Skip to content

RELEASE_NOTES.md

Latest commit

 

History

History
360 lines (261 loc) · 22 KB

RELEASE_NOTES.md

File metadata and controls

360 lines (261 loc) · 22 KB

Feature Release

(v0.212.79)

New Features

1. Audio & Video Processing

  • Audio processing pipeline
    • Integrated Azure Speech transcriptions into document ingestion.
    • Splits transcripts into ~400-word chunks for downstream indexing.
  • Video Indexer settings UI
    • Added input fields in Admin Settings for Video Indexer endpoint, key and locale.

2. Multi-Model Support

  • Users may choose from multiple OpenAI deployments at runtime.
  • Model list is dynamically populated based on Admin settings (including APIM).

3. Advanced Chunking Logic

  • PDF & PPTX: page-based chunks via Document Intelligence.
  • DOC/DOCX: ~400-word chunks via Document Intelligence.
  • Images (jpg/jpeg/png/bmp/tiff/tif/heif): single-chunk OCR.
  • Plain Text (.txt): ~400-word chunks.
  • HTML: hierarchical H1–H5 splits with table rebuilding, 600–1200-word sizing.
  • Markdown (.md): header-based splitting, table & code-block integrity, 600–1200-word sizing.
  • JSON: RecursiveJsonSplitter w/ convert_lists=True, max_chunk_size=600.
  • Tabular (CSV/XLSX/XLS): pandas-driven row chunks (≤800 chars + header), sheets as separate files, formulas stripped.

4. Group Workspace Consolidation

  • Unified all group document logic into functions_documents.js.
  • Removed functions_group_documents.js duplication.

5. Bulk File Uploads

  • Support for uploading up to 10 files in a single operation, with parallel ingestion and processing.

6. GPT-Driven Metadata Extraction

  • Admins can select a GPT model to power metadata parsing.
  • All new documents are processed through the chosen model for entity, keyword, and summary extraction.

7. Advanced Document Classification

  • Admin-configurable classification fields, each with custom color-coded labels.
  • Classification metadata persisted per document for filtering and display.

8. Contextual Classification Propagation

  • When a classified document is referenced in chat, its tags are automatically applied to the conversation as contextual metadata.

9. Chat UI Enhancements

  • Left-docked conversation menu for persistent navigation.
  • Editable conversation titles inline (left & right panes stay in sync).
  • Streamlined new chat flow: click-to-start or type-to-auto-create.
  • User-defined prompts surfaced inline within the message input.

10. Semantic Reranking & Extractive Answers

  • Switched to semantic queries (query_type="semantic") on both user and group indexes.
  • Enabled extractive highlights (query_caption="extractive") to surface the most relevant snippet in each hit.
  • Enabled extractive answers (query_answer="extractive") so the engine returns a concise, context-rich response directly from the index.
  • Automatically falls back to full-text search (query_type="full", search_mode="all") whenever no literal match is found, ensuring precise retrieval of references or other exact phrases.

Bug Fixes

A. AI Search Index Migration

  • Automatically add any missing fields (e.g. author, chunk_keywords, document_classification, page_number, start_time, video_ocr_chunk_text, etc.) on every Admin page load.
  • Fixed SDK usage (Collection attribute) to update index schema without full-index replacement.

B. User & Group Management

  • User search 401 error when adding a new user to a group resolved by:
    • Implementing SerializableTokenCache in MSAL tied to Flask session.
    • Ensuring _save_cache() is called after acquire_token_by_authorization_code.
    • Refactoring get_valid_access_token() to use acquire_token_silent().
  • Restored metadata extraction & classification buttons in Group Workspace.
  • Fixed new role language in Admin settings and published an OpenAPI spec for /api/.

