- Increase sampling timeout from 30s to 300s in semantic.py to accommodate
slower local LLMs like Ollama
- Refactor RAG integration tests to support multiple providers (ollama,
openai, anthropic, bedrock)
- Remove unnecessary embedding_provider fixture since MCP server handles
embeddings internally
- Add --provider flag via tests/integration/conftest.py
- Add provider_fixtures.py with factory functions for generation providers
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
The refactor in fafeaf3 moved background tasks to Starlette server lifespan
but broke nc_get_vector_sync_status because it still looked for streams in
FastMCP's AppContext (lifespan_context).
Add VectorSyncState module singleton to bridge the lifespans:
- starlette_lifespan sets the singleton when starting background tasks
- app_lifespan_basic reads from singleton and includes in AppContext
- MCP tools can now access document_receive_stream for pending count
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
The refactor in fafeaf3 moved background tasks to Starlette server lifespan
but broke nc_get_vector_sync_status because it still looked for streams in
FastMCP's AppContext (lifespan_context).
Add VectorSyncState module singleton to bridge the lifespans:
- starlette_lifespan sets the singleton when starting background tasks
- app_lifespan_basic reads from singleton and includes in AppContext
- MCP tools can now access document_receive_stream for pending count
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Move scanner/processor tasks from FastMCP session lifespan to Starlette
server lifespan (correct architecture: background tasks run once at
server level, not per-session)
- Change default CLI transport from SSE to streamable-http
- Remove SSE transport option from CLI (SSE is deprecated)
- Remove SSE client session factory from test fixtures
- Add tracing instrumentation to BM25 hybrid search operations for
better observability
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Change create_tag() to use WebDAV POST instead of OCS API which
returned 404 in some Nextcloud versions
- Add llm_judge() helper that evaluates system output against ground
truth with simple TRUE/FALSE prompt
- Replace keyword-based assertions in RAG tests with LLM judge for
more flexible semantic evaluation
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add get_file_info() to get file info including file ID via PROPFIND
- Add create_tag() to create system tags via OCS API
- Add get_or_create_tag() for idempotent tag creation
- Add assign_tag_to_file() to assign tags to files via WebDAV
- Add remove_tag_from_file() to remove tags from files
Also refactors RAG evaluation:
- Add indexed_manual_pdf fixture using existing nc_client/nc_mcp_client
- Remove manual tag creation steps from workflow (now handled by fixture)
- Add comprehensive unit tests for new WebDAV methods
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Adds a manually-triggered GitHub Actions workflow for RAG evaluation:
- Builds Nextcloud User Manual PDF from documentation source
- Uploads PDF to Nextcloud via WebDAV
- Tags file with 'vector-index' for vector sync indexing
- Waits for vector sync to complete
- Runs RAG integration tests with OpenAI/GitHub Models API
Inputs:
- embedding_model: OpenAI embedding model (default: openai/text-embedding-3-small)
- generation_model: OpenAI generation model (default: openai/gpt-4o-mini)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Adds OpenAI provider to the unified provider architecture (ADR-015),
supporting:
- OpenAI API (api.openai.com)
- GitHub Models API (models.github.ai/inference)
- OpenAI-compatible endpoints (Fireworks, Together, etc.)
Features:
- Embedding support with text-embedding-3-small/large models
- Text generation via chat completions API
- Automatic retry with exponential backoff for rate limits
- Provider auto-detection in registry (priority after Bedrock)
Environment variables:
- OPENAI_API_KEY: API key (required)
- OPENAI_BASE_URL: Base URL override (optional)
- OPENAI_EMBEDDING_MODEL: Embedding model (default: text-embedding-3-small)
- OPENAI_GENERATION_MODEL: Generation model (default: gpt-4o-mini)
Also adds:
- Integration tests for RAG pipeline with MCP sampling
- MCP client sampling support for integration tests
- Ground truth Q&A pairs for Nextcloud User Manual
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
The Smithery scanner was reporting "0 tools" despite the server returning
valid tool definitions. Root cause: the server was returning SSE-formatted
responses (event: message\ndata: {...}) which the scanner couldn't parse.
Changes:
- Add json_response=True to FastMCP for Smithery stateless mode
- Clean up verbose docstring examples in semantic.py and webdav.py
The MCP spec allows both SSE and plain JSON responses for HTTP transport.
Setting json_response=True returns Content-Type: application/json with
plain JSON-RPC instead of text/event-stream with SSE format.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Replace sequential Qdrant scroll calls with batch retrieve
(50 HTTP requests → 1 request, ~50x faster vector fetch)
- Add point_id to SearchResult to enable batch retrieval by Qdrant point ID
- Reuse query embedding from search algorithm in viz_routes
(eliminates redundant embedding call, saves ~30ms)
- Make BM25 encode() async with thread pool to avoid blocking event loop
(~4.4s was blocking, now properly async)
- Run PCA computation in thread pool to avoid blocking event loop
(~1.2s was blocking, now properly async)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Reorganize README to promote Smithery as the fastest way to get started:
- Quick Start now features Smithery one-click deployment
- Docker instructions moved to separate "Docker (Self-Hosted)" section
- Added note about Smithery's stateless mode limitations
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
ADR-016: For container runtime deployment, Smithery does not auto-generate
the .well-known/mcp-config endpoint like it does for Python CLI runtime.
