Releases: xr843/fojin
Release list
v1.0.0 — Initial Release
FoJin v1.0.0 — Initial Release
The world's encyclopedic Buddhist digital text platform.
Highlights
- 504 data sources aggregated from CBETA, SuttaCentral, BDRC, 84000, SAT, GRETIL, and more
- 9,200+ texts in Pali, Classical Chinese, Tibetan, Sanskrit, and 26 other languages
- 7,600+ full-text readings available online
- 8 UI languages — Chinese, English, Japanese, Korean, Thai, Vietnamese, Sinhala, Burmese
Core Features
- Multi-dimensional search — Full-text search powered by Elasticsearch with ICU tokenizer for CJK, Pali, Sanskrit, and Tibetan
- Parallel reading — Side-by-side comparison of translations across 30 languages
- AI Q&A (XiaoJin) — RAG-based question answering grounded in 38 core sutras (~11M characters), with source citations
- Knowledge graph — 9,600+ entities and 3,800+ relationships, interactive D3 visualization
- Dictionary lookup — 6 authoritative dictionaries, 237K+ entries (DDB, NCPED, Edgerton BHS, Monier-Williams, and more)
- IIIF manuscript viewer — View original manuscripts from BDRC and other collections
- Collections, bookmarks & annotations — Personal study tools for organizing research
- Citation export — BibTeX, RIS, and APA formats for academic use
Tech Stack
- Frontend: React 18 + TypeScript + Ant Design 5
- Backend: FastAPI + SQLAlchemy async
- Database: PostgreSQL 15 (pgvector) + Elasticsearch 8 + Redis 7
- Deployment: Docker Compose one-command setup
Quick Start
git clone https://github.com/xr843/fojin.git
cd fojin
cp .env.example .env
docker compose up -dVisit http://localhost:3000 to start exploring.
Links
- 🌐 Live: fojin.app
- 📖 中文文档: docs/README_zh.md
- 💬 Discord: Join the community
License
Apache License 2.0 — applies to FoJin source code only. Third-party data sources retain their own licenses.
v4.0.0 — Full-featured Buddhist Digital Text Platform
🎉 FoJin v4.0.0
A major milestone — 93 commits, 65+ PRs since v3.4.0, transforming FoJin into a comprehensive, production-ready Buddhist digital text platform.
Live at fojin.app
✨ New Features
AI Chat ("XiaoJin" 小津)
- Real SSE streaming with hybrid RAG retrieval grounded in 11M characters of canonical text (#51, #74)
- Auto-generate session titles from conversation (#62)
- Typing animation with thinking placeholder (#52)
- Paginated session messages (#69)
- Daily quota display for users (#60)
Knowledge Graph
- 9,700+ entities and 4,170+ relations across 6 types (person, text, place, concept, school, dynasty)
- Seeded 25 Buddhist places and 18 core concepts (#53)
- Auto-link texts to concepts via title keyword matching — Pass G (#54)
- pg_trgm GIN index for fast fuzzy search (#54)
- Statistics panel with entity/relation counts (#54)
- Fixed DILA Authority Database integration (#53)
Reader & Annotations
- Annotation button and panel for text reader (#72)
- Bookmark toggle button (#58)
- Mobile-optimized reading view (#19, #47)
Search
- Highlight matches in catalog search results (#64)
- Rate limiting and filter caching (#67)
- Loading skeleton cards (#44)
Data Sources — 571 Global Sources
- 30 new global sources spanning South/Southeast Asia, East Asia, Europe, Americas (#70)
- 28 more sources including AI/NLP tools, Silk Road, Vietnamese Nom, Manchu/Sogdian (#73)
- Added Dharmamitra AI toolkit (#35), 51shu.app Ashoka Index (#55), and more
Internationalization
UX
- Dark mode with Ant Design 5 theme support (#46)
- Citation export in BibTeX, RIS, and APA formats (#48)
🏗 Infrastructure & Quality
- Error handling: Custom
FoJinErrorhierarchy with 20+ specific types (#38, #42, #50) - Security: HSTS header, secrets scanning CI, Docker healthchecks (#41)
- CI/CD: Split type-check/build steps, tightened ESLint threshold (#39)
- Rate limiting: Configurable via environment variables (#43)
- Dependencies: Upgraded alembic 1.14→1.18, pydantic-settings 2.7→2.13 (#75)
- Docker: Debian-based Nginx with DNS re-resolution (#45)
📊 By the Numbers
| Metric | Value |
|---|---|
| Data sources | 571 across 30+ countries |
| Readable fascicles | 4,488 |
| KG entities | 9,700+ |
| KG relations | 4,170+ |
| Dictionary entries | 237,593 |
| Supported languages | 30 |
| PRs merged | 65+ |
🚀 Quick Start
git clone https://github.com/xr843/fojin.git
cd fojin
cp .env.example .env
docker compose up -dVisit http://localhost:3000 — API docs at http://localhost:8000/docs
Full Changelog: v3.4.0...v4.0.0
v3.4.0 — Website UX Improvements
FoJin v3.4.0 — Website UX Improvements
New Features
- Online Reading Page — New TextReaderPage for reading full-text Buddhist scriptures directly on the platform, with juan navigation and serif typography
- Homepage Feature Cards — Quick navigation to Data Sources, Knowledge Graph, AI Q&A, and Collections
- Knowledge Graph Default View — KG page now loads with "玄奘" (Xuanzang) pre-filled, showing results immediately instead of a blank page
- Chat Mobile Responsive — Sidebar hidden on mobile with drawer overlay toggle for session management
Fixes
- Fixed Japanese label "漢籍リポジトリ" → "汉籍全文资料库" in source display
- Updated data source count in README (483 → 440+) to match live API
Technical
- New route
/texts/:id/readwith lazy-loaded TextReaderPage component - Mobile-responsive CSS for chat sidebar (drawer overlay at <768px)
- Feature navigation grid with responsive 4→2 column layout
Full Changelog: v3.3.0...v3.4.0
v3.3.0 — First Open-Source Release
FoJin v3.3.0 — First Open-Source Release
We're excited to open-source FoJin, the world's most comprehensive Buddhist digital text platform.
What's Included
409 data sources from 27 countries, covering 29 languages — from Classical Chinese and Sanskrit to Pali, Tibetan, Gandhari, and more.
Core Features
- Federated Search — Search across local index + 728,000 records from DianJin (Tsinghua/CNKI)
- Full-Text Reading — 4,488 fascicles of Buddhist texts available online
- Parallel Reading — Compare translations side by side in 29 languages
- Dictionary Lookup — 6 authoritative dictionaries, 237,593 entries (DDB, NCPED, Monier-Williams, etc.)
- Knowledge Graph — 9,600+ entities, 3,800+ relationships, interactive visualization
- AI Q&A "XiaoJin" — RAG-powered assistant grounded in 11M characters of canonical text
- Manuscript Viewer — IIIF-connected viewer for BDRC and other digitized manuscripts
Technical Highlights
- FastAPI (async) + React 18 + TypeScript
- PostgreSQL 15 with pgvector + Elasticsearch 8 (ICU)
- Docker Compose one-command deployment
- Non-root containers, security headers, rate limiting
- GitHub Actions CI + automated secret scanning
Quick Start
git clone https://github.com/xr843/fojin.git
cd fojin
cp .env.example .env # edit with your API keys
docker compose up -dVisit http://localhost:3000 to start exploring.
Contributing
We welcome contributions! Whether it's adding a new data source, improving search, or translating the UI — see CONTRIBUTING.md.
If FoJin is useful for your research, please consider giving it a star!