C. Conversation Flow & UI

  • Auto-create a new conversation on first user input, prompt selection or file upload.
  • Custom logo persistence across reboots via Base64 storage in Cosmos (max 100 px height, ≤ 500 KB).
  • Prevent uploaded files from overflowing the chat window (CSS update).
  • Sync conversation title in left pane without manual refresh.
  • Restore missing loadConversations() in chat-input-actions.js.
  • Fix feedback button behavior and ensure prompt selection sends full content.
  • Include original search_query & user_message in AI Search telemetry.
  • Ensure existing documents no longer appear “Not Available” by populating percent_complete.
  • Support Unicode (e.g. Japanese) in text-file chunking.

D. Miscellaneous Fixes

  • Error uploading file (loadConversations is not defined) fixed.
  • Classification disabled no longer displays in documents list or title.
  • Select prompt/upload file now always creates a conversation if none exists.
  • Fix new categories error by seeding missing nested settings with defaults on startup.

Breaking Changes & Migration Notes

  • Index schema must be re-migrated via Admin Settings (admin initiates in the app settings page).

(v0.203.15)

The update introduces "Workspaces," allowing users and groups to store both documents and custom prompts in a shared context. A new prompt selection feature enhances the chat workflow for a smoother experience. Additionally, admin configuration has been streamlined, and the landing page editor now supports improved Markdown formatting.

1. Renaming Documents to Workspaces

  • Your Documents → Your Workspace
  • Group Documents → Group Workspaces
  • All references, routes, and templates updated (documents.html → workspace.html, group_documents.html → group_workspaces.html).
  • New admin settings flags: enable_user_workspace and enable_group_workspaces replaced the old enable_user_documents / enable_group_documents.

2. Custom Prompt Support

  • User Prompts:
    • New backend routes in route_backend_prompts.py (CRUD for user-specific prompts).
  • Group Prompts:
    • New backend routes in route_backend_group_prompts.py (CRUD for group-shared prompts).

3. Chat Page Enhancements

  • Prompt Selection Dropdown:
    • New button (“Prompts”) toggles a dropdown for selecting saved user/group prompts.
    • Eliminates copy-paste; helps users insert larger or more complex prompts quickly.
    • Lays groundwork for future workflow automation.
  • Toast Notifications for errors and status messages (replacing browser alerts).

4. Cosmos Containers

  • Added prompts_container and group_prompts_container.

  • Simplified or standardized the container creation logic in config.py.

(v0.202.41)

  • Azure Government Support:

    • Introduced an AZURE_ENVIRONMENT variable (e.g. "public" or "usgovernment") and logic to handle separate authority hosts, resource managers, and credential scopes.
    # Azure Cosmos DB
AZURE_COSMOS_ENDPOINT="<your-cosmosdb-endpoint>"
AZURE_COSMOS_KEY="<your-cosmosdb-key>"
AZURE_COSMOS_AUTHENTICATION_TYPE="key" # key or managed_identity

# Azure Bing Search
BING_SEARCH_ENDPOINT="https://api.bing.microsoft.com/"

# Azure AD Authentication
CLIENT_ID="<your-client-id>"
TENANT_ID="<your-tenant-id>"
AZURE_ENVIRONMENT="public" #public, usgovernment
SECRET_KEY="32-characters" # Example - "YouSh0uldGener8teYour0wnSecr3tKey!", import secrets; print(secrets.token_urlsafe(32))

  • Admin Settings Overhaul:

    • Route & UI: Added route_backend_settings.py and significantly expanded admin_settings.html to configure GPT, Embeddings, Image Gen, Content Safety, Web Search, AI Search, and Document Intelligence—all from a single Admin page.
    • APIM Toggles: Each service (GPT, Embeddings, Image Generation, Content Safety, etc.) can now be routed through Azure API Management instead of direct endpoints by switching a toggle.
    • “Test Connection” Buttons: Each service (GPT, Embeddings, Image Generation, Content Safety, Bing Web Search, Azure AI Search, and Document Intelligence) now has a dedicated “Test Connection” button that performs a live connectivity check.

  • Improved Safety Features:

    • New pages/sections for “Admin Safety Violations” vs. “My Safety Violations.”

  • Miscellaneous Frontend & Template Updates:

    • All templates now reference an app_settings.app_title for a dynamic page title.
    • Enhanced navigation and labeling in “My Documents,” “My Groups,” and “Profile” pages.