Changes:
- Remove [tool.smithery] from pyproject.toml (not used in container mode)
- Remove smithery_server.py (Python CLI runtime specific)
- Add .well-known/mcp-config endpoint to return JSON Schema config
- Add SmitheryConfigMiddleware to extract config from URL query params
- Use ContextVar to pass session config to tool handlers
The container runtime passes config as URL query parameters to /mcp:
GET /mcp?nextcloud_url=...&username=...&app_password=...
Tested:
- All 164 unit tests passing
- Docker container builds successfully
- .well-known/mcp-config returns valid JSON Schema
- Health endpoints working
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Adds support for Smithery hosted deployment with stateless operation:
- Add DeploymentMode enum with SELF_HOSTED and SMITHERY_STATELESS modes
- Add get_deployment_mode() to detect mode from SMITHERY_DEPLOYMENT env var
- Update get_client() to create per-request clients from session config
- Add conditional tool registration (skip semantic search in Smithery mode)
- Add conditional /app admin UI mounting (skip in Smithery mode)
- Create smithery.yaml with configSchema for user credentials
- Create Dockerfile.smithery for minimal stateless container
- Create smithery_main.py entrypoint for Smithery deployment
In Smithery mode:
- Users provide nextcloud_url, username, app_password via session config
- Each request creates a fresh NextcloudClient (no state between requests)
- Semantic search tools are disabled (no vector database)
- Admin UI (/app) is disabled (no webhooks, vector viz)
All existing self-hosted functionality remains unchanged.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add architecture decision record for supporting Smithery-hosted MCP
server in a stateless mode for multi-user public Nextcloud instances.
Key decisions:
- New SMITHERY_STATELESS deployment mode alongside SELF_HOSTED
- Session-based configuration (nextcloud_url, username, app_password)
- Feature subset excluding semantic search and background sync
- Admin UI (/app) excluded in Smithery mode
- Per-request client creation from session config
This enables users to try the MCP server without self-hosting
infrastructure while supporting multiple Nextcloud instances.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Drawing directly with ImageDraw on RGBA mode doesn't blend alpha
properly. Use Image.alpha_composite() with a transparent overlay
to achieve correct semi-transparent highlight fills.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Replace parallel per-page extraction with single to_markdown(page_chunks=True)
call. This is more efficient as pymupdf4llm can optimize internally for
full-document processing instead of making N separate calls for N pages.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Phase 1 - PDF Highlighting Optimization:
- Render each page ONCE instead of once per chunk (N chunks = 1 render, not N)
- Use PIL to draw bounding boxes on copied base images (fast) instead of
re-rendering page via pymupdf (slow)
- Add _find_chunk_bbox() to extract bbox without modifying page
Phase 2 - Parallel Page Extraction:
- Use anyio task group with run_sync() for parallel page extraction
- Each page extracted in separate thread via anyio.to_thread.run_sync()
- Event loop stays responsive during extraction
- Remove obsolete _process_sync() method
Expected improvement: 30-50% reduction in total PDF processing time.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Previously, pymupdf4llm.to_markdown() was called twice - once in
PyMuPDFProcessor during indexing and again in PDFHighlighter during
visualization. Different image path lengths caused different character
offsets, leading to highlighted pages not matching their chunks.
Also fixed issue where all chunks on the same page showed all highlights
instead of just their own highlight. Now restores original page contents
between chunks using xref stream caching.
Changes:
- Add PDFHighlighter class requiring pre-computed page_boundaries and
full_text from document processor (no fallback extraction)
- Pass pre-computed data from processor to highlighter
- Extract page-relative portion of chunk text for cross-page chunks
- Add bounding box highlighting using text anchor search
- Run highlight generation in parallel with embedding/BM25
- Cache and restore page contents to isolate highlights per chunk
Results: Highlighting success rate improved from 51% to 95% (121/128).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Implements optional context expansion for semantic search results that
fetches adjacent chunks (N-1 and N+1) from Qdrant to provide before/after
context. Removes configurable chunk overlap (default 200 chars) to avoid
duplicate text appearing in both context and excerpt.
Key changes:
- Add include_context and context_chars parameters to nc_semantic_search
and nc_semantic_search_answer tools
- Implement Qdrant cache fast path for chunk retrieval (avoids re-fetching
and re-parsing documents, especially important for PDFs)
- Add _get_chunk_by_index_from_qdrant() to fetch adjacent chunks
- Remove chunk overlap from before_context (last N chars) and after_context
(first N chars) to prevent duplicate text
- Fetch context in parallel with anyio.Semaphore (max 20 concurrent)
- Pass through page_number from SearchResult to SemanticSearchResult
- Remove document-level deduplication (keep chunk-level dedup from algorithm)
Context expansion is opt-in via include_context=true parameter. When enabled:
- Populates has_context_expansion, marked_text, before_context, after_context
- Adds truncation flags when context exceeds context_chars limit
- Falls back to document fetch for legacy data with truncated excerpts
Related: nextcloud_mcp_server/search/context.py:87-382,
nextcloud_mcp_server/server/semantic.py:161-255
The processor was not setting is_placeholder field when writing real
document chunks to Qdrant. This caused the placeholder filter to exclude
all documents (since None != False), resulting in 0 search results.
