candidates = []
+    audio_lang_lower = audio_language.lower()
+
+    for fmt in info['formats']:
+        # Must have both video and audio
+        has_video = fmt.get('vcodec') and fmt.get('vcodec') != 'none'
+        has_audio = fmt.get('acodec') and fmt.get('acodec') != 'none'
+
+        if not (has_video and has_audio):
+            continue
+
+        # Skip HLS/DASH formats
+        protocol = fmt.get('protocol', '')
+        format_id = str(fmt.get('format_id', ''))
+
+        if any(x in protocol.lower() for x in ['m3u8', 'hls', 'dash']):
+            continue
+        if format_id.startswith('9'):  # HLS formats
+            continue
+
+        height = fmt.get('height', 0)
+        if height < min_height:
+            continue
+
+        # Language matching
+        lang = (
+            fmt.get('language') or
+            fmt.get('audio_language') or
+            'en'
+        ).lower()
+
+        lang_match = (
+            lang == audio_lang_lower or
+            lang.startswith(audio_lang_lower[:2]) or
+            audio_lang_lower.startswith(lang[:2])
+        )
+
+        if not lang_match:
+            continue
+
+        # Calculate score
+        score = 0
+
+        # Language match bonus
+        if lang == audio_lang_lower:
+            score += 10000
+        elif lang.startswith(audio_lang_lower[:2]):
+            score += 8000
+        else:
+            score += 5000
+
+        # Quality score
+        quality_diff = abs(height - max_quality)
+        if height >= max_quality:
+            score += 3000 - quality_diff
+        else:
+            score += 2000 - quality_diff
+
+        # Protocol preference
+        if protocol in ('https', 'http'):
+            score += 500
+
+        # Format preference
+        if fmt.get('ext') == 'mp4':
+            score += 100
+
+        candidates.append({
+            'format': fmt,
+            'score': score,
+            'height': height,
+            'lang': lang,
+        })
+
+    if not candidates:
+        logger.debug(f'No unified format found for {max_quality}p + {audio_language}')
+        return None
+
+    # Sort by score and return best
+    candidates.sort(key=lambda x: x['score'], reverse=True)
+    best = candidates[0]
+
+    logger.info(
+        f'Selected unified format: {best["format"].get("format_id")} | '
+        f'{best["lang"]} | {best["height"]}p | score={best["score"]}'
+    )
+
+    return best['format']
+
+
+def clear_cache():
+    """Clear the video info cache."""
+    extract_video_info.cache_clear()
+    logger.info('yt-dlp cache cleared')
+
+
+def get_cache_info() -> Dict[str, Any]:
+    """Get cache statistics."""
+    cache_info = extract_video_info.cache_info()
+    return {
+        'hits': cache_info.hits,
+        'misses': cache_info.misses,
+        'size': cache_info.currsize,
+        'maxsize': cache_info.maxsize,
+        'hit_rate': cache_info.hits / (cache_info.hits + cache_info.misses) if (cache_info.hits + cache_info.misses) > 0 else 0,
+    }