yujunjun/LightRAG
1
0
Fork 0
mirror of https://github.com/HKUDS/LightRAG.git synced 2025-07-04 15:41:03 +00:00
Code Issues Packages Projects Releases Wiki Activity
2,680 Commits 2 Branches 39 Tags
Commit Graph

41 Commits

Author SHA1 Message Date
yangdx
0761af19c6 Files are now processed in batches in auto scan 2025-03-21 13:41:37 +08:00
yangdx
67eee2d2d5 Prioritize OS environment variables over .env file to improve Docker compatibility for the server 2025-03-21 13:27:12 +08:00
yangdx
c42f08c0e6 Add navigation service for centralized login redirect routing 
- Simplify token validation logic in API
- Update axios interceptor to use navigation service
 2025-03-18 19:45:43 +08:00
yangdx
43996656d3 Fix linting 2025-03-18 03:30:43 +08:00
yangdx
f8440c8f80 feat(auth): implement auto guest mode and enhance token system 
- Add role-based token system with metadata support
- Implement automatic guest mode for unconfigured authentication
- Create new /auth-status endpoint for authentication status checking
- Modify frontend to auto-detect auth status and bypass login when appropriate
- Add guest mode indicator in site header for better UX

This change allows users to automatically access the system without manual
login when authentication is not configured, while maintaining secure
authentication when credentials are properly set up.
 2025-03-18 02:56:02 +08:00
yangdx
c2ba7f33ff Adjust concurrency limits more LLM friendly settings for new comers 
- Lowered max async LLM processes to 4
- Enabled LLM cache for entity extraction
- Reduced max parallel insert to 2
 2025-03-16 23:56:34 +08:00
zrguo
2bf40ab636
Merge branch 'main' into neo4j-add-min-degree 2025-03-10 22:24:06 +08:00
yangdx
4065a7df92 Fix linting 2025-03-10 02:07:19 +08:00
yangdx
bc42afe7b6 Unify llm_response_cache and hashing_kv, prevent creating an independent hashing_kv. 2025-03-09 22:15:26 +08:00
Saifeddine ALOUI
04862033d6 Made the defa&ult mode non docling 2025-03-09 13:14:39 +01:00
yangdx
020a6b5ae0 Refactor LLM cache config to use argparse and add status display 2025-03-09 16:45:57 +08:00
zrguo
c67e6aceca
Merge branch 'main' into main 2025-03-09 00:23:06 +08:00
Milin
c015296081 refactor(api): Fix issues reported by pre-commit 
- Modified code layout and formatting in multiple files, improving code readability.
  - Updated import statements, removing unused libraries.
  - Simplified the writing of some functions and exception handling.
 2025-03-06 14:23:52 +08:00
Saifeddine ALOUI
00f3c6c6dd Upgraded document loading engine 2025-03-06 01:11:48 +01:00
Saifeddine ALOUI
39c24f4a59
Update utils_api.py 2025-03-05 15:36:17 +01:00
Saifeddine ALOUI
c62422eade
Update utils_api.py 2025-03-05 15:33:54 +01:00
Milin
59e3b2eec1 refactor(api): Code optimization based on review comments. 
- Removed authentication dependency for the health check endpoint in lightrag_server.py
- Removed the authentication dependency for the entire router in ollama_api.py
- Updated the parameter list and example usage in README.md, removing authentication-related parts
- Removed neo4j and tqdm dependencies from requirements.txt
- Deleted command-line argument parsing code related to authentication from utils_api.py
 2025-03-05 20:38:05 +08:00
Milin
11502fc756
Merge branch 'HKUDS:main' into feat_login-jwt 2025-03-05 20:21:21 +08:00
zrguo
06fc65d9a0 Revert "[EVO] Add language configuration to environment and argument parsing" 
This reverts commit a688b8822a8a9eb3853781bb5c71029a22aa5396.
 2025-03-05 16:26:28 +08:00
Milin
852634e20e
Merge branch 'main' into feat_login-jwt 2025-03-05 12:42:40 +08:00
Milin
63aa4f9dfc feat(api): Add user authentication functionality 
- Implement JWT-based user authentication logic
- Add login endpoint and token validation middleware
- Update API routes with authentication dependencies
- Add authentication-related environment variables
- Optimize requirements.txt with necessary dependencies
 2025-03-05 11:09:31 +08:00
Brocowlee
a688b8822a [EVO] Add language configuration to environment and argument parsing 2025-03-04 10:09:47 +00:00
yangdx
c973498c34 Fix linting 2025-02-28 21:35:04 +08:00
yangdx
c37b1e8aa7 Align Gunicorn configuration with Uvicorn 
- centralize config in gunicorn_config.py
- fix log level handling in Gunicorn
 2025-02-28 20:41:11 +08:00
yangdx
f588cdc5df Optimize logging config & worker handling for different server modes 
• Separate logging config for uvicorn/gunicorn
• Force workers=1 in uvicorn mode
• Add warning for worker count in uvicorn
 2025-02-28 16:50:54 +08:00
yangdx
7436c06f6c Fix linting 2025-02-26 18:11:16 +08:00
yangdx
2752a764ae Refactor storage implementations to support both single and multi-process modes 
• Add shared storage management module
• Support process/thread lock based on mode
 2025-02-26 05:38:38 +08:00
yangdx
7262f61b0e add redis configuration and update workers default value 2025-02-25 10:47:27 +08:00
yangdx
04fc5ce604 Remove unspported endpoint from splash mesages 2025-02-25 09:45:14 +08:00
yangdx
ddc366b672 Optimize display_splash_screen function 
- Merge System Configuration into Server Configuration section
- Add Workers parameter display after Port parameter
 2025-02-25 09:44:17 +08:00
yangdx
d74a23d2cc Add multiple workers support for API Server 2025-02-25 09:37:00 +08:00
yangdx
f5efe5977b Merge branch 'clear-text-before-insert' into simplify-cli-arguments 2025-02-23 17:06:39 +08:00
yangdx
78e0ca7835 Change defautl timeout for LLM to 150s 2025-02-23 17:03:35 +08:00
yangdx
df95f251dc Move server ready message to lifespan 2025-02-23 16:42:31 +08:00
yangdx
d84b90bcd4 Handle special case of CLI argument openai-ollama 2025-02-22 15:27:19 +08:00
yangdx
f9780830ad Revert: get llm-binding and embedding-binding from cli 2025-02-22 10:46:54 +08:00
yangdx
2fdbbc2062 Reorder and improve command line argument definitions for better organization 
- Change verbose to store_true action
- Move verbose flag to logging section
- Move auto-scan flag to end of options
 2025-02-21 20:01:43 +08:00
yangdx
a848884a7b Remove unnesessary CLI arguments, reduce CLI arguments complexity 
• Move storage config from CLI
• Move LLM and embedding binding config from CLI
• Remove chunk config from CLI
 2025-02-21 19:34:17 +08:00
yangdx
5d884f6d3e refactor: centralize configuration and utility functions 
- Move parse_args and display_splash_screen functions from lightrag_server.py to utils_api.py
- Move OllamaServerInfos class and instance from ollama_api.py to utils_api.py
 2025-02-20 12:12:20 +08:00
yangdx
a8abcf14ac Fit linting 2025-02-20 04:12:21 +08:00
yangdx
c0c87edc45 split lightrag_servery.py to smaller files 2025-02-20 03:26:39 +08:00