Now explicitly sets is_placeholder: False in payload when writing real
indexed chunks, allowing search filters to correctly distinguish between
placeholders and real documents.
- Switch from sequential loop to /api/embed batch endpoint
- Use 'input' array parameter instead of individual 'prompt' requests
- Process in chunks of 32 to avoid quality degradation (issue #6262)
- Reduces HTTP overhead: 128 texts = 4 requests instead of 128
- Maintains backward compatibility with embed() for single embeddings
Ref: ollama/ollama#6262
- Changed from 2x (120s) to 5x (300s) scan interval
- Large PDFs take 3-4 minutes to process, need longer threshold
- Prevents premature requeuing of in-flight documents
- Only requeue documents if placeholder is older than 2x scan interval (120s default)
- Prevents scanner from immediately requeuing in-flight documents
- Fixes issue where PDFs were being reprocessed every 60 seconds
- Staleness check applied to both notes and files scanning logic
Qdrant validation rejects None for sparse vectors in named vector dicts.
Use models.SparseVector(indices=[], values=[]) instead to create valid
empty sparse vectors for placeholder points.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Introduces a placeholder-based state tracking system to prevent duplicate
document processing during the gap between scanner queuing and processor
completion.
**Key Changes:**
1. **Placeholder Helper Functions** (`vector/placeholder.py`):
- `write_placeholder_point()` - Creates zero-vector placeholder when queuing
- `query_document_metadata()` - Queries for existing entry (placeholder or real)
- `delete_placeholder_point()` - Removes placeholder before writing real vectors
- `get_placeholder_filter()` - Filters placeholders from user-facing queries
2. **Scanner Updates** (`vector/scanner.py`):
- Replace `indexed_at` comparison with `modified_at` comparison
- Write placeholder before queuing each document
- Query per-document metadata instead of bulk-querying indexed_at
- Fixes bug where files were resubmitted every scan cycle
3. **Processor Updates** (`vector/processor.py`):
- Delete placeholder before upserting real vectors
- Ensures no duplicate points in Qdrant
4. **Query Filters** (all search files):
- Add `get_placeholder_filter()` to all user-facing queries
- Ensures placeholders never appear in search results or visualizations
- Applied to: bm25_hybrid.py, semantic.py, viz_routes.py, algorithms.py
**Architecture:**
- Placeholders use zero vectors with dimension from embedding service
- Payload includes `is_placeholder: True` flag for filtering
- Status field tracks: "pending", "processing", "completed", "failed"
- Deterministic UUIDs using uuid5 for consistent point IDs
**Impact:**
- Eliminates duplicate processing of same documents
- Fixes race condition where long-running documents get queued multiple times
- Prevents scanner from resubmitting files every scan cycle
- Maintains clean separation between in-flight and indexed documents
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
When vector visualization search returns zero results, the code was returning
query_coords: null, which caused JavaScript error "can't access property 0,
queryCoords is null" when the frontend tried to access the array.
Changed to return empty array [] to match expected type and prevent crash.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
This commit fixes two critical issues with PDF processing:
1. **Text extraction mismatch (context expansion bug)**:
- Indexing used pymupdf4llm.to_markdown() producing markdown text
- Context expansion used page.get_text() producing plain text
- Different text formats caused character offset misalignment
- Search would find correct chunk, but expansion showed wrong section
- Fixed by making context.py use pymupdf4llm.to_markdown() consistently
2. **Diagnostic logging for page number assignment**:
- Added logging to verify page_boundaries exist in metadata
- Added logging to verify assign_page_numbers() assigns values
- Helps diagnose why page numbers show as null in search results
3. **mime_type storage bug**:
- Fixed incorrect field reference in processor.py:405
- Was using file_metadata.get("content_type", "")
- Should use content_type from WebDAV response
Changes:
- nextcloud_mcp_server/search/context.py: Use pymupdf4llm.to_markdown()
for PDF text extraction to match indexing method
- nextcloud_mcp_server/vector/processor.py: Add diagnostic logging for
page boundaries and assignment, fix mime_type storage
- tests/unit/client/test_webdav.py: Fix import sorting
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- algorithms.py: Revert SearchResult.id to int (all docs use int IDs now)
- semantic.py: Revert SemanticSearchResult.id to int, remove Union import
- viz_routes.py: Remove str() conversion when querying doc_id from Qdrant
- viz_routes.py: Convert doc_id from query param to int in chunk context
Fixes vector visualization which was collapsing all chunks to a single
point because Qdrant queries were failing to match doc_id (string vs int).
