Media
6 items
Description
A clean reading mode extension. # Greeder - Clean Reading Mode for Chrome ## Overview Greeder is a powerful Chrome extension that transforms cluttered web pages into clean, distraction-free reading experiences. With one click or keyboard shortcut (Alt+G), it extracts article content and presents it in a beautifully formatted, customizable reader view. ## Key Features ### Smart Content Extraction - Intelligent extraction engine: Combines site-specific rules with Mozilla's Readability algorithm for accurate content detection - Automatic article detection: Pre-checks pages before entering reading mode - Fallback mechanism: If one extraction method fails, automatically tries alternatives - Three-tier rule system: Bundled rules + local cache + cloud updates for optimal performance ### Customizable Reading Experience - Three beautiful themes: Light, Dark, and Sepia modes for comfortable reading in any environment - Font customization: Choose from 8+ carefully selected fonts including system fonts, serif, sans-serif, and monospaced options - Adjustable font size: 5 size presets from small to extra-large - Line spacing control: 4 spacing options for optimal readability - Flexible toolbar: Position toolbar on top, bottom, left, or right side - Fullscreen mode: Immersive reading with F11 or toolbar button ### Advanced Annotation Tools - Doodle mode: Draw freehand annotations directly on articles using Fabric.js - Multiple drawing tools: Pen, highlighter, eraser, arrows, and shapes with customizable colors and brush sizes - More shapes for callouts: Triangle, Diamond, Star, and Rounded Rectangle - Perfect strokes: Smooth, natural drawing powered by perfect-freehand library - Non-destructive: Annotations are layered on top without modifying original content ### Content Editing - Live Markdown editing: Edit extracted content in split-screen view - Real-time preview: See changes instantly as you type - Preserve formatting: Markdown syntax for headers, lists, links, code blocks, and more - Syntax highlighting: Code blocks beautifully highlighted with highlight.js ### Image Gallery - Lightbox viewer: Click any image to view in fullscreen gallery - Keyboard navigation: Arrow keys to browse, Esc to close - Smooth transitions: Elegant animations between images ### Export Options - Markdown export: Download original content as .md file - PDF export: Generate professional PDFs with jsPDF - PNG export: Capture full article or selected region as image - Crop tool: Select specific portions to export - Quick sharing: Share captured content to WhatsApp, LinkedIn, Reddit, or Email ### Smart Blacklist System - Three-tier protection: Built-in patterns + user customization + cloud updates - Automatic filtering: Prevents reading mode on web apps, dashboards, and incompatible sites - Memory feature: Remembers when you manually exit reading mode on specific pages - Domain-level blocking: Wildcards supported (e.g., *.example.com) ### Internationalization - Multi-language support: Full interface in English, Chinese (中文), Japanese (日本語), Korean (한국어), German (Deutsch), French (Français), Spanish (Español), Portuguese (Português), Russian (Русский), Arabic (العربية), Hindi (हिन्दी), Bengali (বাংলা), Indonesian (Bahasa Indonesia), Italian (Italiano), and Urdu (اردو) - Auto-detection: Matches browser language preference - Remote config: Donation info and changelog localized via backend API ### Privacy & Performance - Offline-first design: Core features work without internet connection - No tracking: Zero user data collection, no analytics - Minimal permissions: Only requests necessary permissions (activeTab, storage, alarms) - Shadow DOM isolation: Reader UI completely isolated from page styles to prevent conflicts - Per-site settings: Theme and layout preferences saved per hostname ### Developer-Friendly - Open source: Transparent codebase built with modern web technologies - Manifest V3: Future-proof Chrome extension architecture - Error reporting: Optional anonymous error reports to improve extraction accuracy - Regular updates: Cloud-based rule updates for new websites ## How It Works 1. **Activate**: Click extension icon or press Alt+G on any article page 2. **Extract**: Greeder analyzes the page and extracts main content (title, body, author, date) 3. **Render**: Content displays in clean reader view with your preferred theme and settings 4. **Customize**: Adjust fonts, sizes, themes, or add annotations 5. **Export**: Save as Markdown, PDF, or PNG when finished ## Use Cases - Research & studying: Read academic papers and long articles without distractions - Content curation: Extract and save articles for later reference - Accessibility: Improve readability with larger fonts and better contrast - Annotation: Mark up articles with highlights and notes - Offline reading: Extract content once, read without internet ## Technology Stack - UI Framework: Lit (lightweight web components) - Styling: Shadow DOM for complete isolation - Extraction: Custom rule engine + @mozilla/readability - Markdown: marked.js for parsing and rendering - Drawing: Fabric.js 7.1 + perfect-freehand - Export: jsPDF + modern-screenshot - Build: Vite with dual build system (content + background scripts) ## Privacy Policy Greeder respects your privacy: - No data collection: We don't collect, store, or transmit personal information - Local storage only: All settings stored locally