UniqAI
/
open-webui
mirror of https://github.com/open-webui/open-webui
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
1,017 Commits 3 Branches 61 Tags 220 MiB
Commit Graph

1017 Commits

Author SHA1 Message Date
Timothy J. Baek 220530c450 feat: convo tag filtering 2024-01-18 02:55:25 -08:00
Timothy J. Baek 1eec176313 feat: save tags to chat data 2024-01-18 02:17:31 -08:00
Timothy J. Baek 987685dbf9 feat: convo tagging full integration 2024-01-18 02:10:07 -08:00
Timothy J. Baek d5ed119687 feat: convo tagging api added 2024-01-18 01:04:24 -08:00
Timothy J. Baek 077f1fa34b feat: convo tagging backend support 2024-01-18 00:58:45 -08:00
Timothy J. Baek 287668f84e feat: tag add/remove frontend 2024-01-17 22:49:58 -08:00
Timothy J. Baek 4c69e2371e fix: wording 2024-01-17 22:49:38 -08:00
Timothy J. Baek 93ec04003a fix: share modal styling 2024-01-17 21:03:53 -08:00
Timothy J. Baek 4e828b0669 feat: convo tagging frontend 2024-01-17 21:01:30 -08:00
Timothy Jaeryang Baek 541cf36601
Merge pull request #509 from ollama-webui/stop-fix 
feat: terminate request on user stop
 2024-01-17 19:43:29 -08:00
Timothy J. Baek 981ec89e15 fix: cancel chat generation on route change 2024-01-17 19:42:40 -08:00
Timothy J. Baek b8337db9d5 Update +page.svelte 2024-01-17 19:31:55 -08:00
Timothy J. Baek 442e3d978a feat: terminate request on user stop 2024-01-17 19:19:44 -08:00
Timothy Jaeryang Baek 684bdf5151
Merge pull request #506 from ollama-webui/model-size 
feat: display model size
 2024-01-17 17:58:47 -08:00
Timothy J. Baek bb2563f1c4 feat: display model size 2024-01-17 17:56:36 -08:00
Timothy Jaeryang Baek 74da8ebf5b
Merge pull request #505 from ollama-webui/message-ts 
feat: message ts
 2024-01-17 17:47:54 -08:00
Timothy J. Baek cf88652aa8 fix: timestamp styling 2024-01-17 17:47:01 -08:00
Timothy J. Baek 4ec81a897e feat: message ts display 2024-01-17 17:43:45 -08:00
Timothy J. Baek 59724ea9d8 feat: message ts 2024-01-17 15:09:38 -08:00
Timothy Jaeryang Baek 89b35e4556
Merge pull request #504 from mslinn/fix1 
This script can now be run from any directory
 2024-01-17 15:00:24 -08:00
Timothy Jaeryang Baek 578e78cb39
Merge pull request #498 from bhulston/fix/chat-imports 
Feat: Add ChatGPT import functionality
 2024-01-17 14:50:28 -08:00
Timothy J. Baek b6ab357e8c fix: more edge cases 2024-01-17 14:47:56 -08:00
Mike Slinn a39cd633f8 Merge branch 'main' of github.com:mslinn/ollama-webui into fix1 2024-01-17 17:35:32 -05:00
Mike Slinn f9885a15e4 Can now run from any directory 2024-01-17 17:30:58 -05:00
Timothy J. Baek f40147ce58 refac: gpt renamed to openai 2024-01-17 14:23:16 -08:00
Timothy Jaeryang Baek 3cb74ab66a
Merge pull request #503 from nitrodude/fix-md-rag-ui 
Fix: ".md" documents upload only works for drag n. drop in message.
 2024-01-17 14:02:14 -08:00
Timothy Jaeryang Baek 55ba5e6258
Merge pull request #501 from oliverbob/main 
Idiot-Proof level HTTPS doc for apache
 2024-01-17 14:00:03 -08:00
i810697 7f14479567
fixing md document upload bug 2024-01-17 11:47:49 -05:00
Oliver Bob Reyes Lagumen 8e0552f735 Idiot-Proof level HTTPS doc for apache 2024-01-17 20:30:52 +08:00
Brandon Hulston 1f66bbba98 Better solution to handle edge cases 2024-01-17 02:24:17 -07:00
Brandon Hulston 81bfb97c91 Make sure model is saved in DB for imported chats (or chats with no model saved already) 2024-01-17 02:07:29 -07:00
Brandon Hulston 415777eca2 Add error catching for import 2024-01-17 01:19:27 -07:00