- scanner.py: Use file_info['id'] as doc_id instead of file_path
- scanner.py: Pass file_path in DocumentTask for content retrieval
- processor.py: Store file_path in Qdrant payload for later lookup
- context.py: Add _get_file_path_from_qdrant() to resolve file_id → file_path
- context.py: Update get_chunk_with_context() to handle file ID resolution
This makes the system resilient to file renames since file IDs are stable
identifiers in Nextcloud, while file paths can change.
Notes are indexed as "{title}\n\n{content}" in processor.py but were
being retrieved as just content during chunk expansion, causing
chunk_start_offset and chunk_end_offset to be misaligned.
This fix reconstructs the full content structure when fetching notes
for chunk expansion, ensuring the displayed chunks match the excerpts
shown in search results.
Fixes chunk/excerpt mismatch reported in vector visualization.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Major improvements to vector visualization page:
- Refactor PCA to display individual chunks instead of averaged documents
- Add context expansion module for fetching surrounding text from notes and PDFs
- Update deduplication to use (doc_id, doc_type, chunk_start, chunk_end) keys
- Fix Alpine.js rendering with chunk-specific keys including offsets
- Refactor authentication helper to return NextcloudClient for better reuse
- Add async context manager support to NextcloudClient
Technical details:
- viz_routes.py: Fetch specific chunk vectors instead of averaging per document
- context.py: New module supporting both notes and PDF text extraction via PyMuPDF
- search algorithms: Extract page_number, chunk_index, total_chunks from Qdrant
- vector-viz.js/html: Use chunk positions in expansion tracking keys
This enables users to see which specific chunks match their query
and view them with surrounding context in the PCA visualization.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
This commit addresses multiple issues with async operations, PDF metadata
extraction, and type safety in document processing and search.
## Async/Await Fixes
- processor.py:259 - Added await for chunker.chunk_text(content)
- processor.py:270 - Added await for bm25_service.encode_batch(chunk_texts)
- tests/unit/test_document_chunker.py - Converted all 12 test methods to async
## PDF Metadata Enhancement
- pymupdf.py:143 - Added file_size metadata extraction
- pymupdf.py:145-206 - Refactored to extract text page-by-page
- Manually loop through pages instead of using page_chunks=True
- Generate page_boundaries metadata for precise page tracking
- Works around pymupdf.layout.activate() breaking page_chunks=True
- processor.py:32-66 - Added assign_page_numbers() helper function
- Assigns page numbers to chunks based on overlap with page boundaries
- Handles chunks spanning multiple pages
- processor.py:298-300 - Call assign_page_numbers() for PDF files
## Type Safety Fixes
- bm25_hybrid.py:184 - Removed int() conversion of doc_id
- semantic.py:131 - Removed int() conversion of doc_id
- viz_routes.py:275 - Removed int() conversion of doc_id
- Added comments documenting that doc_id can be int (notes) or str (file paths)
## Testing
- All 18 tests passing (12 unit + 6 integration)
- No type errors in modified files
- Container logs show successful processing
- Vector viz searches working correctly
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Get container dimensions before creating Plotly layout to render at correct size immediately
- Add init() method with window resize listener for responsive plot sizing
- Remove post-render resize call (no longer needed with explicit dimensions)
- Improve colorbar positioning and scene domain configuration
This eliminates the visual "jump" during initial render and ensures the plot resizes smoothly when the browser window changes size.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Two fixes for the vector visualization page:
1. **CSS Loading Fix**: Moved CSS <link> from vector_viz.html fragment
to user_info.html <head> block. HTMX fragments don't process <link>
tags in <head>, causing unstyled page. Now CSS loads correctly.
2. **Camera Preservation**: Modified renderPlot() to preserve camera
position when toggling query point visibility. Previously, toggling
the "Show Query Point" checkbox would reset zoom/rotation to default.
Now reads existing camera settings from plot before updating.
Related: nextcloud_mcp_server/auth/static/vector-viz.js:123-130
Related: nextcloud_mcp_server/auth/templates/user_info.html:12
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Extract CSS and JavaScript into separate static files
- Created nextcloud_mcp_server/auth/static/vector-viz.css
- Created nextcloud_mcp_server/auth/static/vector-viz.js
- Updated templates to reference external assets
- Fix vector visualization issues:
- Normalize vectors before PCA to match Qdrant's cosine distance
- Add zero-norm and NaN detection/handling for large datasets
- Enable responsive Plotly sizing (autosize + responsive config)
- Widen plot area to full viewport width with minimized margins
- Improve visualization accuracy:
- Query point now positioned correctly relative to documents
- Handles 200+ points without JSON serialization errors
- Full-width plot maximizes screen space utilization
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
This commit updates the web interface to better align with Nextcloud's
design system and improve the Vector Viz layout.
