* refactor(vite): suppress known warnings
* fix: allow known code filetypes if there is a mismatch of expected type, or originalFile.type is empty
* refactor(useFileHandling): naming, use variable
* chore: bump data-provider
* feat: script to check recent dependency updates
* fix: override vite/rollup version for vite build fix
- also remove unused vite-plugin-html
- add vite build to file output command
* chore: bump rollup override to last known working version (v4.16.0 is breaking)
* chore(vite): increase file size cache for workbox
* fix: resolve openai to last known version using assistants v1 latest features and default header
* chore: update openrouter examples
* added pwa setup via vite config
Added apple status bar meta data
added maskable 512 icon for chrome and android devices
added vite-plugin-pwa
updated vite config to setup the pwa service worker and manifest upon build
* fix(vite): avoid pre-caching generated images
* chore: add manual chunking of larger vendor package
* chore: remove comments
---------
Co-authored-by: davecrab <65996799+davecrab@users.noreply.github.com>
* chore: replace violation cache accessors with enum
* chore: fix test
* chore(fileSchema): index timestamps
* fix(ActionService): use encoding/caching strategy for handling assistant function character length limit
* refactor(actions): async `domainParser` also resolve retrieved model (which is deployment name) to user-defined model
* style(AssistantAction): add `whitespace-nowrap` for ellipsis
* refactor(ActionService): if domain is less than or equal to encoded domain fixed length, return domain with replacement of separator
* refactor(actions): use sessions/transactions for updating Assistant Action database records
* chore: remove TTL from ENCODED_DOMAINS cache
* refactor(domainParser): minor optimization and add tests
* fix(spendTokens): use txData.user for token usage logging
* refactor(actions): add helper function `withSession` for database operations with sessions/transactions
* fix(PluginsClient): logger debug `message` field edge case
* fix(useCreateAssistantMutation): force re-render of assistants map by avoiding use shallow reference of listRes.data
* fix(AppService): regression by not including azure assistant defaults when no assistant endpoint values are set
* chore: bump example config version
* refactor(AppService): issue warnings from separate modules where possible
* refactor(AppService): consolidate AppService logic to separate modules as much as possible
* chore: bump data-provider
* chore: remove unn. variable definition
* chore: warning wording
* Improved and additional German translations
* Improved and added german translations and removed language translations here
* Spelling error fixed
* Use single quotes instead of double
* Use single quotes instead of double
* avatar fix
* chore: ensure `imageOutputType` is always defined
* ci(AppService): extra test for default value
* chore: replace default value for `desiredFormat` with `EImageOutputType` enum
* WIP: gemini-1.5 support
* feat: extended vertex ai support
* fix: handle possibly undefined modelName
* fix: gpt-4-turbo-preview invalid vision model
* feat: specify `fileConfig.imageOutputType` and make PNG default image conversion type
* feat: better truncation for errors including base64 strings
* fix: gemini inlineData formatting
* feat: RAG augmented prompt for gemini-1.5
* feat: gemini-1.5 rates and token window
* chore: adjust tokens, update docs, update vision Models
* chore: add back `ChatGoogleVertexAI` for chat models via vertex ai
* refactor: ask/edit controllers to not use `unfinished` field for google endpoint
* chore: remove comment
* chore(ci): fix AppService test
* chore: remove comment
* refactor(GoogleSearch): use `GOOGLE_SEARCH_API_KEY` instead, issue warning for old variable
* chore: bump data-provider to 0.5.4
* chore: update docs
* fix: condition for gemini-1.5 using generative ai lib
* chore: update docs
* ci: add additional AppService test for `imageOutputType`
* refactor: optimize new config value `imageOutputType`
* chore: bump CONFIG_VERSION
* fix(assistants): avatar upload
* Patch for OpenID username
`username` is generally based on email, rather than `given_name`. The challenge with `given_name` is that it can be a multi-value array (ex: "Nick, Fullname"), which completely breaks the system with:
```
LibreChat | ValidationError: User validation failed: username: Cast to string failed for value "[ 'Nickname', 'Firstname' ]" (type Array) at path "username"
LibreChat | at Document.invalidate (/app/node_modules/mongoose/lib/document.js:3200:32)
LibreChat | at model.$set (/app/node_modules/mongoose/lib/document.js:1459:12)
LibreChat | at model.set [as username] (/app/node_modules/mongoose/lib/helpers/document/compile.js:205:19)
LibreChat | at OpenIDConnectStrategy._verify (/app/api/strategies/openidStrategy.js:127:27)
LibreChat | at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
```
* Update openidStrategy.js
* refactor(openidStrategy): add helper function for stringy username
---------
Co-authored-by: Danny Avila <danny@librechat.ai>
* refactor(getFiledownload): explicit accept of `application/octet-stream`
* chore: test compose file
* chore: test compose file fix
* chore(files/download): add more logs
* Fix proxy_pass URLs in nginx.conf
* fix: proxy_pass URLs in nginx.conf to fix file downloads from URL
* chore: move test compose file to utils dir
* refactor(useFileDownload): simplify API request by passing `file_id` instead of `filepath`
* fix(processModelData): handle `openrouter/auto` edge case
* fix(Tx.create): prevent negative multiplier edge case and prevent balance from becoming negative
* fix(NavLinks): render 0 balance properly
* refactor(NavLinks): show only up to 2 decimal places for balance
* fix(OpenAIClient/titleConvo): fix cohere condition and record token usage for `this.options.titleMethod === 'completion'`
* chore: fix type issue with File Table fakeData
* refactor: new lazy loading image strategy and load images/files as part of Message Container
* feat: resubmit files when editing messages with attached files
* fix(useTextarea): trigger SendButton re-render on undo and clearing text
* refactor(PresetItems): show pin icon for default preset
* fix(ChatRoute): do not use conversation.model for useEffect, do not set default Preset if real model list is not yet fetched
Sometimes Traefik created a race condition where LibreChat was up on tcp/3080, and while Traefik was up on tcp/443, it could not route to the LibreChat container due to the multiple interfaces -- depending on how they came up. This is easily solved by simply using one interface.
* refactor(client): remove double caching of models via recoil to rely exclusively on react-query
* chore(useConversation): add modelsQuery.data dep to callback