Brandon Hulston 27d5eb4684 Add chatGPT chat history Import functionality 2024-01-17 01:01:11 -07:00
Timothy Jaeryang Baek 5e32db1c57
Merge pull request #486 from bhulston/fix/chat-imports 
Fix/chat imports
 2024-01-16 00:36:38 -08:00
Brandon Hulston 4b3acfa491 Remove console logs 2024-01-15 18:07:31 -07:00
Brandon Hulston 3b7775a535 Fix chat import bug 2024-01-15 17:58:51 -07:00
Timothy Jaeryang Baek 26f7a1c682
Merge pull request #474 from EmelyanenkoK/patch-1 
Add Docker apt repository set up for installation instructions
 2024-01-14 14:35:42 -08:00
EmelyanenkoK 723db727d3
Add Docker apt repository set up for installation instructions 2024-01-14 12:10:40 +03:00
Timothy Jaeryang Baek bc3df5ac51
Merge pull request #465 from anuraagdjain/fix/sanitized-model 
fix: sanitized model name
 2024-01-13 21:29:27 -08:00
Anuraag Jain fc4235943d fix: sanitized model name 2024-01-13 09:49:22 +02:00
Timothy Jaeryang Baek 5c18d61d3f
Merge pull request #461 from djismgaming/optimization/reduce-image-size 
reduce docker final image size
 2024-01-12 20:00:35 -08:00
Timothy Jaeryang Baek 6d6d5b4d0d
Merge pull request #454 from anuraagdjain/feat/custom-server-port 
feat: custom port for server
 2024-01-12 19:48:36 -08:00
Timothy J. Baek 0e039492c6 refac: SERVER_PORT renamed to PORT 2024-01-12 19:38:30 -08:00
Ismael 141afb4c20 reduce image size 
Added the option `--no-cache-dir` to the `pip3 install` commands on the Dockerfile to reduce the docker image resulting size to 1.89 GB from 2.27 GB.
 2024-01-12 18:03:02 -04:00
Timothy Jaeryang Baek ed4b3e0b32
Merge pull request #407 from anuraagdjain/feat/parallel-model-downloads 
feat: parallel model downloads
 2024-01-11 12:53:21 -08:00
Timothy J. Baek 12fbe9a491 Update .gitignore 2024-01-11 12:52:13 -08:00
Timothy J. Baek bf6685d887 chore: conflict 2024-01-11 12:51:46 -08:00
Anuraag Jain a63507c21e feat: custom port for server 2024-01-11 21:54:14 +02:00
Timothy Jaeryang Baek 5c5bde3b85
Merge pull request #451 from goecho/main 
Fix bug: Header attributes (Host, Authorization, Origin, Referer) not sanitized.
 2024-01-11 03:57:38 -08:00
goecho 74f91bc74d Fix bug: Header attributes (Host, Authorization, Origin, Referer) not sanitized 
- Resolved an issue where header attributes Host, Authorization, Origin, and Referer were not being sanitized, resulting in two major issues:
  1. Ollama requests inadvertently exposed user information, leading to data leakage.
  2. When Ollama is deployed on different servers, and the intermediary proxy layer uses the host header to locate downstream services, it fails to find them.

Root Cause:
- In FastAPI, when accessing request.headers, all header names are converted to lowercase. This is because FastAPI, and its underlying framework Starlette, adhere to the HTTP/2 standard, which mandates lowercase header field names for performance and consistency.
- In HTTP/2, enforcing lowercase header field names reduces complexity in header processing as case sensitivity is no longer a concern. Thus, regardless of the case used in client-sent header fields, the server processes them uniformly in lowercase.
- This practice is adopted in FastAPI and other modern HTTP frameworks, even in an HTTP/1.1 context, to maintain consistency with HTTP/2 and improve overall performance. As a result, header field names are always presented in lowercase in FastAPI, even if the original request used capitalization or mixed case.
 2024-01-11 14:36:34 +08:00