Changes:
- Replace emoji icons with Material Design SVG icons for better
consistency with Nextcloud apps
- Simplify navigation styling with minimal padding and subtle active
states (250px width)
- Update CSS variables to match Nextcloud design system
- Restructure Vector Viz from two-column to single-column vertical
layout for better plot visibility
- Move search controls to compact horizontal grid at top
- Make navigation toggle always visible (not just on mobile)
- Fix plot container sizing with overflow:visible to prevent colorbar
clipping
- Remove heavy shadows and custom card styling for cleaner aesthetic
- Add error and success page templates with consistent styling
Technical details:
- Preserve Alpine.js for reactive functionality
- Use CSS Grid for responsive horizontal controls layout
- Add smooth transitions for navigation collapse/expand
- Maintain HTMX for dynamic content loading
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Migrates from custom word-based chunking to LangChain's MarkdownTextSplitter
for better semantic search quality. This implements the chunking portion of
ADR-011.
Changes:
- Replace custom regex word chunker with MarkdownTextSplitter
- Optimized for Markdown content (headers, code blocks, lists)
- Convert from word-based (512 words) to character-based (2048 chars) chunking
- Maintain backward-compatible ChunkWithPosition interface
- Update configuration defaults and validation
- Update all unit tests (12/12 passing)
Benefits:
- Respects markdown structure boundaries
- Never breaks code blocks or headers mid-chunk
- Preserves semantic coherence within chunks
- Expected 20-30% improvement in recall quality
- Industry-standard approach (used by production RAG systems)
Note: Full reindex required to apply new chunking to existing documents.
Current vector database still contains old word-based chunks.
Related: ADR-011 (Improving Semantic Search Quality)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
This commit enhances the vector visualization interface with better score
transparency and improved UX:
**Dual-Score Display:**
- Store original algorithm scores before normalization (viz_routes.py:203)
- Display both raw and normalized scores: "Raw Score: 0.842 (89% relative)"
- Update plot hover text with dual scores (userinfo_routes.py:740)
- Fixes issue where all queries showed at least one 100% match regardless
of actual relevance (normalization artifact)
**UI Improvements:**
1. Fusion Method dropdown: Changed from x-show to :disabled
- Prevents jarring layout shift when switching algorithms
- Dropdown stays visible but grayed out when Semantic is selected
- Better UX with opacity: 0.5 and cursor: not-allowed
2. Score Threshold: Changed step from 0.1 to "any"
- Allows arbitrary float precision (0.7, 0.85, 0.123)
- Users can now fine-tune threshold values
3. Document Types: Converted multi-select to checkbox grid
- Replaced clunky Ctrl/Cmd multi-select listbox
- Checkbox grid with cleaner layout
- Positioned left of Score Threshold and Result Limit inputs
- More intuitive UX
**Technical Details:**
- Raw score ranges vary by algorithm:
- Semantic: 0.0-1.0 (cosine similarity)
- BM25 RRF: ~0.001-0.033 (Reciprocal Rank Fusion)
- BM25 DBSF: Can exceed 1.0 (Distribution-Based Score Fusion)
- Normalized scores (0-1) used for visual encoding (marker size, color)
- Original scores preserved in API response via getattr fallback
Files modified:
- nextcloud_mcp_server/auth/viz_routes.py (store original_score)
- nextcloud_mcp_server/auth/templates/vector_viz.html (UI controls)
- nextcloud_mcp_server/auth/userinfo_routes.py (plot hover text)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Added support for two fusion algorithms (RRF and DBSF) to combine dense
semantic and sparse BM25 search results, with comprehensive documentation
and unit tests.
Changes:
- Added fusion parameter to nc_semantic_search and nc_semantic_search_answer tools
- Updated ADR-014 with detailed comparison of RRF vs DBSF fusion algorithms
- Added unit tests for fusion algorithm initialization and validation
- Updated search_method in responses to include fusion type (e.g., "bm25_hybrid_rrf")
Fusion Algorithms:
- RRF (Reciprocal Rank Fusion): Default, rank-based, general-purpose
- DBSF (Distribution-Based Score Fusion): Score normalization using statistics
RRF is recommended for most use cases due to its robustness and established
track record. DBSF may provide better results when retrieval systems have
very different score distributions.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Track character offsets (start_offset, end_offset) for each chunk in vector
database metadata, enabling precise chunk highlighting in visualization pane.
Changes:
- processor.py: Store chunk_start_offset and chunk_end_offset in Qdrant metadata
- processor.py: Added metadata_version=2 to indicate position tracking support
- search/semantic.py: Return chunk positions from search results
- server/semantic.py: Expose chunk positions in API responses (SemanticSearchResult)
Enables viz pane to:
1. Display exact matched chunk with surrounding context
2. Highlight the precise portion of text that matched the query
3. Build user trust by showing what the RAG system actually retrieved
Position tracking uses ChunkWithPosition dataclass from document_chunker.py
which provides character-accurate offsets in the original document.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Extracted vector visualization HTML template to separate file to resolve
syntax conflicts between Jinja2, Alpine.js, and CSS. Added chunk context
endpoint for fetching matched chunks with surrounding text.
