Get cache statistics

Detect all semantic boundaries in text

Calculate dynamic weight for a relationship based on query context (Phase 2.2) Combines multiple factors: - Base weight: relationship.confidence - Se

Extract keywords using simple frequency analysis

Find entity by name (fuzzy matching)

Generate a cache key for a prompt

Get value from cache Checks L1 cache first, then L2 (Redis), with automatic promotion. # Arguments `key` - Cache key # Returns Option with cached v

Get a chunk by ID

Check if documents have been added

Helper function: Replicates the hash-based embedding algorithm This is the same algorithm implemented in CandleEmbedder::embed()

Load GraphRAG configuration from file (supports JSON5, JSON, TOML)

Determine if parallel processing should be used

Add a vector to the store Vectors are added to a pending queue and indexed when `build_index()` is called. # Arguments `id` - Unique identifier for

Calculate improvement percentage

Clear all documents and reset the index

Count paragraphs in text

Create a snapshot of the current graph state Used for delta computation to calculate minimal diffs.

Create a snapshot from current graph state

Get style for error messages

Find all causal chains between cause and effect # Arguments `cause` - Starting entity ID `effect` - Target entity ID `max_depth` - Maximum chain len

Get entity relationships

Check if a word is capitalized

Check if GraphRAG is initialized

Get the number of vectors in the store

Register a new function

Update scrollbar state

Set causal strength for this relationship

Create new system with custom PageRank configuration

Set temporal range for this relationship

Add an intent pattern

Get style for borders (normal)

Get style for borders (focused)

Calculate the required `num_ctx` for a document and its chunks Formula: ```text num_ctx = tokens(prompt_instructions) // ~100 fixed tokens + tokens

Calculate cosine similarity between two embedding vectors

Helper to create a test graph with philosophical entities and relationships

Helper to create a test graph with relationships

Get a value from the cache

Get neighbors of an entity

Get processing statistics

Compute the hash of this cache key based on key_hash and display_key

Helper function to convert IdbRequest to Promise

Parse a markdown string into ratatui `Line` values ready for rendering.

Update application state based on action

Get workspace directory path

Add a document to the knowledge graph

Update document frequencies with a new document (for corpus-level IDF)

Add an edge to the layout

Add extraction rule

Add a response template

Build the hierarchical tree from text chunks

Calculate personalized PageRank scores with performance optimizations

Convenience method: chunk and enrich with auto-detected format

Clear all indexed data

Compute delta between two snapshots

Cosine similarity implementation (Same as in graphrag-wasm/src/lib.rs)

Ensure system is initialized

Get file extension for this format

Helper: Fetch tokenizer JSON from server

Generate cache key for a request

Check if graph has been built

Advanced hybrid query with strategy selection and hierarchical integration

Infer entity type from name (placeholder implementation)

Initialize the GraphRAG system. When `auto_save.enabled = true` and a `base_dir` is configured, attempts to load an existing graph from the workspace

Get neighbors of a node

Register a function in a specific category

Rewrite and expand a query # Arguments `query` - The original query string # Returns RewrittenQuery with expanded terms and detected query type

Score the semantic coherence of a text chunk Returns a score between 0.0 (incoherent) and 1.0 (highly coherent). High coherence = high cosine similar

Get storage statistics

Update the last snapshot to current state Should be called after significant updates to enable delta computation.

Create heading with line number

Set temporal type for this relationship