Bug Fixes

  • Conversation Pipeline:
    • Removed the "image" role from the allowed conversation roles to streamline message handling.
  • Group Management:
    • Now correctly passes and references the current user’s ID in various group actions.

(v0.201.5)

1. Managed Identity Support

  • Azure Cosmos DB (enabled/disabled via environment variable)
  • Azure Document Intelligence (enabled/disabled via app settings)
  • Azure AI Search (enabled/disabled via app settings)
  • Azure OpenAI (enabled/disabled via app settings)

2. Conversation Archiving

  • Introduced a new setting

    enable_conversation_archiving
    • When enabled, deleting a conversation will first copy (archive) the conversation document into an archived_conversations_container before removing it from the main conversations container.
    • Helps preserve conversation history if you want to restore or analyze it later.

3. Configuration & Environment Variable Updates

  • example.env & example_advance_edit_environment_variables.json:
    • Added AZURE_COSMOS_AUTHENTICATION_TYPE to demonstrate how to switch between key-based or managed_identity-based authentication.
    • Cleaned up references to Azure AI Search and Azure Document Intelligence environment variables to reduce clutter and reflect the new approach of toggling authentication modes.
  • Default Settings Updates
    • functions_settings.py has more descriptive defaults covering GPT, Embeddings, and Image Generation for both key-based and managed identity scenarios.
    • New config fields such as content_safety_authentication_type, azure_document_intelligence_authentication_type, and enable_conversation_archiving.

6. Bug Fixes

  • Fixed bug affecting the ability to manage groups
    • Renamed or refactored manage_groups.js to manage_group.js, and updated the template (manage_group.html) to use the new filename.
    • Injected groupId directly via Jinja for improved client-side handling.

7. Architecture Diagram Updates

  • Updated architecture.vsdx and architecture.png to align with the new authentication flow and container usage.

How to Use / Test the New Features

  1. Enable Managed Identity
    • In your .env or Azure App Service settings, set AZURE_COSMOS_AUTHENTICATION_TYPE="managed_identity" (and similarly for azure_document_intelligence_authentication_type, etc.).
    • Ensure the Azure resource (e.g., App Service, VM) has a system- or user-assigned Managed Identity with the correct roles (e.g., “Cosmos DB Account Contributor”).
    • Deploy, and the application will now connect to Azure resources without storing any keys in configuration.
  2. Test Conversation Archiving
    • In the Admin Settings, enable Enable Conversation Archiving.
    • Delete a conversation.
    • Verify the record is copied to archived_conversations_container before being removed from the active container.
  3. Check New Environment Variables
    • Review example.env and example_advance_edit_environment_variables.json for the newly added variables.
    • Update your application settings in Azure or your local .env accordingly to test various authentication modes (key vs. managed identity).

(V0.199.3)

We introduced a robust user feedback system, expanded content-safety features for both admins and end users, added new Cosmos DB containers, and refined route-level permission toggles. These changes help administrators collect feedback on AI responses, manage content safety more seamlessly, and give end users clearer ways to manage their documents, groups, and personal logs. Enjoy the new functionality, and let us know if you have any questions or issues!

  1. New “User Feedback” System
    • Thumbs Up / Thumbs Down: Users can now provide feedback on individual AI responses (when enabled in App Settings)
    • Frontend Feedback Pages:
      • /my_feedback page shows each user’s submitted feedback.
      • /admin/feedback_review page allows admins to review, filter, and manage all feedback.
  2. Extended Content Safety Features
    • New “Safety Violations” Page: Admins can manage safety violations.
    • New “My Safety Violations” Page: Users can view their violations and add personal notes to each violation.
  3. New or Updated Database Containers
    • feedback_container for user feedback.
    • archived_conversations_container / archived_feedback_container / archived_safety_container for long-term archival.
  4. Route-Level Feature Toggles
    • enabled_required(setting_key) Decorator:
      • Dynamically block or allow routes based on an admin setting (e.g., enable_user_documents or enable_group_documents).
      • Reduces scattered if checks; you simply annotate the route.
  5. Conversation & Messaging Improvements
    • Unique message_id for Each Chat Message:
      • Every user, assistant, safety, or image message now includes a message_id.
      • Makes it easier to tie user feedback or safety logs to a specific message.
    • Public vs. Secret Settings:
      • Frontend references a public_settings = sanitize_settings_for_user(settings) to avoid the potential to expose secrets on the client side.
  6. UI/UX Tweaks
    • Chat Layout Updates:
      • “Start typing to create a new conversation…” message if none selected.
      • Automatic creation of new conversation when user tries to send a message with no active conversation.
    • Navigation Bar Adjustments:
      • Consolidated admin links into a dropdown.
      • “My Account” dropdown for quick access to “My Groups,” “My Feedback,” etc., if enabled.