Changes:
- Moved vector_viz.html to templates/ directory (separates Jinja2/Alpine.js/CSS)
- Added /app/chunk-context endpoint for retrieving chunk text with context
- Updated .dockerignore to include HTML files in Docker builds
- Moved anthropic and boto3 to main dependencies (needed for production features)
- Added jinja2 dependency for template rendering
Fixes Jinja2 TemplateSyntaxError caused by CSS colons being parsed as
Jinja2 syntax when template was inline in Python code.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Fixed a critical infinite loop bug in document_chunker.py that occurred
when the overlap parameter caused the chunker to not make forward progress.
Changes:
- Added ChunkWithPosition dataclass to track character positions
- Refactored chunk_text() to use regex word matching for accurate position tracking
- Added safety check to ensure forward progress (next_start_idx > start_idx)
- Changed return type from list[str] to list[ChunkWithPosition]
The bug manifested when:
1. end_idx reached len(word_matches) (processing last chunk)
2. next_start_idx = end_idx - overlap would not advance past start_idx
3. Loop would continue indefinitely without making progress
Fix ensures chunker always terminates by breaking when not advancing.
All 9 unit tests now pass in 1.66s (previously timing out at 180s).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Fix false-positive validation error where DBSF (Distribution-Based Score
Fusion) correctly produces scores > 1.0 but SearchResult validation
incorrectly rejected them.
**Root Cause**: SearchResult.__post_init__() enforced scores in [0.0, 1.0]
range, but DBSF sums normalized scores from multiple retrieval systems
(dense semantic + sparse BM25), resulting in scores like 1.55 when both
systems strongly agree a document is relevant.
**Changes**:
- Relaxed validation to allow any score ≥ 0.0 (algorithms.py:147-157)
- Updated SearchResult and SemanticSearchResult documentation to explain
score ranges for RRF ([0.0, 1.0]) vs DBSF (unbounded)
- Added comprehensive test coverage for both fusion methods
- Added DBSF fusion option to vector visualization UI
- Updated viz routes and vizApp() to support fusion parameter selection
**Testing**: All 157 unit tests pass, type checking passes, ruff passes
Fixes error: "Configuration error: Score must be between 0.0 and 1.0, got 1.1528953"
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Refactored LLM provider infrastructure to support sustainable additions of new providers with both embedding and text generation capabilities.
## Major Changes
### Unified Provider Architecture (ADR-015)
- Created `nextcloud_mcp_server/providers/` with unified Provider ABC
- Providers now support optional capabilities (embeddings and/or generation)
- Auto-detection registry with priority: Bedrock → Ollama → Simple
- Backward compatible - existing code continues to work
### New Providers
- **BedrockProvider**: Full Amazon Bedrock integration
- Embeddings: Titan Embed, Cohere Embed models
- Generation: Claude, Llama, Titan Text, Mistral models
- Model-specific request/response handling
- AWS credential chain integration
- **OllamaProvider**: Migrated with both capabilities support
- **AnthropicProvider**: Moved from test code to production providers
- **SimpleProvider**: Migrated in-memory fallback provider
### Breaking Changes
None - full backward compatibility maintained:
- `embedding.get_embedding_service()` still works
- RAG evaluation tests updated to use unified providers
- All existing tests pass (127 unit tests)
### Testing
- Added 9 comprehensive Bedrock unit tests with mocked boto3
- All existing unit tests pass
- Type checking (ty) and linting (ruff) pass
- Verified backward compatibility
### Documentation
- `docs/ADR-015-unified-provider-architecture.md`: Comprehensive ADR
- `docs/bedrock-setup.md`: AWS setup guide with IAM permissions
- `CLAUDE.md`: Updated with provider architecture section
### Dependencies
- Added `boto3>=1.35.0` to dev dependencies (optional)
## Environment Variables
### Bedrock
- `AWS_REGION`: AWS region (e.g., "us-east-1")
- `BEDROCK_EMBEDDING_MODEL`: Model ID for embeddings
- `BEDROCK_GENERATION_MODEL`: Model ID for generation
- `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`: Optional credentials
### Ollama
- `OLLAMA_BASE_URL`: API URL
- `OLLAMA_EMBEDDING_MODEL`: Embedding model (default: "nomic-embed-text")
- `OLLAMA_GENERATION_MODEL`: Generation model
## AWS Bedrock Permissions Required
Minimal IAM policy:
```json
{
"Effect": "Allow",
"Action": ["bedrock:InvokeModel"],
"Resource": ["arn:aws:bedrock:*::foundation-model/*"]
}
```
See `docs/bedrock-setup.md` for detailed setup instructions.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- viz_routes.py: Extract "dense" vector from named vector dict
- semantic.py: Specify using="dense" for BM25 hybrid collections
- Fixes "X must be 2D array" error in hybrid search
- Fixes "Dense vector is not found" error in semantic search
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
The visualization UI was still using the old 'hybrid' algorithm name and
weight parameters that were replaced by the BM25 hybrid search refactor.
This caused "Unknown algorithm: hybrid" errors when using the search
& visualize feature.