on your device - Optional error reports: If extraction fails, you can optionally send anonymous error reports (URL + rule ID only) to help improve accuracy - Cloud features: Remote rule updates and config use HTTPS and don't include user-identifying data - No third-party tracking: No analytics, ads, or tracking scripts ## Permissions Explained - **activeTab**: Read page content when you activate reading mode (only on active tab) - **storage**: Save your theme preferences and per-site settings locally - **alarms**: Check for rule updates periodically (every 12 hours) - **host_permissions** (greeder.top/api): Fetch extraction rules and config updates from our backend ## Support - Telegram Channel: Join @greeder_ext for updates and support - Discord: Join our community for discussions and feedback - X (Twitter): Follow for latest news and updates - Donations: Support development via Ko-fi, PayPal, or WeChat Pay - Feedback: Report issues or request features through the extension's About page ## What's New (v1.4.2) **Security:** - Cloud extraction rules now validated before storage — new RuleValidator module enforces schema checks, static ReDoS detection (star height + alternation), selector safety, and size/count limits; historical dirty cache filtered on load and cleaned on next write **Performance:** - TurndownService singleton now shared across extraction, UI, and editor — eliminates redundant instantiation; background context menu uses Actions constant for message consistency - AutoReadingManager blacklist pattern matching now pre-compiles regex and caches storage reads — eliminates redundant regex compilation (~52 patterns) and storage I/O on every blacklist check **Bug Fixes:** - Dynamic blacklist from cloud config was silently discarded — ConfigManager sanitized it as a flat string array but AutoReadingManager expected { version, patterns } object structure; now unified to match design spec - Extension action button silently failed on pages where content script was not yet injected (e.g., pages opened before install/update) — now auto-injects content.js via chrome.scripting.executeScript and retries --- ## Previous Releases ### v1.4.1 **Performance:** - TurndownService (HTML-to-Markdown converter) promoted to a shared singleton — eliminates redundant instantiation and rule registration on every page extraction - RuleManager now caches rules in memory and pre-compiles regex patterns — eliminates redundant Map construction, Storage reads, and regex compilation on every rule lookup - ThemeManager now fetches cloud theme CSS assets in parallel (Promise.all) instead of sequentially — reduces theme update time significantly when multiple themes are available - Doodle hint element removal optimized — uses direct reference instead of traversing all Shadow DOM elements, eliminating unnecessary full-tree scanning on every text edit start/end **Doodle Improvements:** - Stroke and fill color indicators redesigned — stroke shown as outlined square, fill shown as solid square (Photoshop-style), replacing text labels for a more compact toolbar ### v1.4.0 **Performance:** - Content parsing and sanitization now cached — significantly reduces CPU usage during theme switches, font changes, menu toggles, and other UI updates on long articles **Doodle Improvements:** - Text tool now supports font size (S/M/L), bold, italic, underline, and text alignment — a formatting sub-row appears contextually when the text tool is active or a text object is selected - Shapes (triangle, rectangle, etc.) now correctly use transparent fill by default instead of black - Fixed: Switching to a drawing tool while a shape is selected now draws a new shape instead of dragging the existing one - Fixed: Shape stroke/fill colors were incorrectly rendered as black after reloading due to color format corruption - Improved: Freehand brush strokes now have smooth, round caps at the start and end instead of tapered points, for a more natural pen feel - Fixed: Doodle toolbar stroke color indicator was out of sync with the actual brush color — corrupted color values from previous sessions are now detected and reset to default on load ### v1.3.0 **New Features:** - Added more doodle shapes: Triangle, Diamond, Star, and Rounded Rectangle - Added sharing targets for WhatsApp, LinkedIn, Reddit, and Email - Improved freehand drawing with smoother pen input and better stylus/touch pressure handling **Bug Fixes & Reliability:** - Fixed race conditions during rapid reader activation to prevent duplicate initialization - Fixed concurrent per-tab icon updates to avoid state corruption across tabs - Fixed copy button animation state leakage during rapid consecutive clicks - Released temporary DOM clones after extraction to reduce memory usage - Capped exited-page history and debounced storage writes to reduce unnecessary storage growth and I/O - Fixed fill color updates for Triangle, Diamond, and Star shapes in doodle mode **Security:** - Hardened reader isolation to better protect internal UI from external scripts - Replaced dynamic `innerHTML` usage with safer DOM updates for injected content - Added sender validation for background messages to reject unknown extensions - Sanitized remote config before saving to filter invalid URLs and malformed data **Extraction & Compatibility:** - Improved InfoQ article extraction by removing the audio player, bookmark bar, and action sidebar from reader view