Struct TextIndex 

pub struct TextIndex { /* private fields */ }

Full-text search index backed by tantivy.

Provides BM25 scoring for text search, designed to work alongside HNSW vector search for hybrid retrieval.

Implementations

impl TextIndex

pub fn open<P: AsRef<Path>>(path: P) -> Result<Self>

Create or open a text index at the given path with default config.

Example

use omendb::text::TextIndex;
let index = TextIndex::open("./text_index").unwrap();

pub fn open_with_config<P: AsRef<Path>>( path: P, config: &TextSearchConfig, ) -> Result<Self>

Create or open a text index with custom configuration.

Example

use omendb::text::{TextIndex, TextSearchConfig};
let config = TextSearchConfig { writer_buffer_mb: 100 };
let index = TextIndex::open_with_config("./text_index", &config).unwrap();

pub fn open_in_memory() -> Result<Self>

Create an in-memory text index with default config.

pub fn open_in_memory_with_config(config: &TextSearchConfig) -> Result<Self>

Create an in-memory text index with custom configuration.

pub fn index_document(&mut self, id: &str, text: &str) -> Result<()>

Index a document with the given ID and text content.

If a document with this ID already exists, it will be updated.

pub fn delete_document(&mut self, id: &str) -> Result<()>

Delete a document by ID.

pub fn commit(&mut self) -> Result<()>

Commit pending changes to the index.

Changes are not visible to searchers until commit is called. This also reloads the reader to see the new changes immediately.

pub fn search( &self, query_str: &str, limit: usize, ) -> Result<Vec<(String, f32)>>

Search for documents matching the query.

Returns a vector of (id, score) tuples, sorted by score descending.

Arguments

• query_str - The search query (supports tantivy query syntax)
• limit - Maximum number of results to return

pub fn num_docs(&self) -> u64

Get the number of documents in the index.

pub fn index(&self) -> &Index

Get a reference to the underlying tantivy index.

pub fn reader(&self) -> &IndexReader

Get a reference to the index reader.