(v0.196.9)

  1. Content Safety Integration
    • New Safety Tab in Admin Settings: A dedicated “Safety” section now appears under Admin Settings, allowing you to enable Azure Content Safety, configure its endpoint and key, and test connectivity.
    • Real-Time Message Scanning: If Content Safety is enabled, user prompts are scanned for potentially disallowed content. Blocked messages are flagged and a “safety” message is added to the conversation log in place of a normal AI reply.
    • Admin Safety Logs: Site admins (with “Admin” role) can view a new “Safety Violations” page (at /admin/safety_violations) showing blocked or flagged messages. Admins can update the status, action taken, or notes on each violation.
  2. Expanded APIM Support for GPT, Embeddings, and Image Generation
    • Fine-Grained APIM Toggles: You can now enable or disable APIM usage independently for GPT, embeddings, and image generation. Each service has its own APIM endpoint, version, and subscription key fields in Admin Settings.
    • UI-Driven Switching: Check/uncheck “Enable APIM” to toggle between native Azure OpenAI endpoints or APIM-managed endpoints, all without redeploying the app.
  3. Workspaces & Documents Configuration
    • User Documents and Group Documents: A new “Workspaces” tab in Admin Settings (replacing the old “Web Search” tab) lets you enable or disable user-specific documents and group-based documents.
    • Group Documents Page: The front-end for Group Documents now checks whether “Enable My Groups” is turned on. If enabled, members can manage shared group files and see group-level search results.
    • My Groups & Group Management: Navigation includes “My Groups” (if group features are enabled). This leads to a new set of pages for viewing groups, managing memberships, transferring ownership, and more.
  4. Search & Extract Tab
    • Azure AI Search & Document Intelligence: Moved Bing Web Search, Azure AI Search, and Azure Document Intelligence settings into a new “Search and Extract” tab (replacing the older “Web Search” tab).
    • Bing Search Toggle: If you enable web search, the user can optionally include Bing results in chat queries.
    • Azure Document Intelligence: Configure endpoints and keys for file ingestion (OCR, form analysis, etc.) in a more structured place within Admin Settings.
  5. Updated UI & Navigation
    • Admin Dropdown: Admin-specific features (App Settings, Safety Violations, etc.) are grouped in an “Admin” dropdown on the main navbar.
    • Safety: For Content Safety (as noted above).
    • Search & Extract: For Bing Search, Azure AI Search, and Document Intelligence.
    • Minor Styling Adjustments: Updated top navbar to show/hide “Groups” or “Documents” links based on new toggles (Enable Your Documents, Enable My Groups).

(v0.191.0)

  1. Azure API Management (APIM) Support

    • New APIM Toggles: In the Admin Settings, you can now enable or disable APIM usage separately for GPT, embeddings, and image generation.
    • APIM Endpoints & Subscription Keys: For each AI service (GPT, Embeddings, Image Generation), you can specify an APIM endpoint, version, deployment, and subscription key—allowing a unified API gateway approach (e.g., rate limiting, authentication) without changing your core service code.
    • Seamless Switching: A single checkbox (Enable APIM) within each tab (GPT, Embeddings, Image Generation) instantly switches the app between native Azure endpoints and APIM-protected endpoints, with no redeployment required.