Changes:
- Update default algorithm from 'hybrid' to 'bm25_hybrid'
- Update default scoreThreshold from 0.7 to 0.0 to match backend
- Remove deprecated semanticWeight, keywordWeight, fuzzyWeight parameters
- Remove weight parameters from search request
Fixes the visualization search functionality after BM25 hybrid refactor.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Remove obsolete search algorithm imports (Fuzzy, Keyword, Hybrid)
- Update UI to only show Semantic and BM25 Hybrid algorithms
- Replace manual weight controls with RRF fusion info message
- Update default algorithm from "hybrid" to "bm25_hybrid"
- Remove weight parameters (semantic_weight, keyword_weight, fuzzy_weight)
- Update score_threshold default from 0.7 to 0.0 for RRF scoring
- Document ty type checker in CLAUDE.md
Fixes unresolved-import type errors after BM25 refactor.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
This commit addresses critical performance issues with vector visualization
search (reducing time from 40s to ~2s) and improves result visualization
through better visual encoding.
## Performance Fixes
### 1. Fix blocking sleep in retry decorator (base.py:51)
- Changed `time.sleep(5)` to `await anyio.sleep(5)` in @retry_on_429
- Prevents entire event loop from freezing during rate limit retries
- Impact: Reduced search time from 22s to 16s initially
### 2. Add concurrency limiting for verification (verification.py:77-93)
- Added `anyio.Semaphore(20)` to limit concurrent HTTP requests
- Prevents connection pool exhaustion (RequestError) from 90+ simultaneous requests
- Fixes false filtering (was filtering 77/90 results incorrectly)
- Note: Semaphore still in code but verification removed from viz endpoint
### 3. Remove unnecessary verification from viz endpoint (viz_routes.py:483-486)
- Visualization only needs Qdrant metadata (title, excerpt), not full content
- Verification only required for sampling (LLM needs full note content)
- Impact: Reduced search time from 43.7s to ~2s (final fix)
### 4. Restore streaming scanner pattern (scanner.py)
- Process notes one-at-a-time using async generator
- Avoids loading all notes into memory
## Visualization Improvements
### 5. Result-relative score normalization (viz_routes.py:489-504)
- Normalize scores within result set: best=1.0, worst=0.0
- Removes arbitrary RRF normalization (theoretical max didn't make sense)
- Makes visual encoding meaningful regardless of algorithm scores
### 6. Power scaling for marker sizes (userinfo_routes.py:743)
- Changed from linear `8 + (score * 12)` to power `6 + (score² * 14)`
- Creates dramatic visual contrast: 0.0→6px, 0.5→9.5px, 1.0→20px
- Combined with opacity (0.2-1.0) for clear visual hierarchy
### 7. Multi-channel visual encoding (userinfo_routes.py:740-745)
- Size: Exponentially scaled with score²
- Opacity: Linear 0.2-1.0 (keeps all points visible)
- Color: Viridis gradient (blue→yellow)
- Effect: Top results are large/bright/opaque, context results small/dim/transparent
## Result
- Search time: 40s → ~2s (20x faster)
- Visual contrast: Subtle → dramatic (clear result hierarchy)
- No arbitrary cutoffs: All results visible, best naturally highlighted
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Replace asyncio primitives with anyio equivalents throughout the codebase
to establish a single async pattern. This provides better structured
concurrency with automatic cancellation on errors and aligns with the
pytest anyio configuration.
Changes:
- hybrid.py: Replace asyncio.gather() with anyio task groups
- token_broker.py: Replace asyncio.Lock() with anyio.Lock()
- storage.py: Replace asyncio.run() with anyio.run()
- app.py: Replace tg.start_soon() with await tg.start() for task status
- processor.py: Add task_status parameter for structured startup
- scanner.py: Add task_status parameter for structured startup
- CLAUDE.md: Update async/await patterns guidance
The change from start_soon() to await tg.start() enables proper task
initialization signaling, ensuring background tasks are ready before
proceeding. This follows anyio best practices for structured concurrency.
All 118 unit tests pass with the new implementation.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Collect all notes to delete first, then delete concurrently
- Use anyio task group with semaphore (20 concurrent deletions)
- Add progress reporting and error tracking for deletions
- Show count of notes found before deletion starts
This significantly improves --force performance when refreshing large
corpuses (e.g., 3,633 notes now delete in ~1 minute instead of ~5 minutes).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add --force flag to delete all existing notes in target category before upload
- Implement concurrent uploads using anyio task groups (20 concurrent max)
- Add semaphore to limit concurrent requests and avoid overwhelming server
- Improve progress reporting with upload count and error tracking
- Update README with --force flag documentation
Performance improvement: Concurrent uploads significantly reduce upload time
from ~10-15 minutes to ~2-3 minutes for 3,633 documents.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- HuggingFace BeIR/nfcorpus only has 'corpus' and 'queries' configs
- Download qrels from original BEIR ZIP file (nfcorpus.zip)
- Use synchronous httpx.Client for download (simpler than async)
- Remove deprecated trust_remote_code parameter
Tested with successful corpus download and qrels extraction.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Use NextcloudClient with BasicAuth instead of raw httpx
- Replace direct HTTP POST with notes.create_note() method
- Add close() method to LLMProvider Protocol for proper cleanup
- Fix type annotations for dataset iteration
This improves code reuse and consistency with the rest of the codebase.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add ADR-013 documenting RAG evaluation architecture
- Implement two-part evaluation: Context Recall (retrieval) + Answer Correctness (generation)
- Create Click CLI for ground truth generation and corpus upload
- Add pytest fixtures and tests for retrieval/generation quality
- Use BeIR/nfcorpus dataset with 5 selected test queries
- Support Ollama and Anthropic LLM providers
- Generate synthetic ground truth answers offline
- Add comprehensive documentation in tests/rag_evaluation/README.md
The framework separates one-time setup (generate/upload) from test execution,
making tests much faster (~6-12 min vs ~15-25 min per run).