  2. Enhanced Admin Settings UI

    • Advanced Fields: Collapsible “Show Advanced” sections for GPT, Embeddings, and Image Generation let you configure API versions or other fine-tuning details only when needed.
    • Test Connectivity: Each service tab (GPT, Embeddings, Image Gen) now has a dedicated “Test Connection” button, providing immediate feedback on whether your settings and credentials are valid.
    • Improved UX for Keys: Updated show/hide password toggles for all key fields (including APIM subscription keys), making it easier to confirm you’ve entered credentials correctly.

  3. Miscellaneous Improvements

    • UI Polishing: Minor styling updates and improved tooltips in Admin Settings to guide first-time users.
    • Performance Tweaks: Reduced initial load time for the Admin Settings page when large model lists are returned from the OpenAI endpoints.
    • Logging & Error Handling: More descriptive error messages and client-side alerts for failed fetches (e.g., if the user tries to fetch GPT models but hasn’t set the endpoint properly).

v0.191.0

  1. Azure API Management (APIM) Support

    • New APIM Toggles: In the Admin Settings, you can now enable or disable APIM usage separately for GPT, embeddings, and image generation.
    • APIM Endpoints & Subscription Keys: For each AI service (GPT, Embeddings, Image Generation), you can specify an APIM endpoint, version, deployment, and subscription key—allowing a unified API gateway approach (e.g., rate limiting, authentication) without changing your core service code.
    • Seamless Switching: A single checkbox (Enable APIM) within each tab (GPT, Embeddings, Image Generation) instantly switches the app between native Azure endpoints and APIM-protected endpoints, with no redeployment required.

  2. Enhanced Admin Settings UI

    • Advanced Fields: Collapsible “Show Advanced” sections for GPT, Embeddings, and Image Generation let you configure API versions or other fine-tuning details only when needed.
    • Test Connectivity: Each service tab (GPT, Embeddings, Image Gen) now has a dedicated “Test Connection” button, providing immediate feedback on whether your settings and credentials are valid.
    • Improved UX for Keys: Updated show/hide password toggles for all key fields (including APIM subscription keys), making it easier to confirm you’ve entered credentials correctly.

  3. Miscellaneous Improvements

    • UI Polishing: Minor styling updates and improved tooltips in Admin Settings to guide first-time users.
    • Performance Tweaks: Reduced initial load time for the Admin Settings page when large model lists are returned from the OpenAI endpoints.
    • Logging & Error Handling: More descriptive error messages and client-side alerts for failed fetches (e.g., if the user tries to fetch GPT models but hasn’t set the endpoint properly).

v0.190.1

  1. Admin Settings UI

    • Configure Azure OpenAI GPT, Embeddings, Image Generation, and Bing Search settings directly through an in-app interface (rather than .env).
    • Choose between key-based or managed identity authentication for GPT, Embeddings, and Image Generation.
    • Dynamically switch models/deployments without redeploying the app.

  2. Multiple Roles & Group Permissions

    • Roles include Owner, Admin, DocumentManager, and User.
    • Group Owners/Admins can invite or remove members, manage documents, and set “active workspace” for group-based search.

  3. One-Click Switching of Active Group

    • Users in multiple groups can quickly switch their active group to see group-specific documents and chat references.

  4. Ephemeral Document Upload

    • Upload a file for a single conversation. The file is not saved in Azure Cognitive Search; instead, it is only used for the session’s RAG context.

  5. Inline File Previews in Chat

    • Files attached to a conversation can be previewed directly from the chat, with text or data displayed in a pop-up.

  6. Optional Bing Web Search

    • Administrators can enable or disable web search. When enabled, the user can toggle “Search the Web” while chatting to incorporate Bing results.

  7. Optional Image Generation

    • Users can toggle an “Image” button to create images via Azure OpenAI (e.g., DALL·E) when configured in Admin Settings.

  8. App Roles & Enterprise Application

    • Provides a robust way to control user access at scale.
    • Admins can assign roles to new users or entire Azure AD groups.