Tests are manual only (not in CI) and require external LLM access.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Improve user comprehension by scaling RRF scores to match the intuitive
0-1 range used by other search algorithms.
## Problem
RRF (Reciprocal Rank Fusion) scores had a drastically different scale
than semantic/keyword/fuzzy scores:
- Semantic similarity: 0.0 to 1.0 (typical: 0.5-0.9)
- RRF scores: 0.0 to ~0.016 (typical: 0.005-0.015)
This caused user confusion - a score of 0.0078 looked terrible but was
actually excellent (near theoretical maximum).
## Solution
Normalize RRF scores using the formula:
`normalized_score = rrf_score * (rrf_k + 1) / total_weight`
Where:
- rrf_k = 60 (RRF constant)
- total_weight = sum of algorithm weights (default: 1.0)
**Example transformation:**
- Before: 0.0078 (confusing)
- After: 0.477 (intuitive)
## Changes
**nextcloud_mcp_server/search/hybrid.py:**
- Store total_weight as instance variable (line 63)
- Calculate normalization factor in _reciprocal_rank_fusion() (line 209)
- Apply normalization to all RRF scores (line 217)
- Preserve raw RRF score in metadata for debugging (line 222)
## Impact
**User Experience:**
- Hybrid search scores now comparable with semantic/keyword/fuzzy
- Score of 0.5 indicates good match across all algorithms
- Consistent scale improves score threshold usability
**Backward Compatibility:**
- Raw RRF scores preserved in metadata["rrf_score_raw"]
- Result ordering unchanged (normalization is linear transformation)
- Breaking change: Existing score thresholds need adjustment
**Performance:**
- Negligible overhead (single multiplication per result)
## Testing
Verified with nc_semantic_search and nc_semantic_search_answer:
- Hybrid scores now 0.47-0.7 range (was 0.003-0.011)
- Semantic scores unchanged (0.75)
- Result ordering preserved
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Move access verification from individual search algorithms to final output
stage, eliminating redundant API calls and improving performance.
## Changes
**New:**
- `search/verification.py`: Centralized verification using anyio task groups
- Deduplicates results by (doc_id, doc_type) before verification
- Verifies all unique documents in parallel using structured concurrency
- Filters out inaccessible documents in single pass
**Modified Search Algorithms:**
- `search/semantic.py`: Removed _deduplicate_and_verify() and _verify_document_access()
- `search/keyword.py`: Removed _verify_access() and parallel verification
- `search/fuzzy.py`: Removed _verify_access() and parallel verification
- `search/hybrid.py`: Removed nextcloud_client parameter passing
All algorithms now return unverified results from Qdrant payload.
**Modified Output Stages:**
- `server/semantic.py`: Added verify_search_results() call after search
- `auth/viz_routes.py`: Added verify_search_results() call after search
Both endpoints now verify access once at final stage with deduplication.
## Performance Impact
**Before:**
- Hybrid mode (limit=10): 30 API calls (10 per algorithm × 3 algorithms)
- Single algorithm: 10-20 API calls (with verification buffer)
**After:**
- Hybrid mode (limit=10): 10 API calls (deduplicated verification)
- Single algorithm: 10 API calls (deduplicated verification)
**Performance Gain:** 3x reduction in API calls for hybrid search
## Architecture Benefits
- **Separation of concerns**: Algorithms handle scoring, output stage handles security
- **Deduplication**: Each document verified exactly once
- **Parallel execution**: All verifications run concurrently via anyio task groups
- **Consistency**: Same verification logic across MCP tools and viz endpoints
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Vector Visualization Improvements:
- Add interactive vector viz tab with Alpine.js and Plotly.js to user info page
- Refactor viz route CSS for better scoping and maintainability
- Remove unused nextcloud_host variable
Performance Optimizations:
- Parallelize access verification in fuzzy and keyword search algorithms
- Use asyncio.gather() to verify multiple documents concurrently
- Add exception handling with return_exceptions=True for resilience
Dependencies:
- Update third_party/oidc submodule to include RFC 9728 resource_url support
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Skip tracing for /app/vector-sync/status to reduce noise from HTMX polling.
Metrics collection continues for this endpoint.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Chris Coutinho